AI.info

Chenglong Wang

Explore Chenglong Wang on AI.info.

• PlugMem: A Task-Agnostic Plugin Memory Module for LLM Agents
• RM-Distiller: Exploiting Generative LLM for Reward Model Distillation
• SERM: Self-Evolving Relevance Model with Agent-Driven Learning from Massive Query Streams
• Probing Preference Representations: A Multi-Dimensional Evaluation and Analysis Method for Reward Models
• MRO: Enhancing Reasoning in Diffusion Language Models via Multi-Reward Optimization