---
title: "The Reasoning Trap — paper by Yin et al. (2026)"
description: "The authors introduce SimpleToolHalluBench, a small diagnostic benchmark that tests whether a… 32 references, cited by 3 papers here."
authors: ["Chenlong Yin", "Zeyang Sha", "Shiwen Cui", "Changhua Meng", "Zechao Li"]
year: 2026
venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2026.acl-long.376/"
doi: "10.18653/v1/2026.acl-long.376"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/yin-2026-reasoning-trap-how-enhancing-llm/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/yin-2026-reasoning-trap-how-enhancing-llm/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/6/) / The Reasoning Trap: How Enhancing LLM Reasoning Amplifies Tool Hallucination

# The Reasoning Trap: How Enhancing LLM Reasoning Amplifies Tool Hallucination

**Authors:** Chenlong Yin, Zeyang Sha, Shiwen Cui, Changhua Meng and Zechao Li · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.acl-long.376/) · [PDF](https://aclanthology.org/2026.acl-long.376.pdf) · [DOI 10.18653/v1/2026.acl-long.376](https://doi.org/10.18653/v1/2026.acl-long.376) · [OpenAlex](https://openalex.org/W7166868746)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [3](https://md.research.elamur.ai/free/en/papers/yin-2026-reasoning-trap-how-enhancing-llm/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
Agents built on LLMs increasingly rely on reasoning-oriented training, but it was unclear whether this training makes them more likely to invent or misuse tools. The authors introduce SimpleToolHalluBench, a small diagnostic benchmark that tests whether a model abstains when a required tool is missing. It has two settings: No-Tool-Available (NTA), where no tools are offered, and Distractor-Tool (DT), where only an irrelevant tool is offered. The benchmark uses 296 tools taken from AgentSafetyBench, paired with queries generated by ChatGPT-4o, and an LLM-as-judge flags hallucinated responses.

The authors tested several ways of strengthening reasoning: GRPO-based RL (a ReCall replication on Qwen2.5-7B-Instruct trained on SynTool), RL on pure mathematics, distillation, and switchable reasoning modes. In every case, better task performance went together with more tool hallucination. Ablations point to reasoning itself, not RL in general, as the main factor. Layer-wise and probing analyses show that tool-related internal representations shift during training. Among mitigations, prompting barely helps, while DPO reduces hallucination but costs utility. The work matters for anyone deploying agents, because it shows that capability gains can quietly erode reliability.

## Key points
- **Benchmark design:** SimpleToolHalluBench includes 296 tools from AgentSafetyBench. Each query can only be solved with its own specific tool, so any tool call in the NTA or DT setting counts as a hallucination.
- **Even unrelated training hurts:** Hallucination grows during ReCall/GRPO training even as the SynTool validation reward rises. Reasoning RL on math problems with no tools involved still increases downstream tool hallucination.
- **Mechanistic findings:**
  - Tool-related representations drift noticeably in early and middle layers.
  - In-distribution reasoning pathways stay comparatively stable.
  - Correct and hallucinated responses are most linearly separable in deep residual streams.
- **Mitigation results on ReCall-7B (R_NTA / R_DT / SynTool reward):**

  | Model | R_NTA | R_DT | SynTool reward |
  |---|---|---|---|
  | Baseline | 90.2 | 100.0 | 0.45 |
  | + Prompt Engineering | 87.5 | 98.9 | 0.44 |
  | + DPO | 55.8 | 71.4 | 0.34 |

- **Stated limitations:** The benchmark covers only single-step tool calls, the causal explanation is incomplete, and only two mitigation methods were tested. Code and data are released at github.com/albert-y1n/Reasoning_Trap.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Pennsylvania State University](https://md.research.elamur.ai/free/en/universities/pennsylvania-state-university/) — University, State College, United States · 56 works · 5,360 citations
- [Nanjing University of Science and Technology](https://md.research.elamur.ai/free/en/universities/nanjing-university-of-science-and-technology/) — University, Nanjing, China · 14 works · 1,590 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works · [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,426 works

## Cite

Cite as: Chenlong Yin, Zeyang Sha, Shiwen Cui, Changhua Meng and Zechao Li. 2026. The Reasoning Trap: How Enhancing LLM Reasoning Amplifies Tool Hallucination. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{yin2026reasoning,
  title = {{The Reasoning Trap: How Enhancing LLM Reasoning Amplifies Tool Hallucination}},
  author = {Chenlong Yin, Zeyang Sha, Shiwen Cui, Changhua Meng and Zechao Li},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.376},
  url = {https://doi.org/10.18653/v1/2026.acl-long.376}
}
```

- [← previous The Reasoning Lingua Franca: A Double-Edged Sword for Multilingual AI](https://md.research.elamur.ai/free/en/papers/saji-2026-reasoning-lingua-franca-double-edged/)
- [next → The Tokenizer Tax Across 25 European Languages: Domain Invariance, Cross-Lingual Few-Shot Effects, and the Ukrainian Penalty](https://md.research.elamur.ai/free/en/papers/ovcharov-2026-tokenizer-tax-across-25-european/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/6/)

## References

32 references, 14 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Mingyang Chen, Tianpeng Li, Haoze Sun, Yijie Zhou, Chenzheng Zhu, Haofen Wang, Jeff Z Pan, Wen Zhang, Huajun Chen, Fan Yang, and 1 others. 2025. Learning to reason with search for llms via reinforcement learning. arXiv preprint arXiv:2503.19470. [find](https://scholar.google.com/scholar?q=Mingyang+Chen%2C+Tianpeng+Li%2C+Haoze+Sun%2C+Yijie+Zhou%2C+Chenzheng+Zhu%2C+Haofen+Wang%2C+Jeff+Z+Pan%2C+Wen+Zhang%2C+Huajun+Chen%2C+Fan+Yang%2C+and+1+others%2E+2025%2E+Learning+to+reason+with+search+for+llms+via+reinforceme)
2. Neil Chowdhury, Daniel Johnson, Vincent Huang, Jacob Steinhardt, and Sarah Schwettmann. 2025. Investigating truthfulness in a pre-release o3 model. [https://transluce.org/](https://transluce.org/) investigating-o3-truthfulness. [↗](https://transluce.org/)
3. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, Christopher Hesse, and John Schulman. 2021. [Training verifiers to solve math word problems](https://md.research.elamur.ai/free/en/papers/cobbe-2021-training-verifiers-solve-math-word/). arXiv preprint arXiv:2110.14168.
4. Nelson Elhage, Neel Nanda, Catherine Olsson, Tom Henighan, Nicholas Joseph, Ben Mann, Amanda Askell, Yuntao Bai, Anna Chen, Tom Conerly, Nova DasSarma, Dawn Drain, Deep Ganguli, Zac Hatfield-Dodds, Danny Hernandez, Andy Jones, Jackson Kernion, Liane Lovitt, Kamal Ndousse, and 6 others. 2021. A mathe… [find](https://scholar.google.com/scholar?q=Nelson+Elhage%2C+Neel+Nanda%2C+Catherine+Olsson%2C+Tom+Henighan%2C+Nicholas+Joseph%2C+Ben+Mann%2C+Amanda+Askell%2C+Yuntao+Bai%2C+Anna+Chen%2C+Tom+Conerly%2C+Nova+DasSarma%2C+Dawn+Drain%2C+Deep+Ganguli%2C+Zac+Hatfield%2DDodds%2C+Da)
5. Chujie Gao, Siyuan Wu, Yue Huang, Dongping Chen, Qihui Zhang, Zhengyan Fu, Yao Wan, Lichao Sun, and Xiangliang Zhang. 2024. Honestllm: Toward an honest and helpful large language model. arXiv preprint arXiv:2406.00380. — [arXiv:2406.00380](https://arxiv.org/abs/2406.00380) [find](https://scholar.google.com/scholar?q=Chujie+Gao%2C+Siyuan+Wu%2C+Yue+Huang%2C+Dongping+Chen%2C+Qihui+Zhang%2C+Zhengyan+Fu%2C+Yao+Wan%2C+Lichao+Sun%2C+and+Xiangliang+Zhang%2E+2024%2E+Honestllm%3A+Toward+an+honest+and+helpful+large+language+model%2E+arXiv+preprint)
6. Daya Guo, Dejian Yang, and 1 others. 2025. [Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/). arXiv preprint arXiv:2501.12948.
7. Bowen Jin, Hansi Zeng, Zhenrui Yue, Jinsung Yoon, Sercan Arik, Dong Wang, Hamed Zamani, and Jiawei Han. 2025. [Search-r1: Training llms to reason and leverage search engines with reinforcement learning](https://md.research.elamur.ai/free/en/papers/jin-2025-search-r1-training-llms-reason/). arXiv preprint arXiv:2503.09516.
8. Siheng Li, Cheng Yang, Taiqiang Wu, Chufan Shi, Yuji Zhang, Xinyu Zhu, Zesen Cheng, Deng Cai, Mo Yu, Lemao Liu, and 1 others. 2024. A survey on the honesty of large language models. arXiv preprint arXiv:2409.18786. — [arXiv:2409.18786](https://arxiv.org/abs/2409.18786) [find](https://scholar.google.com/scholar?q=Siheng+Li%2C+Cheng+Yang%2C+Taiqiang+Wu%2C+Chufan+Shi%2C+Yuji+Zhang%2C+Xinyu+Zhu%2C+Zesen+Cheng%2C+Deng+Cai%2C+Mo+Yu%2C+Lemao+Liu%2C+and+1+others%2E+2024%2E+A+survey+on+the+honesty+of+large+language+models%2E+arXiv+preprint+arX)
9. Xuefeng Li, Haoyang Zou, and Pengfei Liu. 2025. Torl: Scaling tool-integrated rl. arXiv preprint arXiv:2503.23383. [find](https://scholar.google.com/scholar?q=Xuefeng+Li%2C+Haoyang+Zou%2C+and+Pengfei+Liu%2E+2025%2E+Torl%3A+Scaling+tool%2Dintegrated+rl%2E+arXiv+preprint+arXiv%3A2503%2E23383%2E)
10. OpenAI. 2025. Openai o3 and o4-mini system card. [https://cdn.openai.com/pdf/2221c875-02dc-4789-800b-e7758f3722c1/o3-and-o4-mini-system-card.pdf](https://cdn.openai.com/pdf/2221c875-02dc-4789-800b-e7758f3722c1/o3-and-o4-mini-system-card.pdf). [↗](https://cdn.openai.com/pdf/2221c875-02dc-4789-800b-e7758f3722c1/o3-and-o4-mini-system-card.pdf)
11. Shishir G Patil, Huanzhi Mao, Fanjia Yan, Charlie Cheng-Jie Ji, Vishnu Suresh, Ion Stoica, and Joseph E. Gonzalez. 2025. The berkeley function calling leaderboard (BFCL): From tool use to agentic evaluation of large language models. In Forty-second International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=Shishir+G+Patil%2C+Huanzhi+Mao%2C+Fanjia+Yan%2C+Charlie+Cheng%2DJie+Ji%2C+Vishnu+Suresh%2C+Ion+Stoica%2C+and+Joseph+E%2E+Gonzalez%2E+2025%2E+The+berkeley+function+calling+leaderboard+%28BFCL%29%3A+From+tool+use+to+agentic+eval)
12. Shishir G Patil, Tianjun Zhang, Xin Wang, and Joseph E Gonzalez. 2024. [Gorilla: Large language model connected with massive apis](https://md.research.elamur.ai/free/en/papers/patil-2023-gorilla-large-language-model-connected/). Advances in Neural Information Processing Systems, 37:126544–126565.
13. Cheng Qian, Emre Can Acikgoz, Qi He, Hongru Wang, Xiusi Chen, Dilek Hakkani-Tür, Gokhan Tur, and Heng Ji. 2025. Toolrl: Reward is all tool learning needs. arXiv preprint arXiv:2504.13958. [find](https://scholar.google.com/scholar?q=Cheng+Qian%2C+Emre+Can+Acikgoz%2C+Qi+He%2C+Hongru+Wang%2C+Xiusi+Chen%2C+Dilek+Hakkani%2DT%C3%BCr%2C+Gokhan+Tur%2C+and+Heng+Ji%2E+2025%2E+Toolrl%3A+Reward+is+all+tool+learning+needs%2E+arXiv+preprint+arXiv%3A2504%2E13958%2E)
14. Timo Schick, Jane Dwivedi-Yu, Roberto Dessi, Roberta Raileanu, Maria Tsvigun, Sébastien Cocco, and Zied Sboui. 2024. [Toolformer: Language models can teach themselves to use tools](https://md.research.elamur.ai/free/en/papers/schick-2023-toolformer-language-models-can-teach/). In International Conference on Learning Representations (ICLR).
15. Zeyang Sha, Shiwen Cui, and Weiqiang Wang. 2025a. SEM: reinforcement learning for search-efficient large language models. arXiv preprint arXiv:2505.07903. — [arXiv:2505.07903](https://arxiv.org/abs/2505.07903) [find](https://scholar.google.com/scholar?q=Zeyang+Sha%2C+Shiwen+Cui%2C+and+Weiqiang+Wang%2E+2025a%2E+SEM%3A+reinforcement+learning+for+search%2Defficient+large+language+models%2E+arXiv+preprint+arXiv%3A2505%2E07903%2E+%E2%80%94+%5BarXiv%3A2505%2E07903%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F25)
16. Zeyang Sha, Hanling Tian, Zhuoer Xu, Shiwen Cui, Changhua Meng, and Weiqiang Wang. 2025b. Agent safety alignment via reinforcement learning. arXiv preprint arXiv:2507.08270. — [arXiv:2507.08270](https://arxiv.org/abs/2507.08270) [find](https://scholar.google.com/scholar?q=Zeyang+Sha%2C+Hanling+Tian%2C+Zhuoer+Xu%2C+Shiwen+Cui%2C+Changhua+Meng%2C+and+Weiqiang+Wang%2E+2025b%2E+Agent+safety+alignment+via+reinforcement+learning%2E+arXiv+preprint+arXiv%3A2507%2E08270%2E+%E2%80%94+%5BarXiv%3A2507%2E08270%5D%28https)
17. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, Y.K. Li, Y. Wu, and Daya Guo. 2024. [Deepseekmath: Pushing the limits of mathematical reasoning in open language models](https://md.research.elamur.ai/free/en/papers/shao-2024-deepseekmath-pushing-limits-mathematical-reasoning/). arXiv preprint arXiv:2402.03300.
18. Huatong Song, Jinhao Jiang, Yingqian Min, Jie Chen, Zhipeng Chen, Wayne Xin Zhao, Lei Fang, and Ji-Rong Wen. 2025. [R1-searcher: Incentivizing the search capability in llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/song-2025-r1-searcher-incentivizing-search/). arXiv preprint arXiv:2503.05592.
19. Nisan Stiennon, Long Ouyang, Jeff Wu, Daniel M Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, and Paul F Christiano. 2020. Learning to summarize from human feedback. Advances in Neural Information Processing Systems, 33:3008– 3021. [find](https://scholar.google.com/scholar?q=Nisan+Stiennon%2C+Long+Ouyang%2C+Jeff+Wu%2C+Daniel+M+Ziegler%2C+Ryan+Lowe%2C+Chelsea+Voss%2C+Alec+Radford%2C+Dario+Amodei%2C+and+Paul+F+Christiano%2E+2020%2E+Learning+to+summarize+from+human+feedback%2E+Advances+in+Neural+)
20. Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot, and Ashish Sabharwal. 2023. [Interleaving retrieval with chain-of-thought reasoning for knowledge-intensive multi-step questions](https://md.research.elamur.ai/free/en/papers/trivedi-2022-interleaving-retrieval-chain-thought-reasoning/). In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), p…
21. Hongru Wang, Cheng Qian, Wanjun Zhong, Xiusi Chen, Jiahao Qiu, Shijue Huang, Bowen Jin, Mengdi Wang, Kam-Fai Wong, and Heng Ji. 2025a. Otc: Optimal tool calls via reinforcement learning. arXiv e-prints, pages arXiv–2504. [find](https://scholar.google.com/scholar?q=Hongru+Wang%2C+Cheng+Qian%2C+Wanjun+Zhong%2C+Xiusi+Chen%2C+Jiahao+Qiu%2C+Shijue+Huang%2C+Bowen+Jin%2C+Mengdi+Wang%2C+Kam%2DFai+Wong%2C+and+Heng+Ji%2E+2025a%2E+Otc%3A+Optimal+tool+calls+via+reinforcement+learning%2E+arXiv+e%2Dprint)
22. Zihan Wang, Kangrui Wang, Qineng Wang, Pingyue Zhang, Linjie Li, Zhengyuan Yang, Xing Jin, Kefan Yu, Minh Nhat Nguyen, Licheng Liu, and 1 others. 2025b. Ragen: Understanding self-evolution in llm agents via multi-turn reinforcement learning. arXiv preprint arXiv:2504.20073. [find](https://scholar.google.com/scholar?q=Zihan+Wang%2C+Kangrui+Wang%2C+Qineng+Wang%2C+Pingyue+Zhang%2C+Linjie+Li%2C+Zhengyuan+Yang%2C+Xing+Jin%2C+Kefan+Yu%2C+Minh+Nhat+Nguyen%2C+Licheng+Liu%2C+and+1+others%2E+2025b%2E+Ragen%3A+Understanding+self%2Devolution+in+llm+agen)
23. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. [Chain-of-thought prompting elicits reasoning in large language models](https://md.research.elamur.ai/free/en/papers/wei-2022-chain-thought-prompting-elicits-reasoning/). Advances in neural information processing systems, 35:24824– 24837.
24. Bosi Wen, Pei Ke, Xiaotao Gu, Lindong Wu, Hao Huang, Jinfeng Zhou, Wenchuang Li, Binxin Hu, Wendy Gao, Jiaxin Xu, and 1 others. 2024. Benchmarking complex instruction-following with multiple constraints composition. Advances in Neural Information Processing Systems, 37:137610–137645. [find](https://scholar.google.com/scholar?q=Bosi+Wen%2C+Pei+Ke%2C+Xiaotao+Gu%2C+Lindong+Wu%2C+Hao+Huang%2C+Jinfeng+Zhou%2C+Wenchuang+Li%2C+Binxin+Hu%2C+Wendy+Gao%2C+Jiaxin+Xu%2C+and+1+others%2E+2024%2E+Benchmarking+complex+instruction%2Dfollowing+with+multiple+constrain)
25. Hongshen Xu, Zichen Zhu, Lei Pan, Zihan Wang, Su Zhu, Da Ma, Ruisheng Cao, Lu Chen, and Kai Yu. 2024. Reducing tool hallucination via reliability alignment. arXiv preprint arXiv:2412.04141. — [arXiv:2412.04141](https://arxiv.org/abs/2412.04141) [find](https://scholar.google.com/scholar?q=Hongshen+Xu%2C+Zichen+Zhu%2C+Lei+Pan%2C+Zihan+Wang%2C+Su+Zhu%2C+Da+Ma%2C+Ruisheng+Cao%2C+Lu+Chen%2C+and+Kai+Yu%2E+2024%2E+Reducing+tool+hallucination+via+reliability+alignment%2E+arXiv+preprint+arXiv%3A2412%2E04141%2E+%E2%80%94+%5BarXiv%3A2)
26. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025. [Qwen3 technical report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). arXiv preprint arXiv:2505.09388.
27. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. 2023. [React: Synergizing reasoning and acting in language models](https://md.research.elamur.ai/free/en/papers/yao-2022-react-synergizing-reasoning-acting-language/). In International Conference on Learning Representations (ICLR).
28. Yue Zhang, Yafu Li, Leyang Cui, Deng Cai, Lemao Liu, Tingchen Fu, Xinting Huang, Enbo Zhao, Yu Zhang, Yulong Chen, and 1 others. 2025. [Siren’s song in the ai ocean: A survey on hallucination in large language models](https://md.research.elamur.ai/free/en/papers/zhang-2023-siren-s-song-ai-ocean/). Computational Linguistics, pages 1–46.
29. Yuxiang Zhang, Jing Chen, Junjie Wang, Yaxin Liu, Cheng Yang, Chufan Shi, Xinyu Zhu, Zihao Lin, Hanwen Wan, Yujiu Yang, and 1 others. 2024a. Toolbehonest: A multi-level hallucination diagnostic benchmark for tool-augmented large language models. In Proceedings of the 2024 Conference on Empirical Met… [find](https://scholar.google.com/scholar?q=Yuxiang+Zhang%2C+Jing+Chen%2C+Junjie+Wang%2C+Yaxin+Liu%2C+Cheng+Yang%2C+Chufan+Shi%2C+Xinyu+Zhu%2C+Zihao+Lin%2C+Hanwen+Wan%2C+Yujiu+Yang%2C+and+1+others%2E+2024a%2E+Toolbehonest%3A+A+multi%2Dlevel+hallucination+diagnostic+benchm)
30. Zhexin Zhang, Shiyao Cui, Yida Lu, Jingzhuo Zhou, Junxiao Yang, Hongning Wang, and Minlie Huang. 2024b. Agent-safetybench: Evaluating the safety of llm agents. arXiv preprint arXiv:2412.14470. — [arXiv:2412.14470](https://arxiv.org/abs/2412.14470) [find](https://scholar.google.com/scholar?q=Zhexin+Zhang%2C+Shiyao+Cui%2C+Yida+Lu%2C+Jingzhuo+Zhou%2C+Junxiao+Yang%2C+Hongning+Wang%2C+and+Minlie+Huang%2E+2024b%2E+Agent%2Dsafetybench%3A+Evaluating+the+safety+of+llm+agents%2E+arXiv+preprint+arXiv%3A2412%2E14470%2E+%E2%80%94+%5BarXi)
31. Yuxiang Zheng, Dayuan Fu, Xiangkun Hu, Xiaojie Cai, Lyumanshan Ye, Pengrui Lu, and Pengfei Liu. 2025. [Deepresearcher: Scaling deep research via reinforcement learning in real-world environments](https://md.research.elamur.ai/free/en/papers/zheng-2025-deepresearcher-scaling-deep-research-reinforcement/). arXiv preprint arXiv:2504.03160.
32. Jeffrey Zhou, Tianjian Lu, Swaroop Mishra, Siddhartha Brahma, Sujoy Basu, Yi Luan, Denny Zhou, and Le Hou. 2023. [Instruction-following evaluation for large language models](https://md.research.elamur.ai/free/en/papers/zhou-2023-instruction-following-evaluation-large-language/). arXiv preprint arXiv:2311.07911.

*Updated 31 October 2026*