Lizhu Zhang

Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment

Let it Calm: Exploratory Annealed Decoding for Verifiable Reinforcement Learning

StreamMem: Query-Agnostic KV Cache Memory for Streaming Video Understanding

Exploring System-1 and System-2 Communication for Latent Reasoning in LLMs

Thought Communication in Multiagent Collaboration

S'MoRE: Structural Mixture of Residual Experts for Parameter-Efficient LLM Fine-tuning