Salvato in:
Dettagli Bibliografici
Autore principale: Ravindran, Santhosh Kumar
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2510.18895
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909863111557120
author Ravindran, Santhosh Kumar
author_facet Ravindran, Santhosh Kumar
contents We introduce CosmoCore, a neuroscience-inspired reinforcement learning (RL) architecture that integrates affective signals to enhance code generation in large language models (LLMs). Motivated by human and animal learning where embarrassment from mistakes drives rapid correction, as observed in training a puppy to avoid repeating errors after a single scolding CosmoCore tags code generation trajectories with valence and surprise using a lightweight multi-layer perceptron (MLP). High-negative valence (cringe) episodes, such as buggy code outputs, are prioritized in a Dream Queue for five-fold replay during off-policy updates, while low-surprise successes are pruned to prevent overconfidence and buffer bloat. Evaluated on code generation benchmarks like HumanEval and BigCodeBench, alongside simulations with a custom data pipeline environment, CosmoCore reduces hallucinated code (e.g., syntax errors or logical bugs) by 48\% and accelerates self-correction by 45\%. Local experiments using Hugging Face models in a PySpark environment validate these gains, with code snippets provided for replication. Ablations confirm valence tagging boosts curiosity in exploration, and pruning mitigates inefficiency. This framework extends RL from human feedback (RLHF) for more emotionally aware code assistants, with applications in IDEs and data pipelines. Code and the custom mini-world simulation are released.
format Preprint
id arxiv_https___arxiv_org_abs_2510_18895
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CosmoCore Affective Dream-Replay Reinforcement Learning for Code Generation
Ravindran, Santhosh Kumar
Software Engineering
Artificial Intelligence
Human-Computer Interaction
We introduce CosmoCore, a neuroscience-inspired reinforcement learning (RL) architecture that integrates affective signals to enhance code generation in large language models (LLMs). Motivated by human and animal learning where embarrassment from mistakes drives rapid correction, as observed in training a puppy to avoid repeating errors after a single scolding CosmoCore tags code generation trajectories with valence and surprise using a lightweight multi-layer perceptron (MLP). High-negative valence (cringe) episodes, such as buggy code outputs, are prioritized in a Dream Queue for five-fold replay during off-policy updates, while low-surprise successes are pruned to prevent overconfidence and buffer bloat. Evaluated on code generation benchmarks like HumanEval and BigCodeBench, alongside simulations with a custom data pipeline environment, CosmoCore reduces hallucinated code (e.g., syntax errors or logical bugs) by 48\% and accelerates self-correction by 45\%. Local experiments using Hugging Face models in a PySpark environment validate these gains, with code snippets provided for replication. Ablations confirm valence tagging boosts curiosity in exploration, and pruning mitigates inefficiency. This framework extends RL from human feedback (RLHF) for more emotionally aware code assistants, with applications in IDEs and data pipelines. Code and the custom mini-world simulation are released.
title CosmoCore Affective Dream-Replay Reinforcement Learning for Code Generation
topic Software Engineering
Artificial Intelligence
Human-Computer Interaction
url https://arxiv.org/abs/2510.18895