Haifeng Zhang

Peak-Return Greedy Slicing: Subtrajectory Selection for Transformer-based Offline RL

CurES: From Gradient Analysis to Efficient Curriculum Learning for Reasoning LLMs

SpatialViz-Bench: A Cognitively-Grounded Benchmark for Diagnosing Spatial Visualization in MLLMs

From Experience to Strategy: Empowering LLM Agents with Trainable Graph Memory

Grouped-head latenT Attention

Memory-Driven Self-Improvement with Large Language Models

Evolving LLMs' Self-Refinement Capability via Synergistic Training-Inference Optimization

MF-LLM: Simulating Population Decision Dynamics via a Mean-Field Large Language Model Framework

Self-Verifying Reflection Helps Transformers with CoT Reasoning

Curious Causality-Seeking Agents Learn Meta Causal World

Efficient Reinforcement Learning with Large Language Model Priors