Memory systems are essential for personalized Large Language Models (LLMs). However, retrieval within these systems primarily relies on semantic similarity, potentially missing logically critical memories with limited semantic overlap. We refer to this failure as semantic deflection, which remains insufficiently evalua...
Hongxun Ding, Xiang-Yu Yu, Cheng-Bing Wang et al.· 1 citation
Reinforcement learning framework Knowledge Weaver is proposed, a reinforcement learning framework that trains a language model to curate reusable knowledge from agent trajectories and couple feedback inspired by token-wise mutual information with marginal success rewards to guide knowledge accumulation.
Yu-Yang Zhao, Li-Zi Liao, Le-Yang Shen et al.· 0 citations
A series of models are explored that connect item Semantic IDs with natural language in a unified representation space and seek to unlock the potential of natural-language chain-of-thought (CoT) reasoning for recommendation, including OneReason.
Removing unreliable messages recovers part of the lost accuracy, suggesting that communication should be selective based on upstream reliability and the evidence already available to the downstream agent.
Ya-Xin Gong, Gang-Yi Zhang, Chong-Ming Gao et al.· 0 citations
CREDO reads the confidence from a dedicated token pair in the model's output distribution and trains it by differentiable regression, turning the trained confidence into a signal for accuracy, weighting rollouts by how far confidence and outcome disagree, so that accuracy and calibration improve together.
Chen-Xiao Fan, Chong-Ming Gao, Gang-Yi Zhang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.