Similar Items: Beyond Confidence: Rethinking Self-Assessments for Performance Prediction in LLMs
- Beyond Benchmarks: MathArena as an Evaluation Platform for Mathematics with LLMs
- Beyond "I cannot fulfill this request": Alleviating Rigid Rejection in LLMs via Label Enhancement
- LLMs Improving LLMs: Agentic Discovery for Test-Time Scaling
- The First Token Knows: Single-Decode Confidence for Hallucination Detection
- Grounded or Guessing? LVLM Confidence Estimation via Blind-Image Contrastive Ranking
- Misaligned by Reward: Socially Undesirable Preferences in LLMs