The paper investigates the reliability of tool‑using agents, focusing on two failure modes: selecting the wrong tool and constructing incorrect arguments. It introduces a correct‑invocation rate metric to distinguish these errors and evaluates five open‑weight models on multi‑step tasks up to depth 8, finding that by depth 6 about 70% of a model’s clean‑context capability is lost due to earlier mistakes. The study reveals that exact‑match scoring against a fixed gold trajectory forces severity and recovery parameters to extreme values, and proposes a conditional‑on‑state scoring remedy that yields more realistic severity estimates.
By Afiya Noorain, Subhranshu Mohanty, Amritesh Banerjee, Abhijit Dasgupta
arXiv:2608. 02786v1 Announce Type: new Abstract: AI systems can fail silently.
By Priyanka Bajaj (Independent Researcher)
The paper introduces RuVerBench, a benchmark with 2,458 instances for evaluating the reliability of Large Language Models acting as judges (LaaJ) in verifying rubric compliance within agentic scenarios such as deep research and agentic coding. It systematically meta‑evaluates frontier LLMs, revealing that even the most advanced models perform well yet still produce substantial noise. The study also examines how prompt design, batching, and majority voting affect verification accuracy, noting that weaker models are more prompt‑sensitive, batched verification trades accuracy for efficiency, and majority voting offers diminishing returns.
By Yangda Peng, Yunjia Qi, Haotian Xia, Guanzhong He, Xintong Shi, Richeng Xuan, Songyuanyi Lu, Yixian Liu, Zhichao Hu, Yuhong Liu, Hao Peng
arXiv:2607. 16122v1 Announce Type: new Abstract: Evaluations should do more than measure a models current performance.
By Vipul Gupta, Zihao Wang, Razvan-Gabriel Dumitru, MohammadHossein Rezaei, Aakash Sabharwal, Yunzhong He
arXiv:2608.21601v1 Announce Type: new
Abstract: Benchmarks for scientific artificial intelligence are mostly written to be scored: multiple-choice questions, curated agent tasks with reference soluti...
By Aubrey Brueckner, Darshil Patel, Yuhuan He, Timothy Kassis
arXiv:2608.22808v2 Announce Type: replace
Abstract: When can an agent failure be caught? An audit is usually limited by the record rather than by the method. CatchBench therefore puts one auditor's q...
By Yue Zhao
arXiv:2607. 17044v1 Announce Type: cross Abstract: Multi-step enterprise agent tasks fail in a characteristic way: single-pass inference has no checkpoint between deciding an answer and committing to it.
By Arunabh Dastidar (for the Leni Team)
arXiv:2606. 09046v1 Announce Type: new Abstract: Useful audits reveal not only how often a model fails, but also where its failures concentrate.
By Vyzantinos Repantis, Ameya Gawde, Harshvardhan Singh
arXiv:2606. 09165v1 Announce Type: new Abstract: Safety judges are increasingly deployed to evaluate model outputs against evolving criteria, yet recent meta-evaluation work shows they remain brittle under prompt and rubric variation, with false negative-rate swings of up to 0.
By Yongtaek Lim, Hyeji Choi, Minwoo Kim
arXiv:2608. 07528v1 Announce Type: new Abstract: Linear probes detect corrupted context in language models with near-perfect accuracy, yet this does not translate into reliable failure prediction.
By Jyotin Goel, Ipshita Bandyopadhyay, Justin Shenk
arXiv:2602.13576v2 Announce Type: replace-cross
Abstract: Evaluation and alignment pipelines for large language models increasingly rely on LLM-based judges, whose behavior is guided by natural-langu...
By Ruomeng Ding, Yifei Pang, He Sun, Yizhong Wang, Zhiwei Steven Wu, Zhun Deng
arXiv:2607. 28802v1 Announce Type: new Abstract: Existing evaluations often reduce agent failures to system-level outcomes, obscuring where the fault originated and which intervention would improve the agent system.
By Harsh Raj, Vipul Gupta, Anas Mahmoud, Razvan-Gabriel Dumitru, Darvin Yi, Aakash Sabharwal, Yunzhong He