research

My research lies at the intersections of: reliable agents and tool use, language model interpretability and control, red teaming and reliable evaluation, and reasoning. Click on the map below to explore more :)

Reliable Agents & Tool Use

LM Interpretability
& Control

Red Teaming & Reliable Evaluation

Reasoning

TicToc ACL’26Reliable Agents & Tool Use Unreliable Tool Preferences EMNLP’25Reliable Agents & Tool Use STAC SLM-Agents @ NeurIPS’26Reliable Agents & Tool Use Knowing–Doing Gap NeurIPS’26Tool Use · Interpretability & Control SteerBoost Under reviewLM Interpretability & Control ThinkARM ACL’26Interpretability & Control · Reasoning DyePack EMNLP’25Red Teaming & Reliable Evaluation Adversarial Paraphrasing NeurIPS’25Red Teaming & Reliable Evaluation Accessibility Profiles arXiv’26Red Teaming & Reliable Evaluation Thinking Hard, Not Smart Under reviewReliable Evaluation · Reasoning Failing to Explore Under reviewReasoning
Swipe sideways to explore the map.