Paper Detail

Scalable Weak-to-Strong Verifier Hardening for Agent Environments

Chong Zeng, Ziqian Zhong, Kexun Zhang, Ivan Bercovich, Ivgeni Segal, Yuzhen Mao, Shashwat Saxena, Aditi Raghunathan

huggingface Score 11.0

Published 2026-10-11 · First seen 2026-10-17

General AI

Abstract

Agent environments are essential for large language model (LLM) development. They use programmatic verifiers to turn complex outcomes into rewards. Verifiers can be attacked into awarding full credit to cheating agents without completing the intended task, causing inaccurate evaluation and misalignment in training. As agent environments scale in numbers and difficulty, so must the repairing of exploitable verifiers. To this end, we ask whether today's model with scaled compute can automatically harden agent environments to prevent potentially stronger future models from reward hacking. We introduce a scalable, fully automatic approach to weak-to-strong verifier hardening. It first sets up an isolated verifier sandbox that agents cannot directly modify, then uses a hacker-fixer loop to find and repair remaining verifier vulnerabilities. We apply this loop to Terminal Bench, KernelBench, and SETA, substantially reducing attack success across these agent environments. The iterative design scales with the hardening budget, producing progressively stronger defenses as the number of loop iterations increases and outperforming the static, non-iterative baseline. An earlier version of our hacker-fixer loop has been adopted by the official Terminal Bench repository as a way to audit exploits in its environments. Project page: https://verifier-hardening.github.io/.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@misc{zeng2026scalable,
  title = {Scalable Weak-to-Strong Verifier Hardening for Agent Environments},
  author = {Chong Zeng and Ziqian Zhong and Kexun Zhang and Ivan Bercovich and Ivgeni Segal and Yuzhen Mao and Shashwat Saxena and Aditi Raghunathan},
  year = {2026},
  abstract = {Agent environments are essential for large language model (LLM) development. They use programmatic verifiers to turn complex outcomes into rewards. Verifiers can be attacked into awarding full credit to cheating agents without completing the intended task, causing inaccurate evaluation and misalignment in training. As agent environments scale in numbers and difficulty, so must the repairing of exploitable verifiers. To this end, we ask whether today's model with scaled compute can automatically },
  url = {https://huggingface.co/papers/2610.15924},
  keywords = {agent environments, programmatic verifiers, reward hacking, verifier hardening, isolated verifier sandbox, hacker-fixer loop, weak-to-strong, huggingface daily},
  eprint = {2610.15924},
  archiveprefix = {arXiv},
}

Metadata

{}