8 results about "Selective attention" patented technology

A closed-loop selective attention system for resolving conflicts in multi-source or multi-speaker environments, including a plurality of internal attention models, each outputting a probability distribution over candidate sources and an associated confidence score, a fuser detecting conflicts when two or more of said attention models output high-confidence predictions that disagree, a selective sampling policy querying one or more external agents, wherein each external agent possesses a knowledge base, a reliability model, and a communication protocol, a trust and reliability module assigning and updating dynamic trust scores for internal and external agents based on past performance, an efficiency optimizer minimizing communication overhead and decision delay by balancing token usage cost and latency cost, and a dynamical system formulator ensuring convergence of the conflict resolution process under bounded trust, decaying step size, and limited sampling.

A selective attention system for multi-source or multi-speaker environments, including a plurality of independent selective attention (SA) models, each configured to output a probability distribution over a plurality of sources and a confidence score, a modular framework allowing said SA models to be added, replaced, upgraded or blocked in real time without interrupting system operation, an output fuser configured to combine outputs from said SA models based on at least one of model confidence and dynamically learned reliability scores, and a routing mechanism that directs user attention to a single source at a time based on the fused output.

Process for selective attention in a conversation with multiple participants, each participant being either a human or an apparatus, including processing outputs from a plurality of sensors, each sensor sensing a different modality, detecting communication modalities of the multiple participants, a communication modality being either speech or sign language, determining a plurality of probability distributions of source participants to whom a specific participant is paying attention, according to a respective plurality of models, fusing the plurality of probability distributions into a confident probability distribution, defining a confidence level for the confident probability distribution, applying natural language processing to convert each participant's communication to text, clustering the participants into conversation groups, deriving an intended source participant to whom the specific participant is paying attention, according to a source participant with a highest expected utility, and suppressing audio received from participants other than from the intended source participant.

Systems, methods and computer-readable media for providing explainable selective attention in multi-source or multi-speaker environments. A selective attention module receives multimodal sensor data including audio, video, gaze, text, and physiological signals from a plurality of sources. An attention inference engine generates attention distributions over the sources and fuses them into a probabilistic belief state. An explainability module produces interpretable outputs corresponding to the fused belief, including attention matrices, confidence scores, reliability measures, margin-based differentiators, and natural language rationales. The explainability outputs are rendered through visual, auditory, or augmented / virtual reality interfaces to indicate the attended source, suppressed sources, and reasoning for the selection. The system enables user interaction by providing justifications in real time, logging explanations for retrospective analysis, and supporting adaptation of thresholds and model weights based on feedback. The disclosed technology improves transparency, interpretability, and trust in selective attention systems, while maintaining real-time performance in dynamic multi-speaker environments.

A user's attention capabilities can be assessed in a virtual environment. An electronic device, such as a head-mounted display, can display a plurality of visual stimuli concurrently in a 3D virtual environment, and each visual stimulus can be displayed at a position in the 3D virtual environment according to a display scheme. The electronic device can obtain a stream of sensor data measured by the one or more sensors, and can determine a plurality of sequential user responses to the plurality of visual stimuli based on the stream of sensor data. Based on the plurality of sequential user responses, the electronic device can determine an attention indicator indicating an attention capability of the user associated with the electronic device to different visual stimuli.

A context-aware selective attention system for multi-source or multi-speaker environments, including a natural language processor analyzing linguistic context of signals, an attention-shifter that enables transitions between sources even while a prior source remains active, a profile-learner constructing and updating semantic and behavioral profiles of sources, including speech features, emotional tone, and past attentional importance, an adaptive thresholder adjusting attention-decision thresholds based on the profiles, environmental context and model reliability, and a conversation graph manager maintaining a dynamic graph of sources and attentional relationships in real time.

Methods, systems, and apparatuses, including computer programs encoded on computer storage media, for processing an input sequence (or new input token) using a neural network that includes one or more attention layer blocks that each include a selective attention layer. By generating and using aggregated mask scores to modify attention logits used by an attention mechanism, the neural network's performance is improved. Further, by using the aggregated mask scores to also remove one or more context tokens from a context buffer; and by pruning the context buffer based on the aggregated mask scores, the memory cost and computation processing cost of using the neural network to process an input sequence (or new input token) is simultaneously improved.

This invention discloses a cross-modal target perception method based on data feature-based collaborative selective attention. It employs a neuromorphic collaborative selective attention mechanism that combines prior information and modal characteristics to quickly and accurately select effective cross-modal information. Based on this mechanism, a vision-radar selective attention target perception network model is constructed, including a primary and secondary modality definition module, a data-level visual perception information selection module, and a binary corrector for adaptive feature selection. By selecting primary and secondary modal data and converting the data into features, the extracted features are processed using the binary corrector to achieve effective feature focusing. Prediction is then performed, outputting the position and type of radar point targets in the image. This achieves efficient and real-time radar-vision cross-modal target detection, ensuring high accuracy and low latency in target detection.