Valentina Pyatkin

PluriHarms: Benchmarking the Full Spectrum of Human Judgments on AI Harm

RewardBench 2: Advancing Reward Model Evaluation

Tulu 3: Pushing Frontiers in Open Language Model Post-Training

WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild

2 OLMo 2 Furious (COLM’s Version)

SafetyAnalyst: Interpretable, Transparent, and Steerable Safety Moderation for AI Behavior

Diverging Preferences: When do Annotators Disagree and do Models Know?

Diverging Preferences: When do Annotators Disagree and do Models Know?

Hybrid Preferences: Learning to Route Instances for Human vs. AI Feedback

SafetyAnalyst: Interpretable, transparent, and steerable LLM safety moderation