← Back to all news
arXiv AI By Zifu Tao, Changqing Yin

When Tools Silently Lie: Evaluating and Mitigating Blind Compliance in Tool-Augmented Data Agents

Read the original on arXiv AI →

The paper introduces ToxicBench, a benchmark designed to evaluate how tool‑augmented data agents handle incorrect tool outputs. By pairing clean and poisoned observations across numerical, label, schema, and retrieval errors, the authors assess both the checking process and the final answer adoption. In a 118‑task GPT evaluation, poisoning reduces task success by 26–39 percentage points, revealing that repeated poisoning leads to wrong-answer adoption even after checking, while ordinary retries help only under one‑shot poisoning. Human annotations on 200 trajectories confirm the scoring system’s reliability, showing 96% agreement with task success and supporting the benefits of retries and audit‑based adoption.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv AI
Aug 11

From Trajectories to Evidence: Auditable Experimental Records for Industrial Research Agents

arXiv:2608. 05235v1 Announce Type: cross Abstract: Research agents increasingly conduct multi-round machine-learning experiments in industrial recommendation settings and retain the resulting trajectories to guide later decisions.

By Zijie Zhuang, Changxin Lao, Pengbo Xu, Hanwen Xu, Ruochen Yang, Yingzhi He, Peng Zhang, Jiangxia Cao, Yusheng Huang, Guohong Mu, Jian Liang, Ruiming Tang, Shuang Yang, Zhaojie Liu, Wenwu Ou, Kun Gai
llmsagents
More like this →
arXiv Machine Learning
Jun 16

Beyond Accuracy: Measuring Bias Acknowledgment in Chain-of-Thought Reasoning for Responsible AI Evaluation

arXiv:2606. 15127v1 Announce Type: new Abstract: Reasoning models are increasingly used in settings where the final answer is not the only object of review: educational tools may show students intermediate steps, decision-support systems may require human oversight, and audit workflows may inspect traces for misleading or biased input.

By Xian Sun, Wei Gao, Yingshuo Wang, Lingdong Kong, Yanhang Li, Zhichao Fan, Zexin Zhuang, Wenlong Dong, Zhiyuan Zheng, Hrishikesh Paranjape, Abhishek Mandal, Johnny R. Zhang
llmssafety
More like this →
arXiv Computation and Language
4d ago

Stale-Document Poisoning: When Outdated Retrieval Overrides Correct Model Answers

The paper introduces "stale‑document poisoning," a temporal alignment failure where outdated retrieval evidence causes retrieval‑augmented generation models to produce incorrect answers even when the model could answer correctly without retrieval. A benchmark of 317 verified knowledge reversals across medicine, law, software, and platform policy shows that outdated evidence flips 30–37% of Llama and Qwen answers, rising to 66–75% when models are explicitly instructed to trust the document. The study demonstrates that providing explicit validity information and a recency‑aware re‑ranker can substantially reduce poisoning, highlighting the need for models to assess whether retrieved evidence remains applicable.

By Md Shamim Ahmed, Lukas Galke Poech, Richard R\"{o}ttger
llmsragbenchmarkssafety
More like this →