Literary work compression encoding and traversal reconstruction method based on configurable factor decomposition

Abstract

The application discloses a literary work compression encoding and traversal reconstruction method based on configurable factor decomposition. For the first time, the structural characteristics of literary works are expressed as at least two independent and combinable factor encoding sections, at least one factor encoding creative structural decision (emotional pattern intensity, character mobility, value orientation, narrative skeleton, etc.), and at least another factor encoding mechanically traversable external setting (time and space coordinates, style texture, etc.). The factors are separated by delimiters and can be independently stored, transmitted, indexed and replaced. The factor configuration scheme (the allocation relationship of each parameter group to each factor) can be reconfigured in different implementation forms - the same parameter group belongs to the creative factor in one scheme, and can be allocated to the traversable factor in another scheme - so that the same set of encoding framework can adapt to different creative workflows. Taking a short story "Heart of the Arctic" (about 15,000 characters) as an example, the core factor encoding section is 85 characters, the projection factor encoding section is 34 characters, and the combination is a factorized compact encoding of 119 characters, with a compression ratio of 126:1. The application protects any such configurable factor decomposition scheme and its traversal generation method.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary technical field of natural language processing, data compression and computational creativity. Specifically, it relates to a method for encoding the structured features of literary works into two independent and composable compressed code segments through factorization, thereby reducing the storage space requirement, and a method for batch traversal generation of literary works based on this factorization structure. Background Technology

[0002] Existing literary encoding and generation technologies mainly fall into two categories, both with significant shortcomings: The first category is text-level compression. Represented by common compression algorithms such as ZIP and GZIP, this only eliminates byte-level data redundancy, resulting in low compression rates. It also loses all literary structural information after compression, and the compressed results cannot be directly used for content retrieval or automated creation. The compressed data is an uninterpretable binary string, completely losing literary structural information such as the work's emotional tone, plot arc, character relationships, and narrative style. The compressed file cannot be used for content retrieval, work similarity comparison, or automated creation—in these application scenarios, it needs to be decompressed and restored to complete text before further analysis.

[0003] The second category: structural-level encoding and text decoupling representation learning. Existing text decoupling representation learning technologies (such as style-content decoupling models based on VAEs) are mainly aimed at style transfer tasks for short texts at the sentence level. They obtain abstract representation dimensions automatically through model learning, but do not have the ability to parameterize interpretable structures for entire literary works.

[0004] Existing methods for creating natural language prompts also have shortcomings. In the field of automated literary creation, current technologies mainly rely on user-provided natural language prompts, with a large language model (LLM) generating a single piece of work in one go. Natural language prompts are qualitative descriptions and have the following inherent limitations: (1) It is not possible to specify creative parameters in a mathematically precise way. Users can say "make this story sadder", but they cannot specify how much sadness needs to be increased in a numerically precise way, nor can they shift the style from one precise coordinate point to another.

[0005] (2) Content retrieval or similarity comparison cannot be performed in compressed state. Whether two long works generated from similar prompts are structurally similar requires manual reading of the full text to determine—and these prompts themselves do not constitute a computable mathematical representation.

[0006] (3) It cannot support batch traversal generation. If you want to generate 50 variant works of the same story idea that take place in 10 different eras and 5 different locations, you need to write 50 independent natural language prompts. Most of the content (emotional structure, character relationships, plot skeleton) is exactly the same. Each time, you need to manually repeat the description, which is extremely inefficient.

[0007] This invention aims to solve the following technical problems: 1. How to decompose literary encoding into core factors and projection factors that can be stored, transmitted, indexed, and replaced independently, thereby significantly reducing storage requirements; 2. How to achieve mechanical traversal based on factor decomposition structure—fixing the core and batch replacing projections to generate a series of variant works; 3. How to define a cross-factor quantitative indicator (core-projection eccentricity angle) to distinguish between two different types of similarity: "same core, different projection" and "different core, same projection". Summary of the Invention

[0008] The core concept of this invention is to express a literary work as a combination of multiple independent factors for the first time. Furthermore, the structural features of a literary work should not be compressed into an indivisible single data block, but rather decomposed and represented as the product (concatenation) of at least two independent factor coding segments. Each factor coding segment is distinguished from the others by a delimiter within a compact coding string, and each can be operated on independently.

[0009] Taking the core × projection two-factor decomposition scheme in a preferred embodiment of the present invention as an example (see attached diagram) Figure 1 ): Core Factors (C-OTMES, 20-dimensional) encode "what kind of story is told"—the emotional core and narrative framework of a literary work: The literary mode tensor M (10-dimensional): the intensity of 10 emotional modes (tragedy M0 / comedy M1 / ​​satire M2 / poetry M3 / political intrigue M4 / suspense M5 / horror M6 / science fiction M7 / romance M8 / epic M9) in the work, with values ​​ranging from [0, 10]. The dynamism vector N (2-dimensional): [proactivity level N1, passivity level N2], N1+N2=1, with a value range of [0, 1]; Value carrier vector K (2-dimensional): [emotional individual orientation K1, rational supra-individual orientation K2], K1+K2=1, value range [0, 1]; Irreversibility I (1-dimensional): The degree of irreversibility of the core choices made in the work, [0, 1]; Plot skeleton vector B (5 dimensions): [Number of turning points B0, intensity of midpoint reversal B1, proportion of climax location B2, proportion of coda length B3, density of flashbacks B4], each dimension [0, 1].

[0010] Projection factors (P-OTMES, 19-dimensional) encode "the context in which the story is told"—the spatiotemporal setting, character configuration, and stylistic texture of a literary work: Character configuration vector C (5 dimensions): [Age span C0, Adversarial relationship type C1, Supporting character density C2, Key death number C3, Relationship change range C4], each dimension [0, 1]; Spatiotemporal coordinate vector D (5 dimensions): [Date D0, Geographical latitude D1, Urban-rural gradient D2, Historical embedding D3, Social class D4], each dimension [0, 1]; Ending texture vector E (4-dimensional): [Joy / sorrow value E0 ([-1,1]), Ambiguity E1, Epilogue scale E2, Beginning-End Echo E3], E1 / E2 / E3 ∈ [0, 1]; Style texture vector F (5 dimensions): [Sentence complexity F0, Dialogue density F1, Description ratio F2, Narrative distance F3, Temporal rhythm F4], each dimension [0, 1].

[0011] The key technical features of the factorization coding proposed in this invention are as follows.

[0012] Feature 1 – Independent and Replaceable Factors. The core factor encoding segment and the projection factor encoding segment are clearly distinguished by the separator `|` within the compact encoding string. Each of the two encoding segments can be stored, transmitted, indexed, and replaced independently – a core factor encoding segment can be paired with any number of projection factor encoding segments, and vice versa. This means that if a creator designs 10 different cores (10 story skeletons) and 10 different projections (10 spatiotemporal settings), they can freely combine them to create 10 × 10 = 100 new works that are structurally independent and do not infringe on each other's copyrights – but the actual computational load is only 10 core encodings + 10 projection encodings = 20 calculations (O(k+m)), not 100 complete encodings (O(k×m)).

[0013] Feature 2 - The factor configuration scheme is reconfigurable. The allocation relationship of each parameter group to core factors or projection factors is defined by the Factor Configuration Scheme (CFG). The same group can be allocated to different factors in different implementation modes. For example, in the CFG-01 scheme, narrative skeleton B belongs to a creative factor (core); in the CFG-03 scheme, B can be allocated to an independent second factor (narrative skeleton factor), which is juxtaposed with the emotion factor and the style factor.

[0014] Feature 3 - Core structure indicators. The present invention defines two new quantitative indicators that only exist under the factor decomposition structure: Core tension angle τ = atan2(||M_counter||, M_dominant) × 180 / π, wherein M_counter is the norm of all non-dominant modes. τ ∈ [0°, 90°], which characterizes the force contrast between the "dominant emotion mode" and the "counter emotion mode" inside the core factor. τ → 0° represents a pure dominant type (one emotion mode is overwhelmingly dominant), τ→ 45° represents a complex interweaving type (polyphonic narration), and τ → 90° represents a decentralized type (difficult to be classified into a single type).

[0015] Core-projection eccentricity angle φ = atan2(E_proj, E_core) × 180 / π, wherein E_core = ||M||×||N||×||K||×||B|| is the total energy of the core factor, and E_proj = ||C||×||D||×||E'||×||F|| is the total energy of the projection factor. φ ∈ [0°, 90°], which characterizes whether a work is "core-driven" (the story itself is dominant, φ→0°), "balanced" (φ≈45°) or "projection-driven" (space-time setting and style are dominant, φ→90°).

[0016] The factorized encoding process is shown below by taking specific works as an example.

[0017] The following illustrates the factorized encoding process of the present invention by taking a completed short story *The Polar Heart* (Chinese translation: *The Polar Heart*, with a full text of about 15,000 English characters and about 8,000 Chinese characters, which has been submitted as an appendix to the present application) as an example.

[0018] The core plot of the novel is as follows: In 1911, British meteorologist Isabel Meredith disguised herself as a man and joined Scott's Antarctic expedition as a meteorological observer under the name of her deceased brother, Edward Meredith. During the voyage to Antarctica, she met the expedition's sled driver, Ellis Hackett—a former White Russian officer. On the Antarctic ice sheet, she pointed out that his file angle was off by 0.7° when he repaired the sled track; he replied, "22.7°, I'll remember." The two fell in love. After the expedition, she was expelled from the Royal Geographical Society for fraud and had her teaching position at Cambridge University revoked. Standing at Lyttelton Pier, she announced her marriage to Hackett in front of her father. In the spring of the third year of their marriage, World War I broke out, and Hackett was drafted and killed in the Battle of the Somme in 1916. For the remaining twenty-three years of her life, Isabel charted Antarctica for the British Admiralty Hydrographic Office—located at 78°S, 162°E. The unnamed glacier bears a note saying, 'This route is not recommended.' She died in 1971, and her drawer contained letters from Hackett and copies of telegrams from Dr. Wilson to the Royal Geographical Society.

[0019] Step 1: Generation of the core coding segment (C-OTMES, 20-dimensional). (See attached document) Figure 2 .

[0020] The system extracts the following core parameters from the above works: M = [9.0, 0.5, 4.0, 9.0, 3.0, 1.0, 6.5, 0, 8.5, 2.0] (Tragedy 9.0, Poetry 9.0, Romance 8.5 are the three main modes) N = [0.8, 0.2] (Highly proactive) K = [0.4, 0.6] (Predominantly rational supra-individual) I = 1.0 (completely irreversible) B = [0.75, 0.50, 0.85, 0.30, 0.10] (Multiple twists, strong midpoint reversal, climax slightly later, moderate ending, a few flashbacks) Calculate the core indicators based on the above parameters: The total potential energy of literature E = ||M||×||N||×||K|| ≈ 10.41, E_code = int(10.41×10) =104 Dominant mode = M0 (tragedy), M_dominant = 9.0 The adversarial mode norm M_counter = √(0.5²+4²+9²+3²+1²+6.5²+0²+8.5²+2²) ≈15.03 Core tension angle τ = atan2(15.03, 9.0)×180 / π ≈ 59.1°, τ_code = 059 Tensor Rank = 9 (9 active out of 10 modes) The principal component percentage η = 9.0 / 43.5×100 ≈ 20, η_code = 20 Irreversible code I_code = int(1.0 × 10) = 10 Tensor hash = MD5("9_9.00_10.41") First 6 bits = "35526E" Vector B in hexadecimal: Each dimension × 100, then converted to 2 hexadecimal digits = "4B32551E0A" Checksum = MD5("C-OTMES-v1-35526E-104-M0-059-9R2010") First 4 bits = "10AA" The generated core encoding segment is: C-OTMES-v1-35526E-104-M0-059-9R2010-10AA|B:4B32551E0A.

[0021] Step 2: Generation of the projected coded segment (P-OTMES, 19-dimensional). (See attached document.) Figure 3 .

[0022] Extract the following projection parameters from the above works: C = [0.70, 0.80, 0.30, 0.40, 0.80] (Large age range, confrontation with external systems, few supporting characters, a few deaths, extreme relationship changes) D = [0.37, 0.90, 0.20, 0.80, 0.50] (circa 1874, polar latitude, wilderness environment, highly embedded in World War I, middle class) E = [-0.40, 0.10, 0.70, 0.90] (Bitter ending, completely clear, long coda, strong beginning and end echo) F = [0.60, 0.40, 0.70, 0.30, 0.50] (Medium sentence length, little dialogue, more descriptive, near third person, moderate rhythm) Multiply each dimension by 100 and then convert to 2 hexadecimal digits (E0 is first mapped to (E0+1) / 2 = 0.30 → 1E): C_hex = "46501E2850" D_hex = "255A145032" E_hex = "1E0A465A" F_hex = "3C28461E32" The generated projection encoding segment is: P-OTMES-v1|C:46501E2850|D:255A145032|E:1E0A465A|F:3C28461E32.

[0023] Step 3: Combine into a factorized compact encoded string.

[0024] Calculate the core - projection eccentricity angle φ: E_core = ||M||×||N||×||K||×||B|| ≈ 10.41×1.28 ≈ 13.36 E_proj = ||C||×||D||×||E'||×||F|| ≈ 1.50×1.40×1.25×1.08 ≈2.84 φ = atan2(2.84, 13.36)×180 / π ≈ 12.0°, φ_code = 12 The core coding segment, projection coding segment, and eccentricity angle are combined using the separator `|`: C-OTMES-v1-35526E-104-M0-059-9R2010-10AA|B:4B32551E0A|P-OTMES-v1|C:46501E2850|D:255A145032|E:1E0A465A|F:3C28461E32|φ:12.

[0025] Total length 119 characters. The original text summary is approximately 15,000 characters. Even compared to the simplified version of the summary, the compression ratio is still 15,000 / 119 ≈ 126:1.

[0026] φ = 12° Interpretation: φ is located in the 0°–30° range, belonging to the "core-driven type"—the story structure of *Heart of the Arctic* itself strongly dominates, while the time and space setting and style are secondary. This is consistent with the actual situation of the work: whether Isabel is in the Arctic, in Southampton, or in the Admiralty, no matter how the background changes, the core emotional framework of "disguised identity → love → irreversible choice → lifelong drawing" remains unchanged.

[0027] The factor configuration scheme proposed in this invention has flexible configurability.

[0028] The above embodiment uses the preferred scheme CFG-01 (core = M, N, K, I, B, projection = C, D, E, F). This invention is not limited to this single partitioning method. The factor allocation scheme defines the allocation relationship between each group and each factor; the following are some example schemes: Option 1 (CFG-01: Core × Projection): Creativity Factor = M, N, K, I, B (20 dimensions), Erratable Factor = C, D, E, F (19 dimensions);

[0029] Option 2 (CFG-02: Creativity × Pure Errata): Creativity Factor = M, N, K, I, B, C (25 dimensions), Erratatable Factor = D, F (10 dimensions). The erratatable factor in this option only includes spatiotemporal coordinates and style textures—all 10 parameters can be traversed through a regular grid (7 levels of era × 5 levels of latitude × 5 levels of urban / rural × 5 levels of social class × 3 levels of sentence length × 3 levels of dialogue density × 4 levels of narrative distance = 31,500 mechanized combinations), achieving erratatable traversal with "zero creative participation";

[0030] Scheme 3 (CFG-03: Three-Factor Decomposition): First Factor (Emotional Core) = M, N, K, I (15 dimensions), Second Factor (Narrative Skeleton) = B, C, E0 (11 dimensions), Third Factor (Style Texture) = D, F, E1₋3 (13 dimensions). During traversal, any one or two factors can be fixed while traversing the rest.

[0031] The factorization structure of this invention also naturally supports the following traversal generation patterns: Mode 1 – Fixed Factor Traversal. Keeping the creative factor constant, replace the traversable factor with different values ​​to generate a series of projected variant works. For example, by keeping the core coded segment of *Heart of the Polar Regions* fixed, and sequentially combining it with four different projected coded segments (Ancient Equatorial City, Victorian Temperate Countryside, Polar Wilderness, and Jazz Age City), four independent works with the same core structure but vastly different temporal and spatial backgrounds are generated. (See attached reference.) Figure 4 In the four works, the core coding segment is calculated only once, while the projection coding segment is calculated four times and combined into a string concatenation—the computational cost is O(1+4) instead of O(4).

[0032] Mode 2 – Cartesian Product Traversal. This mode combines k core coding segments with m projected coding segments to generate k×m cross-variant works. The computational cost is O(k+m), not O(k×m). (See attached reference.) Figure 5 .

[0033] Mode 3 – Dual Traversal. Perform complementary mapping on any factor encoding segment – ​​M'[i] = 10-M[i], N'=[N2,N1], K'=[K2,K1], I'=1-I, E'0=-E0, D'=1-D – to generate a dual encoding segment, which, combined with the remaining factors, produces an emotionally mirrored work. For example, the dual encoding ¬C0׬P0 of *Heart of the Polar Regions* produces a comedic adventure story set in an equatorial tropical city.

[0034] It is important to note that this invention directly encodes the structural parameters of a literary work into at least two independent factor coding segments, rather than "splitting an existing single code into multiple segments." In this invention, each factor coding segment of the combined coding string is independently extracted and generated directly from the original text of the target literary work—the core factor coding segment extracts emotional patterns and narrative skeleton information from the original text, while the projection factor coding segment extracts spatiotemporal setting and stylistic texture information from the original text; both are independent from each other at the extraction stage. Attached Figure Description

[0035] Figure 1: Factorization structure of 39-dimensional transformation to 20+19. It shows the complete flat 39-dimensional encoding, how to split the core factor (20 dimensions) and projected factor (19 dimensions), and their respective subgroups. It should be noted that 39 dimensions is just one encoding method; we have also experimented with 45-dimensional schemes in our tests, and these specific parameters are not specified.

[0036] Figure 2: Core Coding Segment (C-OTMES) Generation Process. This demonstrates the complete computation process from 20-dimensional core parameters to a compact C-OTMES encoded string, including the calculation steps for E, τ, Rank, η, I, and hash value.

[0037] Figure 3: Projection-OtmES (P-OTMES) generation process. It illustrates the hexadecimal encoding process from 19-dimensional projection parameters to a compact P-OTMES encoded string.

[0038] Figure 4: Fixed Core Traversal Projection. This illustrates the process of generating four projection variant works using a single core code (C1) paired with four different projection codes (P1~P4).

[0039] Figure 5: Cartesian product of core and projection. It shows the complete combination of k=3 cores and m=3 projections, totaling 9 cross-variant works. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0041] Example 1: Compression coding of literary works based on core-projective two-factor decomposition.

[0042] This embodiment uses a completed short story, *Heart of the Arctic* (approximately 15,000 English characters), as an example to illustrate the factorization coding process of this invention. The core plot of the story is as follows: In 1911, British meteorologist Isabelle Meredith disguises herself as a man to join an Antarctic expedition and falls in love with sled driver Ellis Hackett. After the expedition, she is expelled for fraud and chooses to marry Hackett. Three years after their marriage, Hackett is killed in the Battle of the Somme during World War I. Isabelle spends the remaining twenty-three years of her life drawing Antarctic charts for the Admiralty.

[0043] 1. Generation of the core coding segment (C-OTMES, 20-dimensional).

[0044] The system extracts core creative parameter sets from the above works. Extraction can be achieved through one of three methods: (a) manual annotation; (b) automatic model extraction; (c) human-computer collaboration. The extraction results are as follows: The literary mode tensor M (10-dimensional): [9.0, 0.5, 4.0, 9.0, 3.0, 1.0, 6.5, 0, 8.5, 2.0], of which tragedy (9.0), poetry (9.0), and romance (8.5) are the three dominant modes; The agency vector N (2-dimensional): [0.8, 0.2], represents a highly proactive type; The value carrier vector K (2-dimensional): [0.4, 0.6], represents the partially rational supra-individual; Irreversibility I (1-dimensional): 1.0, representing complete irreversibility; The plot skeleton vector B (5-dimensional): [0.75, 0.50, 0.85, 0.30, 0.10], represents multiple twists and turns, a strong midpoint reversal, a climax that comes later in the day, a moderate ending, and a few flashbacks.

[0045] Calculate the core structural indicators based on the above parameters: The total potential energy of literature E = ||M||×||N||×||K|| ≈ 10.41, E_code = 104; The dominant mode is M0 (tragedy), M_dominant = 9.0; The adversarial mode norm M_counter = √(0.5²+4²+9²+3²+1²+6.5²+0²+8.5²+2²) ≈15.03; The core tension angle τ = atan2(15.03, 9.0)×180 / π ≈ 59.1°, τ_code = 059; Tensor rank = 9; The principal component percentage η = 9.0 / 43.5×100 ≈ 20, η_code = 20; Irreversible code I_code = 10; Tensor hash = MD5("9_9.00_10.41") First 6 bits = "35526E"; Vector B in hexadecimal: Each dimension × 100, then converted to 2 hexadecimal digits = "4B32551E0A"; Checksum = MD5("C-OTMES-v1-35526E-104-M0-059-9R2010") First 4 bits = "10AA".

[0046] The generated core encoding segment is: C-OTMES-v1-35526E-104-M0-059-9R2010-10AA|B:4B32551E0A.

[0047] 2. Generation of projection coding segments (P-OTMES, 19-dimensional).

[0048] Extract the projection parameter set from the above works: Character configuration vector C (5 dimensions): [0.70, 0.80, 0.30, 0.40, 0.80]; The spatiotemporal coordinate vector D (5-dimensional): [0.37, 0.90, 0.20, 0.80, 0.50], represents approximately 1874, polar latitude, wilderness environment, highly embedded in World War I, and the middle class; The ending texture vector E (4-dimensional): [-0.40, 0.10, 0.70, 0.90], represents a bitter ending, complete clarity, long coda, and strong beginning and end echo; Style texture vector F (5-dimensional): [0.60, 0.40, 0.70, 0.30, 0.50].

[0049] Multiply each dimension by 100 and convert to 2 hexadecimal digits (E0 is first mapped to (E0+1) / 2 = 0.30 → 1E): C_hex = "46501E2850"; D_hex = "255A145032"; E_hex = "1E0A465A"; F_hex = "3C28461E32".

[0050] The generated projection encoding segment is: P-OTMES-v1|C:46501E2850|D:255A145032|E:1E0A465A|F:3C28461E32.

[0051] 3. Combine into a factorized compact encoded string.

[0052] Calculate the core - projection eccentricity angle φ: E_core = ||M||×||N||×||K||×||B|| ≈ 10.41×1.28 ≈ 13.36; E_proj = ||C||×||D||×||E'||×||F|| ≈ 1.50×1.40×1.25×1.08 ≈2.84; φ = atan2(2.84, 13.36)×180 / π ≈ 12.0°, φ_code = 12.

[0053] Combine the core coding segment, projection coding segment, and eccentricity angle using the separator `|` to generate the complete code: C-OTMES-v1-35526E-104-M0-059-9R2010-10AA|B:4B32551E0A|P-OTMES-v1|C:46501E2850|D:255A145032|E:1E0A465A|F:3C28461E32|φ:12.

[0054] The total length of the encoding is 119 characters. The original text is approximately 15,000 characters, so the compression ratio is 15,000 / 119 ≈ 126:1. The eccentric angle φ = 12° is located in the 0°–30° range, which is "core-driven," consistent with the fact that the core emotional framework of the work remains dominant regardless of changes in the background.

[0055] Example 2: Batch generation of fixed core traversal projection.

[0056] This embodiment uses the core coding segment C0 from Embodiment 1 as a fixed core, and sequentially pairs it with 4 different projective coding segments to verify the traversal generation capability of independent factor substitution.

[0057] The spatiotemporal coordinate parameters of the four preset projection coding segments are set as follows: (1) Ancient equatorial city, D0=0.15, D1=0.20, D2=0.70; (2) Victorian temperate countryside, D0=0.40, D1=0.60, D2=0.30; (3) Polar wilderness (original projection), D0=0.37, D1=0.90, D2=0.20; (4) Jazz age city, D0=0.55, D1=0.50, D2=0.90.

[0058] The combined generation process involves concatenating the core coding segment C0 with each of the four projective coding segments mentioned above, generating four projective variant works from the input text generation model. The four works share the exact same core story framework (the protagonist disguises themselves to enter a male-dominated world, romance drives the plot, an irreversible choice, and a bittersweet but dignified ending), but are respectively portrayed as a Carthaginian priestess, the daughter of an English country gentleman, an Antarctic meteorologist, and a New York female architect. The dialogue, scene descriptions, and specific details in the four works are all independent original creations.

[0059] Efficiency Analysis: In the four generation steps, the core coding segment C0 is calculated only once, the projected coding segment is calculated four times, and the combination step is string concatenation. The original work text is approximately 15,000 characters, the core coding segment is 85 characters, and each projected coding segment is approximately 34 characters (C segment omitted, only D and F are included). The total encoding of the four works = 85 (core segment calculated once) + 34 × 4 (projected segment calculated four times) = 221 characters. Taking four works with 4 × 15,000 = 60,000 characters as an example, the equivalent compression ratio = 60,000 / 221 ≈ 271:1. Compared to the non-factorized O(4) complete encoding calculations, the computational complexity of this invention is significantly reduced.

[0060] Example 3: Three-factor Cartesian product traversal based on the CFG-03 scheme.

[0061] This embodiment uses the CFG-03 scheme (three-factor decomposition) to verify the efficiency of Cartesian product traversal under multi-factor configuration.

[0062] The factors are allocated as follows: Factor 1 F1 (emotional core) = M, N, K, I (15 dimensions); Factor 2 F2 (narrative skeleton) = B, C, E0 (11 dimensions); Factor 3 F3 (style texture) = D, F, E1₋3 (13 dimensions).

[0063] Configure 3 emotion core variants, 2 narrative skeleton variants, and 2 style texture variants. Perform Cartesian product traversal to generate a total of 3×2×2=12 cross-variant works. The encoding segments of each factor are generated independently and combined into 12 complete codes.

[0064] Efficiency Comparison: Under the factorization scheme, the computational cost is only O(3+2+2) = 7 factor encoding calculations plus 12 string concatenations; if the non-factorization flat coding method is used to process three sets of parameters, it requires 7 × 39 = 273 parameter extraction operations and 12 complete compact code generation operations, with a computational cost of O(3 × 2 × 2) = 12. This example demonstrates that the advantage in computational complexity increases exponentially with the increase in the number of factors.

[0065] Example 4: Dual encoding generation.

[0066] This embodiment performs complementary dual mapping on the factorized encoding of "Heart of the Polar Regions" in Embodiment 1 to verify the ability to generate emotional mirror works.

[0067] Perform dual mapping on core coding segment C0: M'=[1.0, 9.5, 6.0, 1.0, 7.0, 9.0, 3.5, 10.0, 1.5, 8.0] (tragedy 9.0 maps to 1.0, comedy 0.5 maps to 9.5, science fiction 0 maps to 10.0, etc.); N'=[0.2,0.8] (active 0.8 maps to 0.2); K'=[0.6, 0.4]; I'=0.0 (irreversible 1.0 maps to 0.0); the B vector remains unchanged. The dominant mode of dual core ¬C0 becomes M7 (science fiction, 10.0), and the core tension angle τ is approximately 55°.

[0068] Perform dual mapping on the projected coded segment P0: D'=[0.63, 0.10, 0.80, 0.20, 0.50] (latitude 0.90 polar region maps to 0.10 equator, urban / rural / wilderness region maps to 0.80 urban region); E'0=+0.40 (bitter ending maps to comedic ending).

[0069] Combining the dual core and dual projection into a dual encoding (¬C0, ¬P0), and inputting it into a text generation model, produces a comedic adventure story set in an equatorial tropical sci-fi metropolis. The Euclidean distance between this dual encoding and the original encoding in 39-dimensional space is approximately 25.3, far exceeding the same-family threshold, verifying that dual mapping can produce a complete emotional reversal and structural complementarity from "polar tragic love" to "equatorial comedic adventure".

[0070] Through the above embodiments, the present invention has the following significant advantages over the prior art: First, compared to existing text compression technologies (such as ZIP), which compress into unreadable binary strings, completely lose literary structure information and do not support independent replacement and reconstruction, with a compression ratio of only 2-5:1; while the present invention compresses each segment to a specific semantic group, fully preserves 39-dimensional literary structure parameters, supports factor-level independent storage and replacement, achieves a compression ratio of 100-270:1, and supports encoding-driven LLM reconstruction of approximate text.

[0071] Secondly, compared to existing LLM Prompt methods, which are based on qualitative descriptions in natural language and cannot specify creation parameters in a mathematically precise manner, the existing methods require O(k×m) manual writing of Prompts for traversal generation and cannot perform structural similarity comparisons in a compressed state. In contrast, this invention achieves quantitative and precise control of parameters through factorization encoding, reduces the computational cost of traversal generation to O(k+m), and supports independent distance calculation and similarity comparison for each factor subspace.

[0072] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. For example, the creative coding group can be selected from at least one of the following groups, without being limited by the specific examples in this specification: emotional pattern intensity, character agency, value orientation, narrative dynamics, narrative skeleton (narrative arc shape), degree of irreversibility, core conflict paradigm, ethical grayness, cognitive asymmetry parameter, core relationship, character arc paradigm, core cost, ultimate questioning, narrative closure, ethical grayness, cognitive asymmetry, temporal structure, narrative reliability, world logic paradigm, core desire paradigm, narrative aggregation form, emotional contract paradigm, boundary breakthrough paradigm, meta-narrative degree, memory reconstruction paradigm, micro-power structure, spatiotemporal perception deformation, symbolic density, violent aesthetics paradigm, rescue The possibilities of redemption, the gaze of the other, information deconstruction paradigm, identity mobility, survival game paradigm, primordial scarcity paradigm, connection isolation paradigm, language constructing reality paradigm, body politics paradigm, cosmic absurdity, time cycle destiny paradigm, existence form paradigm, group awakening, taboo transgression paradigm, sacrifice and compensation paradigm, reality deconstruction level, order collapse paradigm, spatial topology paradigm, destiny empowerment paradigm, breakthrough inducement paradigm, guilt attribution paradigm, core eroticism paradigm, historical depth paradigm, alienated other paradigm, truth production paradigm, intergenerational inheritance paradigm, consciousness base paradigm, comedic misplacement paradigm, narrative scale paradigm, hope maintenance paradigm, mirror intertextuality paradigm, alienated labor paradigm, or a recombination of the above contents. Among them, the emotional pattern intensity parameter: represents the intensity of each of the at least two emotional patterns in the work, reflecting the core emotional tone of the work; Among them, the character agency parameter: a value that represents the degree to which the narrative subject actively initiates behavior or passively endures external forces in the work (such as active resistance and sacrifice versus passive victimization and deception), reflecting the action source structure of the work; Among them, the value orientation parameter: represents the value object attribute (such as the value of the emotional individual and the value of the rational supra-individual) to which the core conflict of the work points, and reflects the value anchor structure of the work; Among them, the narrative dynamics parameter is a numerical value that characterizes the degree to which characters in the work actively drive the plot or are driven by external forces, reflecting the core action structure of the work. Among them, the narrative skeleton parameter is a value that represents the density of plot twists, the position of the climax, or the shape of the narrative arc of the work, reflecting the core story structure of the work. Among them, the irreversibility parameter is a value that characterizes the degree to which the core choices or core losses in the work are irreversible (such as fully reversible trial and error versus absolutely irreversible sacrifice), reflecting the intensity of the work's sense of destiny and the weight of the choices. Among them, the core conflict paradigm parameter is a numerical value that represents the underlying antagonistic logic type (such as individual versus system, individual versus inner self, and human versus nature) that drives the plot development in the work, reflecting the underlying dynamic model of the work. Among them, the ethical grayscale parameter is a numerical value that represents the absoluteness or ambiguity of the core behaviors and choices in the work in terms of moral judgment, reflecting the philosophical core and value tension of the work. Among them, the cognitive asymmetry parameter is a numerical value that characterizes the gap in the degree of information mastery between the narrator, the core character and the audience in the work and the resulting suspense, reflecting the essential narrative perspective of the work. Among them, the core relationship parameter is a numerical value that represents the deepest interaction patterns and interdependence states (such as symbiotic redemption, power oppression, and mirror intertextuality) between the core characters in the work, reflecting the interpersonal dynamics of the work. Among them, the character arc paradigm parameter is a value that represents the direction of the inner spiritual evolution trajectory of the core character in the work from the beginning to the end of the narrative (such as positive redemption, negative depravity and adherence to one's original intention), reflecting the character evolution structure of the work. Among them, the core cost paradigm parameter represents the value of the essential sacrifices (such as physical sacrifice, moral depravity, and identity erasure) that must be made in order to achieve the goal or advance the plot in the work, reflecting the exchange logic structure of the work; Among them, the ultimate question motif parameter: represents the value of the ultimate philosophical or existential propositions (such as the meaning of life, identity and the boundaries of free will) explored beneath the surface plot of the work, reflecting the core ideological structure of the work; Among them, the narrative closure parameter is a numerical value that characterizes the completeness of the plot clues and causal logic at the end of the work (such as tight closure, cyclical repetition, and open suspension), reflecting the logical termination structure of the work. Among them, the ethical grayscale parameter is a value that represents the absoluteness or ambiguity of the core behaviors and choices in the work in terms of moral judgment (such as the clear-cut opposition of good and evil and the moral dilemma of choosing between two evils), reflecting the philosophical core and value tension of the work. Among them, the cognitive asymmetry parameter is a numerical value that characterizes the gap in the degree of information mastery between the narrator, the core character and the audience in the work and the resulting suspense (such as the tragic expectation of an omniscient perspective and the puzzle-solving suspense of a limited perspective), reflecting the essential narrative perspective of the work. Among them, the temporal structure parameter is a numerical value that characterizes the flow logic and transformation mode of the narrative time of the work (such as linear chronological order, fragmented jigsaw puzzle and circular narrative), reflecting the temporal perception structure of the work. Among them, the world logic paradigm parameter represents the value of the underlying law system (such as strict physical realism, magical realism and absurd logic) that governs the operation of the world set in the work, reflecting the causal basis of the work; Among them, the core desire paradigm parameter is a numerical value that represents the deepest inner desire (such as survival, truth, love, and power) that drives the core characters in the work to take action, reflecting the motivational root structure of the work. Among them, the narrative aggregation form parameter is a value that represents the interweaving and combination of multiple plot lines and character relationships in the work (such as single monologue, parallel interweaving of two lines, network group portrait and nested box), reflecting the macro-texture structure of the work. Among them, the emotional contract paradigm parameter is a numerical value that represents the potential experiential commitment and final fulfillment method established between the work and the audience at the beginning (such as the catharsis of tragedy, the verification of comedy, and the shock of reversal), reflecting the expected response structure of the work. Among them, the boundary breakthrough paradigm parameter is a numerical value that characterizes the attempt and result of the core character in the work to cross the boundaries of existing physical or social rules (such as successful prison break, compromise within the rules and breaking through the dimensional wall), reflecting the boundary-crossing dynamics of the work. Among them, the meta-narrative degree parameter is a value that represents the degree of self-reference and exposure of the work's own fictional attributes (such as immersive disregard, play-within-a-play structure and breaking the fourth wall), reflecting the work's reflexive structure of reality and illusion. Among them, the memory reconstruction paradigm parameter is a numerical value that characterizes the way past events in the work invade and reshape the present narrative (such as objective flashback, subjective trauma distortion and memory tampering), reflecting the historical perception structure of the work. Among them, the micro-power structure parameter is a value that represents the implicit domination and obedience relationship between characters in the core scene of the work (such as absolute patriarchal oppression, peer checks and balances, and alienation control), reflecting the field political structure of the work. Among them, the spatiotemporal perception deformation parameter is a value that characterizes the flow quality and spatial oppression (such as stagnation, fleeting moments and claustrophobia) of the narrative time in the subjective experience of the work, reflecting the experiential rhythm structure of the work. Among them, the symbol density parameter is a numerical value that characterizes the density of concrete objects in the work that carry abstract metaphors and thematic meanings (such as realistic depiction, partial symbolism and full allegorical structure), reflecting the depth of the work's meaning. Among them, the parameter of the violent aesthetics paradigm is a numerical value that represents the physical presentation logic and sensory transformation method (such as realistic bloodshed, dance-like freehand style and psychological oppression) when the conflict breaks out in the work, reflecting the aggressive release structure of the work. Among them, the redemption possibility parameter: represents the value that indicates whether there is a path to liberation (such as self-sacrifice for redemption, salvation by the arrival of others, and no redemption) in the work in a state of despair or depravity, and reflects the existential tendency of the work; Among them, the other's gaze parameter: a value that represents the sensitivity and reaction of the core character's behavior in the work to the evaluation of the fictional social group (such as ignoring others and doing one's own thing, extremely craving recognition and being alienated by the gaze), reflecting the social mirror structure of the work. Among them, the information deconstruction paradigm parameter is a numerical value that represents the deconstruction path (such as peeling an onion-like progression, jigsaw puzzle-like reconstruction, and complete subversion) that the work reveals to the audience to reveal the core mystery or the truth of the world, reflecting the cognitive unlocking structure of the work. Among them, the identity disguise and fluidity parameter: represents the degree of separation and disguise between the social identity and the true self of the core character in the work (such as the exposure of the true identity, hiding incognito, switching between multiple identities and the mask being integrated into the flesh and blood), and reflects the identity alienation structure of the work. Among them, the survival game paradigm parameter is a numerical value that represents whether the core plot of the work is subject to explicit mandatory rules or trial mechanisms (such as a free-developing open world, a closed space battle royale, and an infinite flow rule task and puzzle-solving adventure), reflecting the survival game structure of the work. Among them, the Primal Deprivation and Trauma Paradigm Parameter: represents the value of the most fundamental lack or early trauma type (such as loss of loved ones, deprivation of identity, sense of existencelessness and original sin curse) that drives the core character of the work to embark on the journey, reflecting the deep deprivation dynamic structure of the work. Among them, the connection and isolation paradigm parameter: a value that represents the emotional and existential connection state between the core character in the work and the world or others (such as absolute isolation and disconnection, parasitic symbiosis, networked interconnection and being exiled by the group), reflecting the topological structure of the work's sense of belonging. Among them, the language constructs reality parameter: the numerical value that represents the ability of the language symbol system in the work to reshape and intervene in objective reality (such as language as only a descriptive tool, a magical system where words become law, a power mechanism where naming is control, and language virus infecting thinking), reflects the symbolic power structure of the work; Among them, the body politics paradigm parameter is a value that represents the degree of social discipline and alienation of the physical body in the work (such as the body without natural intervention, the prosthetic body modified by technology, the body of a prisoner who is disciplined and punished, and the body as an object of commodity exchange), reflecting the body ontological structure of the work. Among them, the cosmic absurdity parameter: a value that represents the degree of meaningless resistance of the world in which the work is set against human rationality and effort (such as a rational world where effort is rewarded, an unpredictable world dominated by chance, and an absurd world of absolute meaninglessness), reflecting the existentialist underlying structure of the work. Among them, the time cycle and destiny parameters: represent the linear and cyclical properties of the passage of time in the work and the possibility of breaking free from destiny (such as unidirectional irreversible linearity, Sisyphus-like invalid cycle, closed-loop paradox and breaking the cycle), reflecting the time ontology structure of the work. Among them, the species and existence form parameters are numerical values ​​that characterize the life existence form and species boundaries of the core subject in the work (such as absolute human-centeredness, human-machine hybridity, consciousness-uploaded digital life and Cthulhu-like indescribable objects), reflecting the life form structure of the work. Among them, the parameters of group mob and awakening: represent the degree of blindness or awakening of the group characters as a whole consciousness in the work (such as the unconscious background, the incited mob, the collective unconscious sinking and the collective struggle of awakening), reflecting the group dynamics structure of the work. Among them, the taboo transgression paradigm parameter: represents the nature of the core behavior in the work's contact with the strongest taboo in the world (such as the taboo of incest, the taboo of godkilling, the taboo of forbidden fruit of knowledge and the interference of time paradox), and reflects the taboo transgression structure of the work; Among them, the sacrifice-compensation paradigm parameter: represents the value of the compensation level (such as win-win without sacrifice, sacrifice of material interests, sacrifice of intimate relationships and sacrifice of one's own soul) required to achieve the goals of the group or individual in the work, and reflects the compensation ethical structure of the work. Among them, the reality deconstruction level parameter is a numerical value that characterizes the depth of the work's peeling away of the so-called objective reality (such as surface physical reality, psychological projection reality, simulated program illusion and multiple parallel superposition states), reflecting the work's reality ontological structure. Among them, the order collapse paradigm parameter is a value that characterizes the way and speed at which the old order system in the work disintegrates (such as slow decay, instantaneous apocalyptic destruction, internal hollowing out, and replacement by alien rules), reflecting the entropy increase dynamic structure of the work. Among them, the spatial topological paradigm parameter is a numerical value that characterizes the opening and closing and connectivity topological structure of the narrative space of the work (such as the confinement of a closed room, the linear extension of a highway, and the jumping of a network of galaxies), reflecting the physical field tension structure of the work. Among them, the Fate Empowerment Paradigm Parameter: represents the value of the final decision-making power that drives the ending in the work (such as personal will to defy fate, divine / systemic fatalism and random chaos domination), reflecting the free will structure of the work; Among them, the breaking-the-state trigger paradigm parameter is a value that characterizes the nature of the core catalytic event that breaks the initial equilibrium state of the work (such as the intrusion of outsiders, the outbreak of internal rifts, and the triggering of unknown information), reflecting the unbalanced dynamic structure of the work. Among them, the attribution paradigm parameter represents the way the responsibility for evil or tragedy is identified and questioned in the work (such as personal moral depravity, systemic original sin, and the absurd arrival of a subjectless world), and reflects the ethical structure of the work's attribution of responsibility. Among them, the core erotic paradigm parameter is a numerical value that represents the deep essence and destructive tendency (such as redemptive complementarity, pathological symbiosis and destructive possession) of intimate relationships in the work, reflecting the emotional polarization structure of the work. Among them, the historical depth paradigm parameter is a value that characterizes the entanglement thickness between the current events of the work and the past history or myth (such as the absence of historical transients, intergenerational genetic trauma and the recurrence of myths), reflecting the temporal deposition structure of the work. Among them, the alienated other paradigm parameter is a numerical value that characterizes the essential attributes of the other as the opposite or threat in the work (such as comprehensible adversaries, indescribable cosmic fears and mirror self-projection), reflecting the antagonistic essential structure of the work. Among them, the truth production paradigm parameter is a numerical value that characterizes the way the core truth is acquired and established in the work (such as empirical discovery, power discourse construction, and the suspension of the eternally unknowable), reflecting the epistemological structure of the work. Among them, the intergenerational transmission paradigm parameter is a value that represents the transmission and rupture pattern of the core contradiction of the work between generations (such as the passing down of curses from generation to generation, the struggle for inheritance and the complete severance and restart), reflecting the vertical continuity structure of the work. Among them, the consciousness base paradigm parameter is a numerical value that represents the level of consciousness and the anchor point of reality (such as lucid objective reality, dream subconscious projection and delirium) at which the narrative of the work occurs, reflecting the subjective reality structure of the work. Among them, the comedic misalignment paradigm parameter is a numerical value that represents the underlying misalignment mechanism (such as identity misalignment, context misalignment, and logical misalignment) that generates the comedic or absurd feeling of the work, reflecting the absurdity generation structure of the work. Among them, the narrative scale paradigm parameter is a numerical value that represents the scope and level of influence of the core conflict of the work (such as the micro-level individual inner world, the meso-level family community and the macro-level civilization universe), reflecting the scale structure of the work. Among them, the hope-maintaining paradigm parameter is a numerical value that characterizes the way and resilience of the work in maintaining a spark of hope in desperate situations (such as the glimmer of a desperate counterattack, Sisyphus-like hopeless persistence and utter nihilism), reflecting the tragic resilience structure of the work. Among them, the mirror intertextuality paradigm parameter is a numerical value that characterizes the degree of symmetrical mapping and ironic response between the internal or external texts of the work (such as one-dimensional narrative, dual-line mirror contrast and parody subversion of classic texts), reflecting the intertextual structure of the work. Among them, the alienated labor paradigm parameter is a numerical value that characterizes the degree to which the core figure in the work loses control of his creation or labor results and is devoured by them (such as the unity of skill and results, the creator being enslaved by the creation and the complete alienation of labor), reflecting the creator's dilemma structure of the work. The mechanical projection encoding of the work can be selected from at least one of the following groups: spatiotemporal setting (spatiotemporal coordinates), style texture, character external configuration, ending performance parameters, cultural symbol system, language register, sensory emphasis, genre and length, rhythm segmentation, or a combination of the above. Among them, the spatiotemporal setting parameters are values ​​that represent at least one of the following aspects of the work: the historical background, geographical environment, or social class (such as Victorian London, cyberpunk futuristic city, polar wilderness), reflecting the external physical and social spatiotemporal coordinates of the work. Among them, the style texture parameter is a value that represents at least one of the sentence structure features, dialogue density, or narrative distance of the work (such as long and complicated descriptions, fast-paced dialogue, or omniscient perspective), reflecting the surface language texture and narrative tone of the work. Among them, the external configuration parameters of the characters are numerical values ​​that represent the combination of external attributes of the group of characters in the work (such as age span, occupational distribution, supporting character density, number of key characters), reflecting the scale of the cast and the external character design framework of the work. Among them, the ending performance parameter is a value that represents the surface emotional release state and ending rhythm (such as happy ending, bitter blank space, abrupt end, long epilogue) at the end of the work, reflecting the emotional coating of the work's ending. Among them, the cultural symbol system parameter represents the numerical value of the specific cultural matrix and symbol system (such as Western medieval mythology, Eastern martial arts world, Cthulhu indescribable, hard science fiction technology worship) on which the worldview of the work is based, reflecting the surface cultural clothing of the work. Among them, the language register parameter is a value that represents the language variants and lexical color (such as colloquialisms, official written language, internet slang, and classical Chinese) used in the work, reflecting the dialect layer and lexical style of the work. Among them, the sensory emphasis parameter is a value that represents the sensory channels and atmosphere creation tendency that the work emphasizes in its description (such as visual dominance of cool colors, noisy auditory rendering, and olfactory synesthetic stimulation), reflecting the environmental atmosphere coating of the work. Among them, the genre and length parameters are numerical values ​​that characterize the external literary genre and scale of the work (such as a micro-novel of a thousand words, a long serial of a million words, a three-act script format, or a prose poem), reflecting the physical encapsulation form of the work. Among them, the rhythm segmentation parameter is a value that represents the frequency of scene transitions and the granularity of chapter segmentation (such as continuous long shots with slow rhythm and high-frequency montage fragmented switching), reflecting the external arrangement rhythm of the work.

Claims

1. A method for compressing and encoding literary works based on configurable factorization, characterized in that, Includes the following steps: (a) Extract at least two sets of structural feature parameters from the target literary work, wherein the at least two sets of structural feature parameters are: At least the first group is a set of numerical values ​​representing the creative structural decisions of the work (defined as "what kind of story is told"), selected from at least one of the following groups: emotional pattern intensity, character agency, value orientation, narrative dynamics, narrative skeleton (narrative arc shape), degree of irreversibility, core conflict paradigm, ethical grayness, cognitive asymmetry parameter, core relationship, character arc paradigm, core cost, ultimate question, narrative closure, ethical grayness, cognitive asymmetry, temporal structure, narrative reliability, world logic paradigm, core desire paradigm, narrative aggregation form, emotional contract paradigm, boundary breaking paradigm, meta-narrative degree, memory reconstruction paradigm, micro-power structure, spatiotemporal perception deformation, symbolic density, violence. Aesthetic paradigm, possibility of redemption, gaze of the other, information deconstruction paradigm, identity mobility, survival game paradigm, primordial scarcity paradigm, connection isolation paradigm, language constructing reality paradigm, body politics paradigm, cosmic absurdity, time cycle destiny paradigm, existence form paradigm, group awakening, taboo transgression paradigm, sacrifice and compensation paradigm, reality deconstruction level, order collapse paradigm, spatial topology paradigm, destiny empowerment paradigm, breakthrough inducement paradigm, guilt attribution paradigm, core eroticism paradigm, historical depth paradigm, alienated other paradigm, truth production paradigm, intergenerational inheritance paradigm, consciousness base paradigm, comedic misplacement paradigm, narrative scale paradigm, hope maintenance paradigm, mirror intertextuality paradigm, alienated labor paradigm, or a recombination of the above contents. Among them, the emotional pattern intensity parameter: represents the intensity of each of the at least two emotional patterns in the work, reflecting the core emotional tone of the work; Among them, the character agency parameter: a value that represents the degree to which the narrative subject actively initiates behavior or passively endures external forces in the work (such as active resistance and sacrifice versus passive victimization and deception), reflecting the action source structure of the work; Among them, the value orientation parameter: represents the value object attribute (such as the value of the emotional individual and the value of the rational supra-individual) to which the core conflict of the work points, and reflects the value anchor structure of the work; Among them, the narrative dynamics parameter is a numerical value that characterizes the degree to which characters in the work actively drive the plot or are driven by external forces, reflecting the core action structure of the work. Among them, the narrative skeleton parameter is a value that represents the density of plot twists, the position of the climax, or the shape of the narrative arc of the work, reflecting the core story structure of the work. Among them, the irreversibility parameter is a value that characterizes the degree to which the core choices or core losses in the work are irreversible (such as fully reversible trial and error versus absolutely irreversible sacrifice), reflecting the intensity of the work's sense of destiny and the weight of the choices. Among them, the core conflict paradigm parameter is a numerical value that represents the underlying antagonistic logic type (such as individual versus system, individual versus inner self, and human versus nature) that drives the plot development in the work, reflecting the underlying dynamic model of the work. Among them, the ethical grayscale parameter is a numerical value that represents the absoluteness or ambiguity of the core behaviors and choices in the work in terms of moral judgment, reflecting the philosophical core and value tension of the work. Among them, the cognitive asymmetry parameter is a numerical value that characterizes the gap in the degree of information mastery between the narrator, the core character and the audience in the work and the resulting suspense, reflecting the essential narrative perspective of the work. Among them, the core relationship parameter is a numerical value that represents the deepest interaction patterns and interdependence states (such as symbiotic redemption, power oppression, and mirror intertextuality) between the core characters in the work, reflecting the interpersonal dynamics of the work. Among them, the character arc paradigm parameter is a value that represents the direction of the inner spiritual evolution trajectory of the core character in the work from the beginning to the end of the narrative (such as positive redemption, negative depravity and adherence to one's original intention), reflecting the character evolution structure of the work. Among them, the core cost paradigm parameter represents the value of the essential sacrifices (such as physical sacrifice, moral depravity, and identity erasure) that must be made in order to achieve the goal or advance the plot in the work, reflecting the exchange logic structure of the work; Among them, the ultimate question motif parameter: represents the value of the ultimate philosophical or existential propositions (such as the meaning of life, identity and the boundaries of free will) explored beneath the surface plot of the work, reflecting the core ideological structure of the work; Among them, the narrative closure parameter is a numerical value that characterizes the completeness of the plot clues and causal logic at the end of the work (such as tight closure, cyclical repetition, and open suspension), reflecting the logical termination structure of the work. Among them, the ethical grayscale parameter is a value that represents the absoluteness or ambiguity of the core behaviors and choices in the work in terms of moral judgment (such as the clear-cut opposition of good and evil and the moral dilemma of choosing between two evils), reflecting the philosophical core and value tension of the work. Among them, the cognitive asymmetry parameter is a numerical value that characterizes the gap in the degree of information mastery between the narrator, the core character and the audience in the work and the resulting suspense (such as the tragic expectation of an omniscient perspective and the puzzle-solving suspense of a limited perspective), reflecting the essential narrative perspective of the work. Among them, the temporal structure parameter is a numerical value that characterizes the flow logic and transformation mode of the narrative time of the work (such as linear chronological order, fragmented jigsaw puzzle and circular narrative), reflecting the temporal perception structure of the work; Among them, the world logic paradigm parameter represents the value of the underlying law system (such as strict physical realism, magical realism and absurd logic) that governs the operation of the world set in the work, reflecting the causal basis of the work; Among them, the core desire paradigm parameter is a numerical value that represents the deepest inner desire (such as survival, truth, love, and power) that drives the core characters in the work to take action, reflecting the motivational root structure of the work. Among them, the narrative aggregation form parameter is a value that represents the interweaving and combination of multiple plot lines and character relationships in the work (such as single monologue, parallel interweaving of two lines, network group portrait and nested box), reflecting the macro-texture structure of the work. Among them, the emotional contract paradigm parameter is a numerical value that represents the potential experiential commitment and final fulfillment method established between the work and the audience at the beginning (such as the catharsis of tragedy, the verification of comedy, and the shock of reversal), reflecting the expected response structure of the work. Among them, the boundary breakthrough paradigm parameter is a numerical value that characterizes the attempt and result of the core character in the work to cross the boundaries of existing physical or social rules (such as successful prison break, compromise within the rules and breaking through the dimensional wall), reflecting the boundary-crossing dynamics of the work. Among them, the meta-narrative degree parameter is a value that represents the degree of self-reference and exposure of the work's own fictional attributes (such as immersive disregard, play-within-a-play structure and breaking the fourth wall), reflecting the work's reflexive structure of reality and illusion. Among them, the memory reconstruction paradigm parameter is a numerical value that characterizes the way in which past events in the work invade and reshape the present narrative (such as objective flashback, subjective trauma distortion and memory tampering), reflecting the historical perception structure of the work. Among them, the micro-power structure parameter is a value that represents the implicit dominance and obedience relationship between characters in the core scene of the work (such as absolute patriarchal oppression, peer checks and balances, and alienation control), reflecting the field political structure of the work. Among them, the spatiotemporal perception deformation parameter is a value that characterizes the flow quality and spatial oppression (such as stagnation, fleeting moments and claustrophobia) of the narrative time in the subjective experience of the work, reflecting the experiential rhythm structure of the work. Among them, the symbol density parameter is a numerical value that characterizes the density of concrete objects in the work that carry abstract metaphors and thematic meanings (such as realistic depiction, partial symbolism and full allegorical structure), reflecting the depth of the work's meaning. Among them, the parameter of the violent aesthetics paradigm is a numerical value that represents the physical presentation logic and sensory transformation method (such as realistic bloodshed, dance-like freehand style and psychological oppression) when the conflict breaks out in the work, reflecting the aggressive release structure of the work. Among them, the redemption possibility parameter: represents the value that indicates whether there is a path to liberation (such as self-sacrifice for redemption, salvation by the arrival of others, and no redemption) in the work in a state of despair or depravity, and reflects the existential tendency of the work; Among them, the other's gaze parameter: a value that represents the sensitivity and reaction of the core character's behavior in the work to the evaluation of the fictional social group (such as ignoring others and doing one's own thing, extremely craving recognition and being alienated by the gaze), reflecting the social mirror structure of the work. Among them, the information deconstruction paradigm parameter is a numerical value that represents the deconstruction path (such as peeling an onion-like progression, jigsaw puzzle-like reconstruction, and complete subversion) that the work reveals to the audience to reveal the core mystery or the truth of the world, reflecting the cognitive unlocking structure of the work. Among them, the identity disguise and fluidity parameter: represents the degree of separation and disguise between the social identity and the true self of the core character in the work (such as the exposure of the true identity, hiding incognito, switching between multiple identities and the mask being integrated into the flesh and blood), and reflects the identity alienation structure of the work. Among them, the survival game paradigm parameter is a numerical value that represents whether the core plot of the work is subject to explicit mandatory rules or trial mechanisms (such as a free-developing open world, a closed space battle royale, and an infinite flow rule task and puzzle-solving adventure), reflecting the survival game structure of the work. Among them, the Primal Deprivation and Trauma Paradigm Parameter: represents the value of the most fundamental lack or early trauma type (such as loss of loved ones, deprivation of identity, sense of existencelessness and original sin curse) that drives the core character of the work to embark on the journey, reflecting the deep deprivation dynamic structure of the work. Among them, the connection and isolation paradigm parameter: a value that represents the emotional and existential connection state between the core character in the work and the world or others (such as absolute isolation and disconnection, parasitic symbiosis, networked interconnection and being exiled by the group), reflecting the topological structure of the work's sense of belonging. Among them, the language constructs reality parameter: the numerical value that represents the ability of the language symbol system in the work to reshape and intervene in objective reality (such as language as only a descriptive tool, a magical system where words become law, a power mechanism where naming is control, and language virus infecting thinking), reflects the symbolic power structure of the work; Among them, the body politics paradigm parameter is a value that represents the degree of social discipline and alienation of the physical body in the work (such as the body without natural intervention, the prosthetic body modified by technology, the body of a prisoner who is disciplined and punished, and the body as an object of commodity exchange), reflecting the body ontological structure of the work. Among them, the cosmic absurdity parameter: a value that represents the degree of meaningless resistance of the world in which the work is set against human rationality and effort (such as a rational world where effort is rewarded, an unpredictable world dominated by chance, and an absurd world of absolute meaninglessness), reflecting the existentialist underlying structure of the work. Among them, the time cycle and destiny parameters: represent the linear and cyclical properties of the passage of time in the work and the possibility of breaking free from destiny (such as unidirectional irreversible linearity, Sisyphus-like invalid cycle, closed-loop paradox and breaking the cycle), reflecting the time ontology structure of the work. Among them, the species and existence form parameters are numerical values ​​that characterize the life existence form and species boundaries of the core subject in the work (such as absolute human-centeredness, human-machine hybridity, consciousness-uploaded digital life and Cthulhu-like indescribable objects), reflecting the life form structure of the work. Among them, the parameters of group mob and awakening: represent the degree of blindness or awakening of the group characters as a whole consciousness in the work (such as the unconscious background, the incited mob, the collective unconscious sinking and the collective struggle of awakening), reflecting the group dynamics structure of the work. Among them, the taboo transgression paradigm parameter: represents the nature of the core behavior in the work's contact with the strongest taboo in the world (such as the taboo of incest, the taboo of godkilling, the taboo of forbidden fruit of knowledge and the interference of time paradox), and reflects the taboo transgression structure of the work; Among them, the sacrifice-compensation paradigm parameter: represents the value of the compensation level (such as win-win without sacrifice, sacrifice of material interests, sacrifice of intimate relationships and sacrifice of one's own soul) required to achieve the goals of the group or individual in the work, and reflects the compensation ethical structure of the work. Among them, the reality deconstruction level parameter is a numerical value that characterizes the depth of the work's peeling away of the so-called objective reality (such as surface physical reality, psychological projection reality, simulated program illusion and multiple parallel superposition states), reflecting the work's reality ontological structure. Among them, the order collapse paradigm parameter is a value that characterizes the way and speed at which the old order system in the work disintegrates (such as slow decay, instantaneous apocalyptic destruction, internal hollowing out, and replacement by alien rules), reflecting the entropy increase dynamic structure of the work. Among them, the spatial topological paradigm parameter is a numerical value that characterizes the opening and closing and connectivity topological structure of the narrative space of the work (such as the confinement of a closed room, the linear extension of a highway, and the jumping of a network of galaxies), reflecting the physical field tension structure of the work. Among them, the Fate Empowerment Paradigm Parameter: represents the value of the final decision-making power that drives the ending in the work (such as personal will to defy fate, divine / systemic fatalism and random chaos domination), reflecting the free will structure of the work; Among them, the breaking-the-state trigger paradigm parameter is a value that characterizes the nature of the core catalytic event that breaks the initial equilibrium state of the work (such as the intrusion of outsiders, the outbreak of internal rifts, and the triggering of unknown information), reflecting the unbalanced dynamic structure of the work. Among them, the attribution paradigm parameter represents the way the responsibility for evil or tragedy is identified and questioned in the work (such as personal moral depravity, systemic original sin, and the absurd arrival of a subjectless world), and reflects the ethical structure of the work's attribution of responsibility. Among them, the core erotic paradigm parameter is a numerical value that represents the deep essence and destructive tendency (such as redemptive complementarity, pathological symbiosis and destructive possession) of intimate relationships in the work, reflecting the emotional polarization structure of the work. Among them, the historical depth paradigm parameter is a value that characterizes the entanglement thickness between the current events of the work and the past history or myth (such as the absence of historical transients, intergenerational genetic trauma and the recurrence of myths), reflecting the temporal deposition structure of the work. Among them, the alienated other paradigm parameter is a numerical value that characterizes the essential attributes of the other as the opposite or threat in the work (such as comprehensible adversaries, indescribable cosmic fears and mirror self-projection), reflecting the antagonistic essential structure of the work. Among them, the truth production paradigm parameter is a numerical value that characterizes the way the core truth is acquired and established in the work (such as empirical discovery, power discourse construction, and the suspension of the eternally unknowable), reflecting the epistemological structure of the work. Among them, the intergenerational transmission paradigm parameter is a value that represents the transmission and rupture pattern of the core contradiction of the work between generations (such as the passing down of curses from generation to generation, the struggle for inheritance and the complete severance and restart), reflecting the vertical continuity structure of the work. Among them, the consciousness base paradigm parameter is a numerical value that represents the level of consciousness and the anchor point of reality (such as lucid objective reality, dream subconscious projection and delirium) in which the narrative of the work occurs, reflecting the subjective reality structure of the work. Among them, the comedic misalignment paradigm parameter is a numerical value that represents the underlying misalignment mechanism (such as identity misalignment, context misalignment, and logical misalignment) that generates the comedic or absurd feeling of the work, reflecting the absurdity generation structure of the work. Among them, the narrative scale paradigm parameter is a numerical value that represents the scope and level of influence of the core conflict of the work (such as the micro-level individual inner world, the meso-level family community and the macro-level civilization universe), reflecting the scale structure of the work. Among them, the hope-maintaining paradigm parameter is a numerical value that characterizes the way and resilience of the work in maintaining a spark of hope in desperate situations (such as the glimmer of a desperate counterattack, Sisyphus-like hopeless persistence and utter nihilism), reflecting the tragic resilience structure of the work. Among them, the mirror intertextuality paradigm parameter is a numerical value that characterizes the degree of symmetrical mapping and ironic response between the internal or external texts of the work (such as one-dimensional narrative, dual-line mirror contrast and parody subversion of classic texts), reflecting the intertextual structure of the work. Among them, the alienated labor paradigm parameter is a numerical value that characterizes the degree to which the core figure in the work loses control of his creation or labor results and is devoured by them (such as the unity of skill and results, the creator being enslaved by the creation and the complete alienation of labor), reflecting the creator's dilemma structure of the work. At least the second group is a set of numerical values ​​(defining "under what external conditions the story is told") that characterize the mechanically traversable setting features of the work, selected from at least one of the following groups: spatiotemporal setting (spatiotemporal coordinates), style and texture, character external configuration, ending performance parameters, cultural symbol system, language register, sensory emphasis, genre and length, rhythm segmentation, or a combination of the above. Among them, the spatiotemporal setting parameter is a value that represents at least one of the following aspects of the work: the historical background, geographical environment, or social class (such as Victorian London, cyberpunk futuristic city, polar wilderness), reflecting the external physical and social spatiotemporal coordinates of the work. Among them, the style texture parameter is a value that represents at least one of the sentence structure features, dialogue density, or narrative distance of the work (such as long and complicated descriptions, fast-paced dialogue, or omniscient perspective), reflecting the surface language texture and narrative tone of the work. Among them, the external configuration parameters of the characters are numerical values ​​that represent the combination of external attributes of the group of characters in the work (such as age range, occupational distribution, supporting character density, number of key characters), reflecting the scale of the cast and the external character design framework of the work. Among them, the ending performance parameter is a value that represents the surface emotional release state and ending rhythm (such as happy ending, bitter blank space, abrupt end, long epilogue) at the end of the work, reflecting the emotional coating of the work's finale. Among them, the cultural symbol system parameter represents the numerical value of the specific cultural matrix and symbol system (such as Western medieval mythology, Eastern martial arts world, Cthulhu indescribable, hard science fiction technology worship) on which the worldview of the work is based, reflecting the surface cultural clothing of the work. Among them, the language register parameter is a value that represents the language variants and lexical color (such as colloquialisms, official written language, internet slang, and classical Chinese) used in the work, reflecting the dialect layer and lexical style of the work. Among them, the sensory emphasis parameter is a value that represents the sensory channels and atmosphere creation tendency that the work emphasizes in its description (such as visual dominance of cool colors, noisy auditory rendering, and olfactory synesthetic stimulation), reflecting the environmental atmosphere coating of the work. Among them, the genre and length parameters are numerical values ​​that characterize the external literary genre and scale of the work (such as a micro-novel of a thousand words, a long serial of a million words, a three-act script format, or a prose poem), reflecting the physical encapsulation form of the work. Among them, the rhythm segmentation parameter is a value that represents the frequency of scene transitions and the granularity of chapter segmentation (such as continuous long shots with slow rhythm and high-frequency montage fragmented switching), reflecting the external arrangement rhythm of the work. There is no mandatory correspondence between the groups contained in the first group and the second group. The same group can be assigned to the first group or the second group in different implementations. The allocation relationship given in a single implementation constitutes the factor configuration scheme of that encoding. (b) Generate the first set of structural feature parameters into a first factor coding segment and the second set of structural feature parameters into a second factor coding segment—the first factor coding segment and the second factor coding segment are distinguishable from each other in a compact coding string and can be stored, transmitted, indexed and replaced independently. (c) Combining at least two factor coding segments, such as the first factor coding segment and the second factor coding segment mentioned above, to generate a factorized compact coding string—the length of the factorized compact coding string is less than the length of the original work text, and the number of factor coding segments may be greater than or equal to 2 but is not limited to 2; (d) Output the factorized compact encoded string as a structured compressed representation of the literary work.

2. A method for reconstructing a literary work based on the encoding described in claim 1, characterized in that: (a) Receive or acquire a factorized compact encoded string generated according to the method of claim 1; (b) Based on the factor configuration scheme, recover the first set of structural feature parameters and the second set of structural feature parameters from the factorized compact encoded string, respectively; (c) Convert the first set of structural feature parameters into a first generation control command and the second set of structural feature parameters into a second generation control command; (d) The first generation control instruction and the second generation control instruction are combined and input into the text generation model to produce a reconstructed text that approximates the target literary work at the structural skeleton level.

3. A method for generating literary works based on the encoding described in claim 1, characterized in that: (a) Receive a first factorization code—containing a first factor code segment F1 and a second factor code segment S1; (b) Fixed factor traversal—keeping F1 unchanged, replacing S1 with at least one different second factor encoding segment S2, and generating a factor variant work for each replacement combination (F1, Sᵢ) according to the method of claim 2—the factor variant work inherits the creative structural decision represented by F1, but occurs under a different external setting represented by Sᵢ; (c) Alternatively, keep S1 unchanged and replace F1 to generate a reverse traversal variant; (d) Alternatively, combine the k first factor coding segments with the m second factor coding segments by Cartesian product, and generate a cross-variant work for each combination, for a total of k × m works; (e) Record the first and second factor sources of each variant work in the genealogical database.

4. A system for compression encoding and traversal reconstruction of literary works based on configurable factorization, characterized in that, include: (a) A parameter extraction module, used to extract the first and second set of structural feature parameters as described in claim 1 from the target literary work according to a preset factor configuration scheme; (b) A factor encoding module, used to generate a first factor encoding segment from the first set of parameters, generate a second factor encoding segment from the second set of parameters, and combine the two into a factorized compact encoding string; (c) Decoding and reconstruction module, used to recover the sets of parameters from the factorized compact encoded string and convert them into generation control instructions; (d) A traversal generation module, used to perform fixed factor traversal or Cartesian product cross combination on the first factor coding segment and the second factor coding segment according to the method of claim 3; (e) A factor configuration module for receiving and storing the factor configuration scheme, which defines the allocation relationship of each group to a first factor or a second factor, and the allocation relationship can be reconfigured in different implementations.

5. The method according to claim 1, characterized in that: The at least two sets of structural feature parameters are two groups: the first group (creativity factor) includes at least four groups from emotional pattern intensity, character agency, value orientation, narrative arc shape and irreversibility; the second group (traversability factor) includes at least one group from spatiotemporal coordinates and style texture.

6. The method according to claim 1, characterized in that: The at least two sets of structural feature parameters are two groups: the first group (creativity factor) includes the intensity of emotional patterns and character agency; the second group (traversability factor) includes all five groups in the narrative arc shape, character external configuration, spatiotemporal coordinates, ending performance parameters, and style texture, wherein the narrative arc shape and character external configuration are preset traversed by statistically analyzing common templates of best-selling novels.

7. The method according to claim 1, characterized in that: The at least two sets of structural feature parameters are three or more sets—including a first factor composed of emotional pattern intensity, a second factor composed of narrative arc shape and character configuration, and a third factor composed of spatiotemporal coordinates and style texture; each set is distinguished from the other by a separator, and one or two sets can be arbitrarily fixed while traversing the remaining sets during generation.

8. The method according to claim 1, characterized in that: The factorized compact encoding string also includes a factor configuration identifier, which identifies the factor configuration scheme used when generating the encoding. The factor configuration identifier enables the decoding module to correctly map each encoding segment to the corresponding semantic group when recovering parameters, and supports the simultaneous use of multiple different factor configuration schemes in the same system.

9. The method according to claim 2, characterized in that: In step (d), the temperature parameter is set to zero and a fixed random seed is used for the text generation model, so that multiple reconstructions of the same factorized compact encoded string are substantially consistent in text content.

10. The method according to claim 3, characterized in that: The traversal generation supports dual traversal—for any first factor encoding segment F, its dual encoding segment ¬F is generated by calculating complementary values ​​for each dimension contained therein (mapping the value of each dimension to the opposite endpoint of its allowed range); and a dual variant work that is a mirror image of the original work in terms of emotional structure is generated by combining (¬F, S).

Images