← authors · overview

• MixSD: Mixed Contextual Self-Distillation for Knowledge Injection
  2026
• STRIDE: Training Data Attribution via Sparse Recovery from Subset Perturbations
  2026
• One Word at a Time: Incremental Completion Decomposition Breaks LLM Safety
  2026
• Difficulty-Based Preference Data Selection by DPO Implicit Reward Gap
  2025
• When Seeing Overrides Knowing: Disentangling Knowledge Conflicts in Vision-Language Models
  2025
• Improving Large Language Model Safety with Contrastive Representation Learning
  2025
• Revisiting Multi-Agent Debate as Test-Time Scaling: A Systematic Study of Conditional Effectiveness
  2025
• Accidental Vulnerability: Factors in Fine-Tuning that Shift Model Safeguards
  2025
• DARS: Dynamic Action Re-Sampling to Enhance Coding Agent Performance by Adaptive Tree Traversal
  2025
• OPT-R: Exploring the Role of Explanations in Finetuning and Prompting for Reasoning Skills of Large Language Models
  2023
• Can Large Language Models Infer Causation from Correlation?
  2023