PROBE is an algorithm that trains language‑model auditors to probe deployed tool‑using agents for policy violations through realistic, multi‑turn, tool‑mediated dialogue. By framing auditing as a two‑player Markov game and using a composite reward that balances violation discovery, behavioral plausibility, conversation progress, and policy diversity, PROBE outperforms zero‑shot adversaries in uncovering broader and higher‑yield policy violations. The study compares GRPO, DPO, and GFlowNet optimizations, demonstrating that multi‑turn behavioral auditing is a distinct and necessary axis of agent evaluation, complementing existing capability and safety benchmarks.
By Kshitij Mishra, Abhijith Sharma, Nils Lukas, Salem Lahlou
The paper presents a reinforcement learning approach to enhance large language model (LLM) auditors for alignment tasks. By training policies that investigate target models for hidden behaviors and using an LLM judge to compare investigations, the method improves audit realism and reduces false positives. Experiments show better performance on adversarially fine‑tuned targets and a low false‑positive rate below 1%.
By Paul Rosu, Rowan Wang
Users of a deployed language model routinely encounter behaviours that testing almost never surfaces, since deployment puts the model through orders of magnitude more interactions than any evaluation...
arXiv:2601. 16398v3 Announce Type: replace-cross Abstract: Algorithmic audits are essential tools for examining systems for properties required by regulators or desired by operators.
By Hannah Cyberey, Yangfeng Ji, David Evans
arXiv:2608.31105v1 Announce Type: new
Abstract: Users of a deployed language model routinely encounter behaviours that testing almost never surfaces, since deployment puts the model through orders of...
By Adrians Skapars, Edoardo Manino
arXiv:2608. 05212v1 Announce Type: new Abstract: Deep search agents tackle challenging questions through long-horizon web interactions, a process that is both complex and fragile: small reasoning errors may propagate through long, noisy trajectories into fluent but incorrect answers.
By Zhixiang Liang, Yifei Liu, Yidan Huang, Haozhe Zhao, Beichen Huang, Jiaqi Wang, Nan Duan, Qiong Cao
arXiv:2510. 06096v3 Announce Type: replace Abstract: The objectives that Large Language Models (LLMs) implicitly optimize remain dangerously opaque, making trustworthy alignment and auditing a grand challenge.
By Matthieu Bou, Nyal Patel, Arjun Jagota, Satyapriya Krishna, Sonali Parbhoo
arXiv:2609.33676v2 Announce Type: replace
Abstract: LLM agents increasingly take consequential actions through interactions with users, policies, and external tools. Auditing these agents requires au...
By Yifan Liu, Praveen Venkateswaran, Abdulhamid Adebayo, Dong Wang
Safety tuning pipelines judge only the final answer, which makes it difficult to distinguish robust refusal from two undesirable shortcuts: blanket refusal on benign requests and polished but unfaithf...
AUDITPLAN introduces a plan-then-answer method for safety alignment in language models, where the model first generates a compact structured safety plan before responding. The plan includes a threat label, intended action, and explicit constraints, allowing machine‑checkable auditing while remaining hidden from end users. Training combines supervised fine‑tuning with reinforcement learning using the FAITHGATE reward, which only rewards correct plans, thereby reducing unsafe shortcuts and improving robustness across Qwen model variants.
By Sai Sri Pushpa Jampani, Kshitij Mishra, Asif Ekbal
AuditKIT is a source‑available evaluation library that unifies benchmark accuracy, judge‑based assessment, runtime performance, tool‑use behavior, and retrieval quality into a single result representation. It offers direction‑aware metric contracts, configuration fingerprints, and extensible metrics and backends, while providing turn‑aware tool‑call matching and distinguishing verified task outcomes from trajectory‑level diagnostics. Seven case studies demonstrate how answer extraction, runtime behavior, and metric selection can alter evaluation outcomes across model comparison, serving runtimes, retrieval, tool calling, and interoperability.
By Shreeyans Arora, Utsav Avaiya, Zera Lyngkhoi, Hem Gosalia, Ram Mohan Rao Kadiyala, Vinay Kumar Sankarapu, Pratinav Seth
arXiv:2606. 19714v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as judges for open-ended generation, as large-scale human evaluation is often expensive and difficult to scale, yet their preferences remain imperfect proxies for human judgment.
By Zilong Zhang, Yi-Ting Hung, Weiyi He, Junxi Zhang, Lei Ding, Chi-Kuang Yeh