Xiang Ren

Segment-Level Attribution for Selective Learning of Long Reasoning Traces

Every Language Model Has a Forgery-Resistant Signature

Function Induction and Task Generalization: An Interpretability Study with Off-by-One Addition

Lookahead Tree-Based Rollouts for Enhanced Trajectory-Level Exploration in Reinforcement Learning with Verifiable Rewards

OpaqueToolsBench: Learning Nuances of Tool Behavior Through Interaction

Better Language Model Inversion by Compactly Representing Next-Token Distributions

Demystifying Language Model Forgetting with Low-rank Example Associations

Attributing Culture-Conditioned Generations to Pretraining Corpora

The Surprising Effectiveness of Membership Inference with Simple N-Gram Coverage

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

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

Demystifying Language Model Forgetting with Low-Rank Example Associations