---
title: "When and What to Ask — paper by Zhao et al. (2026)"
description: "The authors present AskBench, a benchmark that automatically turns ordinary (query… 0.010 accuracy. 34 references, cited by 1 paper here."
authors: ["Jiale Zhao", "Ke Fang", "Lu Cheng"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.845/"
doi: "10.18653/v1/2026.findings-acl.845"
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/zhao-2026-when-what-ask-askbench-rubric/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/zhao-2026-when-what-ask-askbench-rubric/)

[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/9/) / When and What to Ask: AskBench and Rubric-Guided RLVR for LLM Clarification

# When and What to Ask: AskBench and Rubric-Guided RLVR for LLM Clarification

**Authors:** [Jiale Zhao](https://md.research.elamur.ai/free/en/authors/jiale-zhao/), Ke Fang and [Lu Cheng](https://md.research.elamur.ai/free/en/authors/lu-cheng/) · **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.845/) · [PDF](https://aclanthology.org/2026.findings-acl.845.pdf) · [DOI 10.18653/v1/2026.findings-acl.845](https://doi.org/10.18653/v1/2026.findings-acl.845) · [OpenAlex](https://openalex.org/W7166870529)

**Citations:** 1 citations (OpenAlex, 2026-10-14) · 1 in our library, of the 149,439 papers whose bibliographies we read

## Summary
LLMs tend to answer right away, even when a prompt leaves out decisive details or contains confidently stated errors. This leads to hallucinated assumptions or reinforced user misconceptions. The authors present AskBench, a benchmark that automatically turns ordinary (query, answer) pairs into multi-turn dialogues with itemized rubric checkpoints. It has two settings. In AskMind, key facts are removed or blurred, so the model must ask for them. In AskOverconfidence, misleading claims are inserted, so the model must spot and correct them. A single LLM judge does three jobs: it classifies each reply as a question or a final answer, grades final answers, and plays the user by revealing only what is explicitly requested. The authors also propose rubric-guided RLVR, which uses checkpoint-based rewards to favor correct answers and targeted questions and to penalize answering too early. Trained models improve accuracy and checkpoint coverage, and these gains transfer to unseen benchmarks. The work treats clarification as a measurable, trainable skill, which matters for high-stakes uses such as medicine.

## Key points
- AskBench is built from Math500, MedQA, BBH, and GPQA-d, with 100 items per domain in each setting. This gives 400 AskMind and 400 AskOverconfidence dialogues, for 800 instances in total.
- On QuestBench-Math, the AskMind-trained policy raises accuracy from 0.320 to 0.539 and coverage from 0.379 to 0.835, though unnecessary questions also increase (0.097).
- On IN3, the trained models ask on vague tasks far more often (0.941 and 1.000). However, they answer clear tasks directly less often (0.625 and 0.500, versus 1.000 for Qwen), which shows a caution-versus-efficiency trade-off.
- The Hard protocol allows one clarification turn before a forced answer. Under it, Gemini answered immediately in 387 of 400 AskOverconfidence cases and scored 0.010 accuracy despite 0.735 coverage. The RLVR policies reached roughly 0.20–0.27 accuracy.
- The authors list several limitations:
  - The user is simulated by an LLM rather than a real person.
  - All rewards and evaluations come from a single judge model.
  - The domains are concentrated in mathematics and medicine.

*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

- [University of Illinois Chicago](https://md.research.elamur.ai/free/en/universities/university-of-illinois-chicago/) — University, Chicago, United States · 61 works · 20,107 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: Jiale Zhao, Ke Fang and Lu Cheng. 2026. When and What to Ask: AskBench and Rubric-Guided RLVR for LLM Clarification. 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{zhao2026when,
  title = {{When and What to Ask: AskBench and Rubric-Guided RLVR for LLM Clarification}},
  author = {Jiale Zhao, Ke Fang and Lu Cheng},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.845},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.845}
}
```

- [← previous What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time](https://md.research.elamur.ai/free/en/papers/yan-2026-what-if-consensus-lies-selective/)
- [next → When Flores Bloomz Wrong: Cross-Direction Contamination in Machine Translation Evaluation](https://md.research.elamur.ai/free/en/papers/tan-2026-when-flores-bloomz-wrong-cross/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/9/)

## References

34 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Mohammad Aliannejadi, Hamed Zamani, Fabio Crestani, and W Bruce Croft. 2019. Asking clarifying questions in open-domain information-seeking conversations. In Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 475–484. [find](https://scholar.google.com/scholar?q=Mohammad+Aliannejadi%2C+Hamed+Zamani%2C+Fabio+Crestani%2C+and+W+Bruce+Croft%2E+2019%2E+Asking+clarifying+questions+in+open%2Ddomain+information%2Dseeking+conversations%2E+In+Proceedings+of+the+42nd+International+ACM+)
2. Rahul K Arora, Jason Wei, Rebecca Soskin Hicks, Preston Bowman, Joaquin Quiñonero-Candela, Foivos Tsimpourlas, Michael Sharman, Meghan Shah, Andrea Vallone, Alex Beutel, and 1 others. 2025. [Healthbench: Evaluating large language models towards improved human health](https://md.research.elamur.ai/free/en/papers/arora-2025-healthbench-evaluating-large-language-models/). arXiv preprint arXiv:2505.08775.
3. Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, and 1 others. 2022. [Constitutional AI: Harmlessness from AI feedback](https://md.research.elamur.ai/free/en/papers/bai-2022-constitutional-ai-harmlessness-ai-feedback/). arXiv preprint arXiv:2212.08073.
4. Chuanruo Fu and Yuncheng Du. 2025. First ask then answer: A framework design for ai dialogue based on supplementary questioning with large language models. arXiv preprint arXiv:2508.08308. — [arXiv:2508.08308](https://arxiv.org/abs/2508.08308) [find](https://scholar.google.com/scholar?q=Chuanruo+Fu+and+Yuncheng+Du%2E+2025%2E+First+ask+then+answer%3A+A+framework+design+for+ai+dialogue+based+on+supplementary+questioning+with+large+language+models%2E+arXiv+preprint+arXiv%3A2508%2E08308%2E+%E2%80%94+%5BarXiv%3A25)
5. Yifan Gao, Jingjing Li, Chien-Sheng Wu, Michael R Lyu, and Irwin King. 2021. Open-retrieval conversational machine reading. arXiv preprint arXiv:2102.08633. — [arXiv:2102.08633](https://arxiv.org/abs/2102.08633) [find](https://scholar.google.com/scholar?q=Yifan+Gao%2C+Jingjing+Li%2C+Chien%2DSheng+Wu%2C+Michael+R+Lyu%2C+and+Irwin+King%2E+2021%2E+Open%2Dretrieval+conversational+machine+reading%2E+arXiv+preprint+arXiv%3A2102%2E08633%2E+%E2%80%94+%5BarXiv%3A2102%2E08633%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F)
6. Ruohao Guo, Wei Xu, and Alan Ritter. 2025. How to protect yourself from 5g radiation? investigating llm responses to implicit misinformation. arXiv preprint arXiv:2503.09598. [find](https://scholar.google.com/scholar?q=Ruohao+Guo%2C+Wei+Xu%2C+and+Alan+Ritter%2E+2025%2E+How+to+protect+yourself+from+5g+radiation%3F+investigating+llm+responses+to+implicit+misinformation%2E+arXiv+preprint+arXiv%3A2503%2E09598%2E)
7. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. 2021. [Measuring mathematical problem solving with the math dataset](https://md.research.elamur.ai/free/en/papers/hendrycks-2021-measuring-mathematical-problem-solving-math/). arXiv preprint arXiv:2103.03874.
8. Pengcheng Jiang, Cao Xiao, Minhao Jiang, Parminder Bhatia, Taha Kass-Hout, Jimeng Sun, and Jiawei Han. 2024. Reasoning-enhanced healthcare predictions with knowledge graph community retrieval. arXiv preprint arXiv:2410.04585. — [arXiv:2410.04585](https://arxiv.org/abs/2410.04585) [find](https://scholar.google.com/scholar?q=Pengcheng+Jiang%2C+Cao+Xiao%2C+Minhao+Jiang%2C+Parminder+Bhatia%2C+Taha+Kass%2DHout%2C+Jimeng+Sun%2C+and+Jiawei+Han%2E+2024%2E+Reasoning%2Denhanced+healthcare+predictions+with+knowledge+graph+community+retrieval%2E+arXiv+p)
9. Di Jin, Eileen Pan, Nassim Oufattole, Wei-Hung Weng, Hanyi Fang, and Peter Szolovits. 2021. [What disease does this patient have? a large-scale open domain question answering dataset from medical exams](https://md.research.elamur.ai/free/en/papers/jin-2020-disease-this-patient-have-large/). Applied Sciences, 11(14):6421.
10. Hyoungwook Jin, Seonghee Lee, Hyungyu Shin, and Juho Kim. 2024. Teach ai how to code: Using large language models as teachable agents for programming education. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems, pages 1–28. [find](https://scholar.google.com/scholar?q=Hyoungwook+Jin%2C+Seonghee+Lee%2C+Hyungyu+Shin%2C+and+Juho+Kim%2E+2024%2E+Teach+ai+how+to+code%3A+Using+large+language+models+as+teachable+agents+for+programming+education%2E+In+Proceedings+of+the+2024+CHI+Conferen)
11. Saurav Kadavath, Tom Conerly, Amanda Askell, Tom Henighan, Dawn Drain, Ethan Perez, Nicholas Schiefer, Zac Hatfield-Dodds, Nova DaSilva, Eli Elhage, and 1 others. 2022. Language models (mostly) know what they know. arXiv preprint arXiv:2207.05221. [find](https://scholar.google.com/scholar?q=Saurav+Kadavath%2C+Tom+Conerly%2C+Amanda+Askell%2C+Tom+Henighan%2C+Dawn+Drain%2C+Ethan+Perez%2C+Nicholas+Schiefer%2C+Zac+Hatfield%2DDodds%2C+Nova+DaSilva%2C+Eli+Elhage%2C+and+1+others%2E+2022%2E+Language+models+%28mostly%29+know+w)
12. Vaibhav Kumar and Alan W Black. 2020. ClarQ: A large-scale and diverse dataset for clarification question generation. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7296–7301. [find](https://scholar.google.com/scholar?q=Vaibhav+Kumar+and+Alan+W+Black%2E+2020%2E+ClarQ%3A+A+large%2Dscale+and+diverse+dataset+for+clarification+question+generation%2E+In+Proceedings+of+the+58th+Annual+Meeting+of+the+Association+for+Computational+Lin)
13. Belinda Z. Li, Been Kim, and Zi Wang. 2025a. Questbench: Can llms ask the right question to acquire information in reasoning tasks? arXiv preprint arXiv:2503.22674. — [arXiv:2503.22674](https://arxiv.org/abs/2503.22674) [find](https://scholar.google.com/scholar?q=Belinda+Z%2E+Li%2C+Been+Kim%2C+and+Zi+Wang%2E+2025a%2E+Questbench%3A+Can+llms+ask+the+right+question+to+acquire+information+in+reasoning+tasks%3F+arXiv+preprint+arXiv%3A2503%2E22674%2E+%E2%80%94+%5BarXiv%3A2503%2E22674%5D%28https%3A%2F%2Farxiv%2E)
14. Dawei Li, Bohan Jiang, Liangjie Huang, Alimohammad Beigi, Chengshuai Zhao, Zhen Tan, Amrita Bhattacharjee, Yuxuan Jiang, Canyu Chen, Tianhao Wu, and 1 others. 2025b. From generation to judgment: Opportunities and challenges of llm-asa-judge. In Proceedings of the 2025 Conference on Empirical Methods… [find](https://scholar.google.com/scholar?q=Dawei+Li%2C+Bohan+Jiang%2C+Liangjie+Huang%2C+Alimohammad+Beigi%2C+Chengshuai+Zhao%2C+Zhen+Tan%2C+Amrita+Bhattacharjee%2C+Yuxuan+Jiang%2C+Canyu+Chen%2C+Tianhao+Wu%2C+and+1+others%2E+2025b%2E+From+generation+to+judgment%3A+Oppor)
15. Sunzhu Li, Jiale Zhao, Miteto Wei, Huimin Ren, Yang Zhou, Jingwen Yang, Shunyu Liu, Kaike Zhang, and Wei Chen. 2026. [Rubrichub: A comprehensive and highly discriminative rubric dataset via automated coarse-to-fine generation](https://md.research.elamur.ai/free/en/papers/li-2026-rubrichub-comprehensive-highly-discriminative-rubric/). Preprint, arXiv:2601.08430.
16. Yang Liu, Dan Iter, Yichong Xu, Shuohang Wang, Ruochen Xu, and Chenguang Zhu. 2023. [G-Eval: NLG evaluation using GPT-4 with better human alignment](https://md.research.elamur.ai/free/en/papers/liu-2023-g-eval-nlg-evaluation-gpt/). arXiv preprint arXiv:2303.16634.
17. Dehai Min, Kailin Zhang, Tongtong Wu, and Lu Cheng. 2025. [Quco-rag: Quantifying uncertainty from the pre-training corpus for dynamic retrieval-augmented generation](https://md.research.elamur.ai/free/en/papers/min-2026-quco-rag-quantifying-uncertainty-pre/). arXiv preprint arXiv:2512.19134.
18. Hussein Mozannar, Valerie Chen, Mohammed Alsobay, Subhro Das, Sebastian Zhao, Dennis Wei, Manish Nagireddy, Prasanna Sattigeri, Ameet Talwalkar, and David Sontag. 2024. The realhumaneval: Evaluating large language models’ abilities to support programmers. arXiv preprint arXiv:2404.02806. — [arXiv:2404.02806](https://arxiv.org/abs/2404.02806) [find](https://scholar.google.com/scholar?q=Hussein+Mozannar%2C+Valerie+Chen%2C+Mohammed+Alsobay%2C+Subhro+Das%2C+Sebastian+Zhao%2C+Dennis+Wei%2C+Manish+Nagireddy%2C+Prasanna+Sattigeri%2C+Ameet+Talwalkar%2C+and+David+Sontag%2E+2024%2E+The+realhumaneval%3A+Evaluating+l)
19. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, and 1 others. 2022. [Training language models to follow instructions with human feedback](https://md.research.elamur.ai/free/en/papers/ouyang-2022-training-language-models-follow-instructions/). Advances in neural information processing systems, 35:27730– 27744.
20. Ankit Pal, Logesh Kumar Umapathi, and Malaikannan Sankarasubbu. 2022. [Medmcqa: A large-scale multi-subject multi-choice dataset for medical domain question answering](https://md.research.elamur.ai/free/en/papers/pal-2022-medmcqa-large-scale-multi-subject/). In Conference on Health, Inference, and Learning, pages 248–260. PMLR.
21. Cheng Qian, Bingxiang He, Zhong Zhuang, Jia Deng, Yujia Qin, Xin Cong, Zhong Zhang, Jie Zhou, Yankai Lin, Zhiyuan Liu, and Maosong Sun. 2024. Tell me more! towards implicit user intention understanding of language model driven agents. arXiv preprint arXiv:2402.09205. [find](https://scholar.google.com/scholar?q=Cheng+Qian%2C+Bingxiang+He%2C+Zhong+Zhuang%2C+Jia+Deng%2C+Yujia+Qin%2C+Xin+Cong%2C+Zhong+Zhang%2C+Jie+Zhou%2C+Yankai+Lin%2C+Zhiyuan+Liu%2C+and+Maosong+Sun%2E+2024%2E+Tell+me+more%21+towards+implicit+user+intention+understandin)
22. David Rein, Betty Li Hou, Asa Cooper Stickland, Jackson Petty, Richard Yuanzhe Pang, Julien Dirani, Julian Michael, and Samuel R Bowman. 2024. [Gpqa: A graduate-level google-proof q&a benchmark](https://md.research.elamur.ai/free/en/papers/rein-2023-gpqa-graduate-level-google-proof/). In First Conference on Language Modeling.
23. Marco Tulio Ribeiro, Tongshuang Wu, Carlos Guestrin, and Sameer Singh. 2020. Beyond accuracy: Behavioral testing of nlp models with checklist. arXiv preprint arXiv:2005.04118. [find](https://scholar.google.com/scholar?q=Marco+Tulio+Ribeiro%2C+Tongshuang+Wu%2C+Carlos+Guestrin%2C+and+Sameer+Singh%2E+2020%2E+Beyond+accuracy%3A+Behavioral+testing+of+nlp+models+with+checklist%2E+arXiv+preprint+arXiv%3A2005%2E04118%2E)
24. Marzieh Saeidi, Max Bartolo, Patrick Lewis, Sameer Singh, Tim Rocktäschel, Mike Sheldon, Guillaume Bouchard, and Sebastian Riedel. 2018. [Interpretation of natural language rules in conversational machine reading](https://md.research.elamur.ai/free/en/papers/saeidi-2018-interpretation-natural-language-rules-conversational/). arXiv preprint arXiv:1809.01494.
25. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, YK Li, Yang Wu, and 1 others. 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.
26. Jiayuan Su, Jing Luo, Hongwei Wang, and Lu Cheng. 2024. Api is enough: Conformal prediction for large language models without logit-access. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 979–995. [find](https://scholar.google.com/scholar?q=Jiayuan+Su%2C+Jing+Luo%2C+Hongwei+Wang%2C+and+Lu+Cheng%2E+2024%2E+Api+is+enough%3A+Conformal+prediction+for+large+language+models+without+logit%2Daccess%2E+In+Findings+of+the+Association+for+Computational+Linguistics)
27. Mirac Suzgun, Nathan Scales, Nathanael Schärli, Sebastian Gehrmann, Yi Tay, Hyung Won Chung, Aakanksha Chowdhery, Quoc Le, Ed Chi, Denny Zhou, and 1 others. 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 Findings of the Association for Computational Linguistics: A…
28. Xin Xu, Jiaxin Zhang, Tianhao Chen, Zitong Chao, Jishan Hu, and Can Yang. 2025. Ugmathbench: A diverse and dynamic benchmark for undergraduate-level mathematical reasoning with large language models. arXiv preprint arXiv:2501.13766. — [arXiv:2501.13766](https://arxiv.org/abs/2501.13766) [find](https://scholar.google.com/scholar?q=Xin+Xu%2C+Jiaxin+Zhang%2C+Tianhao+Chen%2C+Zitong+Chao%2C+Jishan+Hu%2C+and+Can+Yang%2E+2025%2E+Ugmathbench%3A+A+diverse+and+dynamic+benchmark+for+undergraduate%2Dlevel+mathematical+reasoning+with+large+language+models%2E+)
29. Yuqing Yang, Ethan Chern, Xipeng Qiu, Graham Neubig, and Pengfei Liu. 2023. Alignment for honesty. arXiv preprint arXiv:2312.07000. [find](https://scholar.google.com/scholar?q=Yuqing+Yang%2C+Ethan+Chern%2C+Xipeng+Qiu%2C+Graham+Neubig%2C+and+Pengfei+Liu%2E+2023%2E+Alignment+for+honesty%2E+arXiv+preprint+arXiv%3A2312%2E07000%2E)
30. Qiying Yu, Zheng Zhang, Ruofei Zhu, Yufeng Yuan, Xiaochen Zuo, Yu Yue, Weinan Dai, Tiantian Fan, Gaohong Liu, Lingjun Liu, and 1 others. 2025. [Dapo: An open-source llm reinforcement learning system at scale](https://md.research.elamur.ai/free/en/papers/yu-2025-dapo-open-source-llm-reinforcement/). arXiv preprint arXiv:2503.14476.
31. Xuan Zhang, Yongliang Shen, Zhe Zheng, Linjuan Wu, Wenqi Zhang, Yuchen Yan, Qiuying Peng, Jun Wang, and Weiming Lu. 2025. Asktoact: Enhancing llms tool use via self-correcting clarification. arXiv preprint arXiv:2503.01940. [find](https://scholar.google.com/scholar?q=Xuan+Zhang%2C+Yongliang+Shen%2C+Zhe+Zheng%2C+Linjuan+Wu%2C+Wenqi+Zhang%2C+Yuchen+Yan%2C+Qiuying+Peng%2C+Jun+Wang%2C+and+Weiming+Lu%2E+2025%2E+Asktoact%3A+Enhancing+llms+tool+use+via+self%2Dcorrecting+clarification%2E+arXiv+pre)
32. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric Xing, and 1 others. 2023. [Judging llm-as-a-judge with mt-bench and chatbot arena](https://md.research.elamur.ai/free/en/papers/zheng-2023-judging-llm-judge-mt-bench/). Advances in neural information processing systems, 36:46595–46623.
33. Xiaofan Zhou and Lu Cheng. 2025. [Robust uncertainty quantification for self-evolving large language models via continual domain pretraining](https://md.research.elamur.ai/free/en/papers/zhou-2026-robust-uncertainty-quantification-self-evolving/). arXiv preprint arXiv:2510.22931.
34. Yang Zhou, Sunzhu Li, Shunyu Liu, Wenkai Fang, Kongcheng Zhang, Jiale Zhao, Jingwen Yang, Yihe Zhou, Jianwei Lv, Tongya Zheng, Hengtong Lu, Wei Chen, Yan Xie, and Mingli Song. 2026. Breaking the exploration bottleneck: Rubric-scaffolded reinforcement learning for general llm reasoning. Preprint, arX… — [arXiv:2508.16949](https://arxiv.org/abs/2508.16949) [find](https://scholar.google.com/scholar?q=Yang+Zhou%2C+Sunzhu+Li%2C+Shunyu+Liu%2C+Wenkai+Fang%2C+Kongcheng+Zhang%2C+Jiale+Zhao%2C+Jingwen+Yang%2C+Yihe+Zhou%2C+Jianwei+Lv%2C+Tongya+Zheng%2C+Hengtong+Lu%2C+Wei+Chen%2C+Yan+Xie%2C+and+Mingli+Song%2E+2026%2E+Breaking+the+explo)

*Updated 31 October 2026*