Karthik R Narasimhan

$\tau^2$-bench: : Evaluating Conversational Agents in a Dual-Control Environment

Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting

VideoGameBench: Can Vision-Language Models complete popular video games?

IMPersona: Evaluating Individual Level LLM Impersonation

When Models Know More Than They Can Explain: Quantifying Knowledge Transfer in Human-AI Collaboration

{$\tau$}-bench: A Benchmark for \underline{T}ool-\underline{A}gent-\underline{U}ser Interaction in Real-World Domains

EnIGMA: Interactive Tools Substantially Assist LM Agents in Finding Security Vulnerabilities

SWE-bench Multimodal: Do AI Systems Generalize to Visual Software Domains?

Contextual Experience Replay for Continual Learning of Language Agents

Code-of-thought prompting: Probing AI Safety with Code