Similar Items: Misaligned by Reward: Socially Undesirable Preferences in LLMs
- Mitigating Misalignment Contagion by Steering with Implicit Traits
- LLMs Improving LLMs: Agentic Discovery for Test-Time Scaling
- Correct Is Not Enough: Training Reasoning Planners with Executor-Grounded Rewards
- Beyond Benchmarks: MathArena as an Evaluation Platform for Mathematics with LLMs
- Beyond Confidence: Rethinking Self-Assessments for Performance Prediction in LLMs
- SERE: Structural Example Retrieval for Enhancing LLMs in Event Causality Identification