---
title: "Can LLMs Really Judge? A Progressive Argumentation-Mining"
description: "The authors propose a progressive argumentation-mining diagnostic framework. 39 references."
authors: ["Fuyu Wang", "Jiangtong Li", "Kun Zhu", "Changjun Jiang"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.1473/"
doi: "10.18653/v1/2026.findings-acl.1473"
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/wang-2026-can-llms-really-judge-progressive/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wang-2026-can-llms-really-judge-progressive/)

[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/24/) / Can LLMs Really Judge? A Progressive Argumentation-Mining Framework for Distinguishing Understanding from Aggregation

# Can LLMs Really Judge? A Progressive Argumentation-Mining Framework for Distinguishing Understanding from Aggregation

**Authors:** Fuyu Wang, [Jiangtong Li](https://md.research.elamur.ai/free/en/authors/jiangtong-li/), Kun Zhu and Changjun Jiang · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/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.findings-acl.1473/) · [PDF](https://aclanthology.org/2026.findings-acl.1473.pdf) · [DOI 10.18653/v1/2026.findings-acl.1473](https://doi.org/10.18653/v1/2026.findings-acl.1473) · [OpenAlex](https://openalex.org/W7166816309)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
The paper argues that benchmark accuracy measures how well LLMs generate answers, not whether they can tell a sound solution from a convincing wrong one. Pilot experiments show two problems. In a best-of-n setup, a model choosing among its own candidates mostly lands on the majority answer. In standard multi-agent debate (MAD), adding agents inflates the context and eventually lowers accuracy. Both effects hide a model's own judgment.

The authors propose a progressive argumentation-mining diagnostic framework. It groups debate responses by final answer and keeps only the single best-supported rationale for each answer, as chosen by a stronger judge model. This keeps the context compact and makes argument quality, not volume, the variable under test.

They apply it to reasoning-centric tasks (e.g., math) and knowledge-centric tasks (e.g., professional QA) and find a sharp contrast:
- **Knowledge tasks:** models are easily swayed by plausible misinformation, even with good evidence.
- **Reasoning tasks:** models can improve when shown strong arguments, but this ability weakens as convincing wrong paths accumulate.

The work suggests that judgment stability should be evaluated alongside generation accuracy.

## Key points
- **Best-of-n study (n = 7):** five Qwen models (qwen-2.5-7b/14b/32b-instruct, qwen3-4b, qwen3-8b) were tested. Agreement between self-selection and majority voting ranged from 85.3% to 98.5%. This indicates that self-selection behaves like aggregation rather than independent judgment.
- **MAD scaling (two rounds, 3/5/7 agents):** results were non-monotonic.
  - qwen3-4b went 69.7 → 71.9 → 72.2 → 69.7.
  - qwen3-8b went 78.2 → 78.9 → 75.7 → 73.8.
- **Candidate queue design:** a stronger judge model maintains one retained rationale per distinct answer. In each round, it either discards new candidates or uses them to replace the retained one.
- **Human validation on MMLU-Professional Medicine:** human annotators repeated the five-round candidate updates on half of the instances. Human–model selection overlap was 0.90–0.93 per round, and accuracy trends under human-curated and LLM-curated queues were closely aligned.
- **Stated limitations:**
  - extra compute cost from debate and judge models;
  - no explanation of the knowledge/reasoning gap at the representation or training level;
  - a small set of datasets and configurations.

*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

- [Tongji University](https://md.research.elamur.ai/free/en/universities/tongji-university/) — University, Shanghai, China · 19 works · 2,162 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 (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works

## Cite

Cite as: Fuyu Wang, Jiangtong Li, Kun Zhu and Changjun Jiang. 2026. Can LLMs Really Judge? A Progressive Argumentation-Mining Framework for Distinguishing Understanding from Aggregation. Findings of the Association for Computational Linguistics: ACL 2026.

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

```bibtex
@inproceedings{wang2026can,
  title = {{Can LLMs Really Judge? A Progressive Argumentation-Mining Framework for Distinguishing Understanding from Aggregation}},
  author = {Fuyu Wang, Jiangtong Li, Kun Zhu and Changjun Jiang},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1473},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1473}
}
```

- [← previous Can LLMs Faithfully Explain Themselves in Low-Resource Languages? A Case Study on Emotion Detection in Persian](https://md.research.elamur.ai/free/en/papers/mehrazar-2026-can-llms-faithfully-explain-themselves/)
- [next → Can LLMs Reason Like Doctors? Exploring the Limits of Large Language Models in Complex Medical Reasoning](https://md.research.elamur.ai/free/en/papers/merenda-2026-can-llms-reason-like-doctors/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/24/)

## References

39 references, 18 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Jose Blanchet, Peng Cui, Jiajin Li, and Jiashuo Liu. 2024. Stability evaluation via distributional perturbation analysis. In ICLM 2024. [find](https://scholar.google.com/scholar?q=Jose+Blanchet%2C+Peng+Cui%2C+Jiajin+Li%2C+and+Jiashuo+Liu%2E+2024%2E+Stability+evaluation+via+distributional+perturbation+analysis%2E+In+ICLM+2024%2E)
2. Yan Cai, Linlin Wang, Ye Wang, Gerard de Melo, Ya Zhang, Yanfeng Wang, and Liang He. 2024. Medbench: A large-scale chinese benchmark for evaluating medical large language models. In AAAI 2024, volume 38, pages 17709–17717. [find](https://scholar.google.com/scholar?q=Yan+Cai%2C+Linlin+Wang%2C+Ye+Wang%2C+Gerard+de+Melo%2C+Ya+Zhang%2C+Yanfeng+Wang%2C+and+Liang+He%2E+2024%2E+Medbench%3A+A+large%2Dscale+chinese+benchmark+for+evaluating+medical+large+language+models%2E+In+AAAI+2024%2C+volume+)
3. Chi-Min Chan, Weize Chen, Yusheng Su, Jianxuan Yu, Wei Xue, Shanghang Zhang, Jie Fu, and Zhiyuan Liu. 2024. [Chateval: Towards better llm-based evaluators through multi-agent debate](https://md.research.elamur.ai/free/en/papers/chan-2023-chateval-better-llm-based/). In ICLR 2024.
4. Cléa Chataigner, Rebecca Ma, Prakhar Ganesh, Yuhao Chen, Afaf Taïk, Elliot Creager, and Golnoosh Farnadi. 2025. Say it another way: Auditing llms with a user-grounded automated paraphrasing framework. In NeurIPS 2025. [find](https://scholar.google.com/scholar?q=Cl%C3%A9a+Chataigner%2C+Rebecca+Ma%2C+Prakhar+Ganesh%2C+Yuhao+Chen%2C+Afaf+Ta%C3%AFk%2C+Elliot+Creager%2C+and+Golnoosh+Farnadi%2E+2025%2E+Say+it+another+way%3A+Auditing+llms+with+a+user%2Dgrounded+automated+paraphrasing+framework%2E)
5. Elena Chistova, Philipp Cimiano, Shima Haddadan, Gabriella Lapesa, and Ramon Ruiz-Dolz, editors. 2025. Proceedings of the 12th Workshop on Argument Mining. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Elena+Chistova%2C+Philipp+Cimiano%2C+Shima+Haddadan%2C+Gabriella+Lapesa%2C+and+Ramon+Ruiz%2DDolz%2C+editors%2E+2025%2E+Proceedings+of+the+12th+Workshop+on+Argument+Mining%2E+Association+for+Computational+Linguistics%2E)
6. Hyeong Kyu Choi, Xiaojin Zhu, and Sharon Li. 2025. Debate or vote: Which yields better decisions in multi-agent large language models? In NeurIPS 2025. [find](https://scholar.google.com/scholar?q=Hyeong+Kyu+Choi%2C+Xiaojin+Zhu%2C+and+Sharon+Li%2E+2025%2E+Debate+or+vote%3A+Which+yields+better+decisions+in+multi%2Dagent+large+language+models%3F+In+NeurIPS+2025%2E)
7. Heejin Do, Jaehui Hwang, Dongyoon Han, Seong Joon Oh, and Sangdoo Yun. 2025. What defines good reasoning in llms? dissecting reasoning steps with multi-aspect evaluation. arXiv preprint arXiv:2510.20603. — [arXiv:2510.20603](https://arxiv.org/abs/2510.20603) [find](https://scholar.google.com/scholar?q=Heejin+Do%2C+Jaehui+Hwang%2C+Dongyoon+Han%2C+Seong+Joon+Oh%2C+and+Sangdoo+Yun%2E+2025%2E+What+defines+good+reasoning+in+llms%3F+dissecting+reasoning+steps+with+multi%2Daspect+evaluation%2E+arXiv+preprint+arXiv%3A2510%2E206)
8. Yilun Du, Shuang Li, Antonio Torralba, Joshua B. Tenenbaum, and Igor Mordatch. 2024. [Improving factuality and reasoning in language models through multiagent debate](https://md.research.elamur.ai/free/en/papers/du-2023-improving-factuality-reasoning-language-models/). In ICML 2024.
9. Andrew Estornell and Yang Liu. 2024. Multi-llm debate: Framework, principals, and interventions. In NeurIPS 2024. [find](https://scholar.google.com/scholar?q=Andrew+Estornell+and+Yang+Liu%2E+2024%2E+Multi%2Dllm+debate%3A+Framework%2C+principals%2C+and+interventions%2E+In+NeurIPS+2024%2E)
10. Zhiwei Fei, Xiaoyu Shen, Dawei Zhu, Fengzhe Zhou, Zhuo Han, Alan Huang, Songyang Zhang, Kai Chen, Zhixin Yin, Zongwen Shen, et al. 2024. [Lawbench: Benchmarking legal knowledge of large language models](https://md.research.elamur.ai/free/en/papers/fei-2023-lawbench-benchmarking-legal-knowledge-large/). In EMNLP 2024, pages 7933–7962.
11. Team GLM, Aohan Zeng, Bin Xu, Bowen Wang, Chenhui Zhang, Da Yin, Dan Zhang, Diego Rojas, Guanyu Feng, Hanlin Zhao, et al. 2024. [Chatglm: A family of large language models from glm-130b to glm-4 all tools](https://md.research.elamur.ai/free/en/papers/glm-2024-chatglm-family-large-language-models/). arXiv preprint arXiv:2406.12793.
12. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2021. [Measuring massive multitask language understanding](https://md.research.elamur.ai/free/en/papers/hendrycks-2020-measuring-massive-multitask-language-understanding/).
13. Yuzhen Huang, Yuzhuo Bai, Zhihao Zhu, Junlei Zhang, Jinghan Zhang, Tangjun Su, Junteng Liu, Chuancheng Lv, Yikai Zhang, Yao Fu, et al. 2023. [C-eval: A multi-level multi-discipline chinese evaluation suite for foundation models](https://md.research.elamur.ai/free/en/papers/huang-2023-c-eval-multi-level-multi/). Advances in Neural Information Processing Systems, 36:62991–63010.
14. Dongwei Jiang, Jingyu Zhang, Orion Weller, Nathaniel Weir, Benjamin Van Durme, and Daniel Khashabi. 2025. Self-[in] correct: Llms struggle with discriminating self-generated responses. In AAAI 2025, volume 39, pages 24266–24275. [find](https://scholar.google.com/scholar?q=Dongwei+Jiang%2C+Jingyu+Zhang%2C+Orion+Weller%2C+Nathaniel+Weir%2C+Benjamin+Van+Durme%2C+and+Daniel+Khashabi%2E+2025%2E+Self%2D%5Bin%5D+correct%3A+Llms+struggle+with+discriminating+self%2Dgenerated+responses%2E+In+AAAI+2025%2C+v)
15. Lars Benedikt Kaesberg, Jonas Becker, Jan Philip Wahle, Terry Ruas, and Bela Gipp. 2025. Voting or consensus? decision-making in multi-agent debate. In ACL 2025, pages 11640–11671. [find](https://scholar.google.com/scholar?q=Lars+Benedikt+Kaesberg%2C+Jonas+Becker%2C+Jan+Philip+Wahle%2C+Terry+Ruas%2C+and+Bela+Gipp%2E+2025%2E+Voting+or+consensus%3F+decision%2Dmaking+in+multi%2Dagent+debate%2E+In+ACL+2025%2C+pages+11640%E2%80%9311671%2E)
16. Mehran Kazemi, Bahare Fatemi, Hritik Bansal, John Palowitch, Chrysovalantis Anastasiou, Sanket Vaibhav Mehta, Lalit K Jain, Virginia Aglietti, Disha Jindal, Yuanzhu Peter Chen, et al. 2025. [Big-bench extra hard](https://md.research.elamur.ai/free/en/papers/kazemi-2025-big-bench-extra-hard/). In ACL 2025, pages 26473–26501.
17. Akbir Khan, John Hughes, Dan Valentine, Laura Ruis, Kshitij Sachan, Ansh Radhakrishnan, Edward Grefenstette, Samuel R. Bowman, Tim Rocktäschel, and Ethan Perez. 2024. Debating with more persuasive llms leads to more truthful answers. In ICML 2024. [find](https://scholar.google.com/scholar?q=Akbir+Khan%2C+John+Hughes%2C+Dan+Valentine%2C+Laura+Ruis%2C+Kshitij+Sachan%2C+Ansh+Radhakrishnan%2C+Edward+Grefenstette%2C+Samuel+R%2E+Bowman%2C+Tim+Rockt%C3%A4schel%2C+and+Ethan+Perez%2E+2024%2E+Debating+with+more+persuasive+llm)
18. Jiangtong Li, Liu Liu, Li Niu, and Liqing Zhang. 2021. Memorize, associate and match: Embedding enhancement via fine-grained alignment for image-text retrieval. IEEE Transactions on Image Processing, pages 9193–9207. [find](https://scholar.google.com/scholar?q=Jiangtong+Li%2C+Liu+Liu%2C+Li+Niu%2C+and+Liqing+Zhang%2E+2021%2E+Memorize%2C+associate+and+match%3A+Embedding+enhancement+via+fine%2Dgrained+alignment+for+image%2Dtext+retrieval%2E+IEEE+Transactions+on+Image+Processing%2C+)
19. Tian Liang, Zhiwei He, Wenxiang Jiao, Xing Wang, Yan Wang, Rui Wang, Yujiu Yang, Shuming Shi, and Zhaopeng Tu. 2024. [Encouraging divergent thinking in large language models through multi-agent debate](https://md.research.elamur.ai/free/en/papers/liang-2023-encouraging-divergent-thinking-large-language/). In EMNLP 2024, pages 17889–17904.
20. Hunter Lightman, Vineet Kosaraju, Yura Burda, Harri Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe. 2024. [Let’s verify step by step](https://md.research.elamur.ai/free/en/papers/lightman-2023-let-s-verify-step-step/). In ICLR 2024.
21. Chin-Yew Lin. 2004. [ROUGE: A package for automatic evaluation of summaries](https://md.research.elamur.ai/free/en/papers/lin-2004-rouge-package-automatic-evaluation-summaries/). In Text Summarization Branches Out, pages 74–81.
22. Zicheng Lin, Zhibin Gou, Tian Liang, Ruilin Luo, Haowei Liu, and Yujiu Yang. 2024. Criticbench: Benchmarking llms for critique-correct reasoning. In ACL 2024, pages 1552–1587. [find](https://scholar.google.com/scholar?q=Zicheng+Lin%2C+Zhibin+Gou%2C+Tian+Liang%2C+Ruilin+Luo%2C+Haowei+Liu%2C+and+Yujiu+Yang%2E+2024%2E+Criticbench%3A+Benchmarking+llms+for+critique%2Dcorrect+reasoning%2E+In+ACL+2024%2C+pages+1552%E2%80%931587%2E)
23. Xinxin Liu, Aaron Thomas, Cheng Zhang, Jianyi Cheng, Yiren Zhao, and Xitong Gao. 2025a. Refining salience-aware sparse fine-tuning strategies for language models. In ACL 2025, pages 31932–31945. [find](https://scholar.google.com/scholar?q=Xinxin+Liu%2C+Aaron+Thomas%2C+Cheng+Zhang%2C+Jianyi+Cheng%2C+Yiren+Zhao%2C+and+Xitong+Gao%2E+2025a%2E+Refining+salience%2Daware+sparse+fine%2Dtuning+strategies+for+language+models%2E+In+ACL+2025%2C+pages+31932%E2%80%9331945%2E)
24. Yexiang Liu, Jie Cao, Zekun Li, Ran He, and Tieniu Tan. 2025b. Breaking mental set to improve reasoning through diverse multi-agent debate. In ICLR 2025. [find](https://scholar.google.com/scholar?q=Yexiang+Liu%2C+Jie+Cao%2C+Zekun+Li%2C+Ran+He%2C+and+Tieniu+Tan%2E+2025b%2E+Breaking+mental+set+to+improve+reasoning+through+diverse+multi%2Dagent+debate%2E+In+ICLR+2025%2E)
25. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. [Bleu: a method for automatic evaluation of machine translation](https://md.research.elamur.ai/free/en/papers/papineni-2002-bleu-method-automatic-evaluation-machine/). In ACL 2002, pages 311–318.
26. Kai Xiong, Xiao Ding, Yixin Cao, Ting Liu, and Bing Qin. 2023. [Examining inter-consistency of large language models collaboration: An in-depth analysis via debate](https://md.research.elamur.ai/free/en/papers/xiong-2023-examining-inter-consistency-large-language/). In EMNLP 2023, pages 7572–7590.
27. Paloma Piot, David Otero, Patricia Martín-Rodilla, and Javier Parapar. 2025. [Can llms evaluate what they cannot annotate? revisiting llm reliability in hate speech detection](https://md.research.elamur.ai/free/en/papers/piot-2026-can-llms-evaluate-what-they/). arXiv preprint arXiv:2512.09662.
28. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, et al. 2025. [Qwen3 technical report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). arXiv preprint arXiv:2505.09388.
29. Vyas Raina, Adian Liusie, and Mark Gales. 2024. [Is LLM-as-a-judge robust? investigating universal adversarial attacks on zero-shot LLM assessment](https://md.research.elamur.ai/free/en/papers/raina-2024-is-llm-as-judge-robust/). In EMNLP 2024, pages 7499–7517.
30. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, et al. 2024. Qwen2. 5 technical report. arXiv preprint arXiv:2412.15115. [find](https://scholar.google.com/scholar?q=An+Yang%2C+Baosong+Yang%2C+Beichen+Zhang%2C+Binyuan+Hui%2C+Bo+Zheng%2C+Bowen+Yu%2C+Chengyuan+Li%2C+Dayiheng+Liu%2C+Fei+Huang%2C+Haoran+Wei%2C+et+al%2E+2024%2E+Qwen2%2E+5+technical+report%2E+arXiv+preprint+arXiv%3A2412%2E15115%2E)
31. Eva Sánchez Salido, Julio Gonzalo, and Guillermo Marco. 2025. None of the others: a general technique to distinguish reasoning from memorization in multiple-choice llm evaluation benchmarks. arXiv preprint arXiv:2502.12896. — [arXiv:2502.12896](https://arxiv.org/abs/2502.12896) [find](https://scholar.google.com/scholar?q=Eva+S%C3%A1nchez+Salido%2C+Julio+Gonzalo%2C+and+Guillermo+Marco%2E+2025%2E+None+of+the+others%3A+a+general+technique+to+distinguish+reasoning+from+memorization+in+multiple%2Dchoice+llm+evaluation+benchmarks%2E+arXiv+pre)
32. Andrei Shleifer. 2012. Psychologists at the gate: a review of daniel kahneman’s thinking, fast and slow. Journal of Economic Literature, 50(4):1080–1091. [find](https://scholar.google.com/scholar?q=Andrei+Shleifer%2E+2012%2E+Psychologists+at+the+gate%3A+a+review+of+daniel+kahneman%E2%80%99s+thinking%2C+fast+and+slow%2E+Journal+of+Economic+Literature%2C+50%284%29%3A1080%E2%80%931091%2E)
33. Karan Singhal, Tao Tu, Juraj Gottweis, Rory Sayres, Ellery Wulczyn, Mohamed Amin, Le Hou, Kevin Clark, Stephen R Pfohl, Heather Cole-Lewis, et al. 2025. Toward expert-level medical question answering with large language models. Nature Medicine, 31(3):943–950. [find](https://scholar.google.com/scholar?q=Karan+Singhal%2C+Tao+Tu%2C+Juraj+Gottweis%2C+Rory+Sayres%2C+Ellery+Wulczyn%2C+Mohamed+Amin%2C+Le+Hou%2C+Kevin+Clark%2C+Stephen+R+Pfohl%2C+Heather+Cole%2DLewis%2C+et+al%2E+2025%2E+Toward+expert%2Dlevel+medical+question+answering+)
34. Mirac Suzgun, Nathan Scales, Nathanael Schärli, Sebastian Gehrmann, Yi Tay, Hyung Won Chung, Aakanksha Chowdhery, Quoc Le, Ed Chi, Denny Zhou, et al. 2023. [Challenging big-bench tasks and whether chain-of-thought can solve them](https://md.research.elamur.ai/free/en/papers/suzgun-2023-challenging-big-bench-tasks-whether/). In ACL 2023, pages 13003–13051.
35. Gladys Tyen, Hassan Mansoor, Victor Carbune, Peter Chen, and Tony Mak. 2024. [LLMs cannot find reasoning errors, but can correct them given the error location](https://md.research.elamur.ai/free/en/papers/tyen-2024-llms-cannot-find-reasoning-errors/). In ACL 2024, pages 13894–13908.
36. Fuyu Wang, Jiangtong Li, Kun Zhu, and Changjun Jiang. 2025. InspireDebate: Multi-dimensional subjective-objective evaluation-guided reasoning and optimization for debating. In ACL 2025, pages 27525–27544. [find](https://scholar.google.com/scholar?q=Fuyu+Wang%2C+Jiangtong+Li%2C+Kun+Zhu%2C+and+Changjun+Jiang%2E+2025%2E+InspireDebate%3A+Multi%2Ddimensional+subjective%2Dobjective+evaluation%2Dguided+reasoning+and+optimization+for+debating%2E+In+ACL+2025%2C+pages+27525%E2%80%9327)
37. Xiaoxuan Wang, Ziniu Hu, Pan Lu, Yanqiao Zhu, Jieyu Zhang, Satyen Subramaniam, Arjun R. Loomba, Shichang Zhang, Yizhou Sun, and Wei Wang. 2024. [SciBench: Evaluating College-Level Scientific Problem-Solving Abilities of Large Language Models](https://md.research.elamur.ai/free/en/papers/wang-2023-scibench-evaluating-college-level-scientific/). In ICML 2024.
38. Peter West, Ximing Lu, Nouha Dziri, Faeze Brahman, Linjie Li, Jena D. Hwang, Liwei Jiang, Jillian Fisher, Abhilasha Ravichander, Khyathi Chandu, Benjamin Newman, Pang Wei Koh, Allyson Ettinger, and Yejin Choi. 2024. The generative ai paradox: "what it can create, it may not understand". In ICLR 2024… [find](https://scholar.google.com/scholar?q=Peter+West%2C+Ximing+Lu%2C+Nouha+Dziri%2C+Faeze+Brahman%2C+Linjie+Li%2C+Jena+D%2E+Hwang%2C+Liwei+Jiang%2C+Jillian+Fisher%2C+Abhilasha+Ravichander%2C+Khyathi+Chandu%2C+Benjamin+Newman%2C+Pang+Wei+Koh%2C+Allyson+Ettinger%2C+and+Ye)
39. Zhenyu Wu, Qingkai Zeng, Zhihan Zhang, Zhaoxuan Tan, Chao Shen, and Meng Jiang. 2024. Large language models can self-correct with key condition verification. In EMNLP 2024, pages 12846–12867. [find](https://scholar.google.com/scholar?q=Zhenyu+Wu%2C+Qingkai+Zeng%2C+Zhihan+Zhang%2C+Zhaoxuan+Tan%2C+Chao+Shen%2C+and+Meng+Jiang%2E+2024%2E+Large+language+models+can+self%2Dcorrect+with+key+condition+verification%2E+In+EMNLP+2024%2C+pages+12846%E2%80%9312867%2E)

*Updated 1 November 2026*