collaborators

10 papers

cs.CR2026

Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops

Ziqian Zhong, Ivgeni Segal, Ivan Bercovich +3

Agent benchmarks score submissions with outcome verifiers that are typically hand-written and brittle, leaving them open to reward hacking. We audit 1,968 tasks across five termina…

cs.LG2026

Self-Trained Verification for Training- and Test-Time Self-Improvement

Chen Henry Wu, Aditi Raghunathan

Self-improvement at scale has been a longstanding goal for reasoning models, and there are two natural places to do it: at test time, through verification-refinement (V-R) loops; a…

cs.AI2026

Understanding and Mitigating Premature Confidence for Better LLM Reasoning

Jingchu Gai, Guanning Zeng, Christina Baek +4

Long chains of thought (CoT) from current language models frequently contain logical gaps and unjustified leaps, limiting the gains from additional test-time compute. Improving rea…

cs.LG2026

Watch the Weights: Unsupervised monitoring and control of fine-tuned LLMs

Ziqian Zhong, Aditi Raghunathan

The releases of powerful open-weight large language models (LLMs) are often not accompanied by access to their full training data. Existing interpretability methods, particularly t…

cs.CR2026

Terminal Wrench: A Dataset of 331 Reward-Hackable Environments and 3,632 Exploit Trajectories

Ivan Bercovich, Ivgeni Segal, Kexun Zhang +3

We release Terminal Wrench, a subset of 331 terminal-agent benchmark environments, copied from the popular open benchmarks that are demonstrably reward-hackable. The data set inclu…

cs.AI2026

Hodoscope: Unsupervised Monitoring for AI Misbehaviors

Ziqian Zhong, Shashwat Saxena, Aditi Raghunathan

Existing approaches to monitoring AI agents rely on supervised evaluation: human-written rules or LLM-based judges that check for known failure modes. However, novel misbehaviors m…