Public articles linked to the same research event.
arXiv The work frames cognitive-algorithm discovery as a program refinement problem: human-created cognitive models are expressed as probabilistic programs and given to a system of LLM agents that identify mismatches between model and behavior, propose code-level modifications within researcher-specified constraints, and verify structural fidelity, with revisions propagated to a probabilistic inference module for latent-variable inference and data likelihood computation; evaluated on human behavior in a problem-solving paradigm that exposes a variety of cognitive algorithms, revised models consistently improve model fit relative to ancestral models and reveal a small set of recurring innovations that capture meaningful behavioral variability in this task.
The work frames cognitive-algorithm discovery as a program refinement problem: human-created cognitive models are expressed as probabilistic programs and given to a system of LLM agents that identify mismatches between model and behavior, propose code-level modifications within researcher-specified constraints, and verify structural fidelity, with revisions propagated to a probabilistic inference module for latent-variable inference and data likelihood computation; evaluated on human behavior in a problem-solving paradigm that exposes a variety of cognitive algorithms, revised models consistently improve model fit relative to ancestral models and reveal a small set of recurring innovations that capture meaningful behavioral variability in this task.
The work frames cognitive-algorithm discovery as a program refinement problem: human-created cognitive models are expressed as probabilistic programs and given to a system of LLM agents that identify mismatches between model and behavior, propose code-level modifications within researcher-specified constraints, and verify structural fidelity, with revisions propagated to a probabilistic inference module for latent-variable inference and data likelihood computation; evaluated on human behavior in a problem-solving paradigm that exposes a variety of cognitive algorithms, revised models consistently improve model fit relative to ancestral models and reveal a small set of recurring innovations that capture meaningful behavioral variability in this task.
The work frames cognitive-algorithm discovery as a program refinement problem: human-created cognitive models are expressed as probabilistic programs and given to a system of LLM agents that identify mismatches between model and behavior, propose code-level modifications within researcher-specified constraints, and verify structural fidelity, with revisions propagated to a probabilistic inference module for latent-variable inference and data likelihood computation; evaluated on human behavior in a problem-solving paradigm that exposes a variety of cognitive algorithms, revised models consistently improve model fit relative to ancestral models and reveal a small set of recurring innovations that capture meaningful behavioral variability in this task.