PPL
Xiang Wang
Papers (11)
PetroBench: A Benchmark for Large Language Models in Petroleum EngineeringWhat Makes Chain-of-Thought Work at Probe Time? Local Co-occurrence Rather Than Global DerivationTiny Brains, Giant Impact: Uncovering the Keystone Neurons of LLM with Just a Few PromptsFaithful-MR1: Faithful Multimodal Reasoning via Anchoring and Reinforcing Visual AttentionReasoning Can Be Restored by Correcting a Few Decision TokensTo Call or Not to Call: Diagnosing Intrinsic Over-Calling Bias in LLM AgentsTeaching Large Language Models When Not to Know: Learning Temporal Critique for Ex-Ante ReasoningR^2-Mem: Reflective Experience for Memory SearchUniVLR: Unifying Text and Vision in Visual Latent Reasoning for Multimodal LLMsSelf-ReSET: Learning to Self-Recover from Unsafe Reasoning TrajectoriesInternalizing Safety Understanding in Large Reasoning Models via Verification