---
title: "RbtAct — paper by Wu et al. (2026) — summary and ideas"
description: "The authors define a new task, perspective-conditioned segment-level review feedback generation. 35 references, cited by 2 papers here."
authors: ["Sihong Wu", "Yiling Ma", "Yilun Zhao", "Tiansheng Hu", "Owen Jiang", "Manasi Patwardhan", "Arman Cohan"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.1696/"
doi: "10.18653/v1/2026.findings-acl.1696"
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/wu-2026-rbtact-rebuttal-as-supervision-actionable/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wu-2026-rbtact-rebuttal-as-supervision-actionable/)

[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/8/) / RbtAct: Rebuttal as Supervision for Actionable Review Feedback Generation

# RbtAct: Rebuttal as Supervision for Actionable Review Feedback Generation

**Authors:** [Sihong Wu](https://md.research.elamur.ai/free/en/authors/sihong-wu/), [Yiling Ma](https://md.research.elamur.ai/free/en/authors/yiling-ma/), [Yilun Zhao](https://md.research.elamur.ai/free/en/authors/yilun-zhao/), [Tiansheng Hu](https://md.research.elamur.ai/free/en/authors/tiansheng-hu/), [Owen Jiang](https://md.research.elamur.ai/free/en/authors/owen-jiang/), [Manasi Patwardhan](https://md.research.elamur.ai/free/en/authors/manasi-patwardhan/) and [Arman Cohan](https://md.research.elamur.ai/free/en/authors/arman-cohan/) · **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.1696/) · [PDF](https://aclanthology.org/2026.findings-acl.1696.pdf) · [DOI 10.18653/v1/2026.findings-acl.1696](https://doi.org/10.18653/v1/2026.findings-acl.1696) · [OpenAlex](https://openalex.org/W7166840325)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [2](https://md.research.elamur.ai/free/en/papers/wu-2026-rbtact-rebuttal-as-supervision-actionable/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
LLM-drafted peer reviews often sound fluent but stay generic, so authors get little guidance they can actually implement. RbtAct addresses this by treating author rebuttals as an implicit signal of which reviewer comments are useful. A comment that led authors to revise their paper, or to commit to a concrete plan, is taken as more actionable than one they simply argued against. The authors define a new task, perspective-conditioned segment-level review feedback generation. Given a full paper and a target aspect such as experiments or writing, a model must write one focused comment. They also build RMR-75K from ICLR 2024 OpenReview data. It links individual review points to the rebuttal passages answering them and labels each with a perspective and an impact category. Llama-3.1-8B-Instruct is first fine-tuned on review segments, then trained with DPO on preference pairs ranked by impact. On an ICLR 2025 test set, both human experts and an LLM judge rate its output as the most actionable and specific among strong baselines, with grounding and relevance unchanged. This makes rebuttals a cheap, large-scale form of human feedback.

## Key points
- **RMR-75K** contains 75,542 examples, each with a review segment, a perspective label, the matching author response, and an impact category (revision done, revision planned, or defense only). It is roughly 150× larger than DISAPERE, which has 506 pairs.
- **Training pipeline:** SFT of Llama-3.1-8B-Instruct, then DPO. Preference pairs are formed from segments that share a paper and perspective but differ in impact category.
- **Actionability scores:** RbtAct scored 3.46/5 from humans and 3.38/5 from the LLM judge, the best results. It also led on average pairwise win rate across seven perspectives, ahead of GPT-5-chat, Llama-3.1-70B, DeepSeek-V3.2, Qwen-3-32B, DeepReviewer-14b, MARG and LimGen.
- **Human–judge agreement:** Model rankings from humans and the LLM judge agreed strongly on actionability (Spearman ρ=0.94). Agreement at the level of individual items was moderate (ρ=0.52).
- **Limitations:** Rebuttals reflect only short-term author uptake and may contain strategic promises. The data also covers only OpenReview computer-science venues.

*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

- [New York University](https://md.research.elamur.ai/free/en/universities/new-york-university/) — University, New York, United States · 201 works · 100,180 citations
- [Yale University](https://md.research.elamur.ai/free/en/universities/yale-university/) — University, New Haven, United States · 95 works · 12,765 citations
- [Tata Consultancy Services](https://md.research.elamur.ai/free/en/companies/tata-consultancy-services/) — Company, Pune, India · 8 works · 35 citations

📚 Shelves (5): [🖥️ 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 · [🎭 History and Philosophy of Science](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/history-and-philosophy-of-science/) · 53 works

## Cite

Cite as: Sihong Wu, Yiling Ma, Yilun Zhao, Tiansheng Hu, Owen Jiang, Manasi Patwardhan and Arman Cohan. 2026. RbtAct: Rebuttal as Supervision for Actionable Review Feedback Generation. 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{wu2026rbtact,
  title = {{RbtAct: Rebuttal as Supervision for Actionable Review Feedback Generation}},
  author = {Sihong Wu, Yiling Ma, Yilun Zhao, Tiansheng Hu, Owen Jiang, Manasi Patwardhan and Arman Cohan},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1696},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1696}
}
```

- [← previous Query Decomposition for RAG: Balancing Exploration-Exploitation](https://md.research.elamur.ai/free/en/papers/petcu-2026-query-decomposition-rag-balancing-exploration/)
- [next → ReContraster: Making Your Posters Stand Out with Regional Contrast](https://md.research.elamur.ai/free/en/papers/zhang-2026-recontraster-making-your-posters-stand/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/8/)

## References

35 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Satanjeev Banerjee and Alon Lavie. 2005. [METEOR: An automatic metric for MT evaluation with improved correlation with human judgments](https://md.research.elamur.ai/free/en/papers/banerjee-2005-meteor-automatic-metric-mt-evaluation/). In Proceedings of the ACL Workshop on Intrinsic and Extrinsic Evaluation Measures for Machine Translation and/or Summarization, pages 65–72, Ann Arbor, Michigan. Ass…
2. Eric Chamoun, Michael Schlichtkrull, and Andreas Vlachos. 2024. Automated focused feedback generation for scientific writing assistance. In Findings of the Association for Computational Linguistics: ACL 2024, pages 9742–9763, Bangkok, Thailand. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Eric+Chamoun%2C+Michael+Schlichtkrull%2C+and+Andreas+Vlachos%2E+2024%2E+Automated+focused+feedback+generation+for+scientific+writing+assistance%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+A)
3. Gautam Choudhary, Natwar Modani, and Nitish Maurya. 2021. ReAct: A Review Comment Dataset for Actionability (and more), page 336–343. Springer International Publishing. [find](https://scholar.google.com/scholar?q=Gautam+Choudhary%2C+Natwar+Modani%2C+and+Nitish+Maurya%2E+2021%2E+ReAct%3A+A+Review+Comment+Dataset+for+Actionability+%28and+more%29%2C+page+336%E2%80%93343%2E+Springer+International+Publishing%2E)
4. Mike D’Arcy, Tom Hope, Larry Birnbaum, and Doug Downey. 2024a. Marg: Multi-agent review generation for scientific papers. [find](https://scholar.google.com/scholar?q=Mike+D%E2%80%99Arcy%2C+Tom+Hope%2C+Larry+Birnbaum%2C+and+Doug+Downey%2E+2024a%2E+Marg%3A+Multi%2Dagent+review+generation+for+scientific+papers%2E)
5. Mike D’Arcy, Alexis Ross, Erin Bransom, Bailey Kuehl, Jonathan Bragg, Tom Hope, and Doug Downey. 2024b. [Aries: A corpus of scientific paper edits made in response to peer reviews](https://md.research.elamur.ai/free/en/papers/arcy-2024-aries-corpus-scientific-paper-edits/).
6. DeepSeek-AI. 2025. Introducing deepseek-v3.2-exp. [https://api-docs.deepseek.com/](https://api-docs.deepseek.com/) news/news250929. Accessed: 2025-10-07. [↗](https://api-docs.deepseek.com/)
7. Nils Dycke, Ilia Kuznetsov, and Iryna Gurevych. 2023. [NLPeer: A unified resource for the computational study of peer review](https://md.research.elamur.ai/free/en/papers/dycke-2023-nlpeer-unified-resource-computational-study/). In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 5049– 5073, Toronto, Canada. Association for Computat…
8. Abdur Rahman Bin Md Faizullah, Ashok Urlana, and Rahul Mishra. 2024. Limgen: Probing the llms for generating suggestive limitations of research papers. [find](https://scholar.google.com/scholar?q=Abdur+Rahman+Bin+Md+Faizullah%2C+Ashok+Urlana%2C+and+Rahul+Mishra%2E+2024%2E+Limgen%3A+Probing+the+llms+for+generating+suggestive+limitations+of+research+papers%2E)
9. Zhaolin Gao, Kianté Brantley, and Thorsten Joachims. 2024. [Reviewer2: Optimizing review generation through prompt generation](https://md.research.elamur.ai/free/en/papers/gao-2024-reviewer2-optimizing-review-generation-through/).
10. Tirthankar Ghosal, Shubham Kumar, Prasenjit Kumar Bharti, and Asif Ekbal. 2022. Peer review analyze: A novel benchmark resource for computational analysis of peer reviews. PLOS ONE, 17(1):e0259238. [find](https://scholar.google.com/scholar?q=Tirthankar+Ghosal%2C+Shubham+Kumar%2C+Prasenjit+Kumar+Bharti%2C+and+Asif+Ekbal%2E+2022%2E+Peer+review+analyze%3A+A+novel+benchmark+resource+for+computational+analysis+of+peer+reviews%2E+PLOS+ONE%2C+17%281%29%3Ae0259238%2E)
11. M. Hosseini and S. P. J. M. Horbach. 2023. Fighting reviewer fatigue or amplifying bias? considerations and recommendations for use of chatgpt and other large language models in scholarly peer review. Research Integrity and Peer Review, 8(4):1–12. [find](https://scholar.google.com/scholar?q=M%2E+Hosseini+and+S%2E+P%2E+J%2E+M%2E+Horbach%2E+2023%2E+Fighting+reviewer+fatigue+or+amplifying+bias%3F+considerations+and+recommendations+for+use+of+chatgpt+and+other+large+language+models+in+scholarly+peer+review%2E)
12. Yiqiao Jin, Qinlin Zhao, Yiyang Wang, Hao Chen, Kaijie Zhu, Yijia Xiao, and Jindong Wang. 2024. AgentReview: Exploring peer review dynamics with LLM agents. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 1208–1226, Miami, Florida, USA. Association fo… [find](https://scholar.google.com/scholar?q=Yiqiao+Jin%2C+Qinlin+Zhao%2C+Yiyang+Wang%2C+Hao+Chen%2C+Kaijie+Zhu%2C+Yijia+Xiao%2C+and+Jindong+Wang%2E+2024%2E+AgentReview%3A+Exploring+peer+review+dynamics+with+LLM+agents%2E+In+Proceedings+of+the+2024+Conference+on+Em)
13. Dongyeop Kang, Waleed Ammar, Bhavana Dalvi, Madeleine van Zuylen, Sebastian Kohlmeier, Eduard Hovy, and Roy Schwartz. 2018. [A dataset of peer reviews (PeerRead): Collection, insights and NLP applications](https://md.research.elamur.ai/free/en/papers/kang-2018-dataset-peer-reviews-peerread-collection/). In Proceedings of the 2018 Conference of the North American Chapter of the Association for Comp…
14. Neha Nayak Kennard, Tim O’Gorman, Rajarshi Das, Akshay Sharma, Chhandak Bagchi, Matthew Clinton, Pranay Kumar Yelugam, Hamed Zamani, and Andrew McCallum. 2022. DISAPERE: A dataset for discourse structure in peer review discussions. In Proceedings of the 2022 Conference of the North American Chapter… [find](https://scholar.google.com/scholar?q=Neha+Nayak+Kennard%2C+Tim+O%E2%80%99Gorman%2C+Rajarshi+Das%2C+Akshay+Sharma%2C+Chhandak+Bagchi%2C+Matthew+Clinton%2C+Pranay+Kumar+Yelugam%2C+Hamed+Zamani%2C+and+Andrew+McCallum%2E+2022%2E+DISAPERE%3A+A+dataset+for+discourse+struct)
15. 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, Barcelona, Spain. Association for Computational Linguistics.
16. Jialiang Lin, Jiaxin Song, Zhangping Zhou, Yidong Chen, and Xiaodong Shi. 2023. Moprd: A multidisciplinary open peer review dataset. Neural Computing and Applications, 35(34):24191–24206. [find](https://scholar.google.com/scholar?q=Jialiang+Lin%2C+Jiaxin+Song%2C+Zhangping+Zhou%2C+Yidong+Chen%2C+and+Xiaodong+Shi%2E+2023%2E+Moprd%3A+A+multidisciplinary+open+peer+review+dataset%2E+Neural+Computing+and+Applications%2C+35%2834%29%3A24191%E2%80%9324206%2E)
17. Ryan Liu and Nihar B. Shah. 2023. Reviewergpt? an exploratory study on using large language models for paper reviewing. [find](https://scholar.google.com/scholar?q=Ryan+Liu+and+Nihar+B%2E+Shah%2E+2023%2E+Reviewergpt%3F+an+exploratory+study+on+using+large+language+models+for+paper+reviewing%2E)
18. Meta AI. 2024. Introducing llama 3.1: Our most capable models to date. [https://ai.meta.com/](https://ai.meta.com/) blog/meta-llama-3-1/. Accessed: 2025-10-07. [↗](https://ai.meta.com/)
19. OpenAI. 2025. Gpt-5. [https://openai.com/](https://openai.com/) gpt-5/. Accessed: 2025-10-07. [↗](https://openai.com/)
20. 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 Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational…
21. Maja Popović. 2015. [chrF: character n-gram F-score for automatic MT evaluation](https://md.research.elamur.ai/free/en/papers/popovic-2015-chrf-character-n-gram-f/). In Proceedings of the Tenth Workshop on Statistical Machine Translation, pages 392–395, Lisbon, Portugal. Association for Computational Linguistics.
22. Sukannya Purkayastha, Anne Lauscher, and Iryna Gurevych. 2023. Exploring jiu-jitsu argumentation for writing peer review rebuttals. [find](https://scholar.google.com/scholar?q=Sukannya+Purkayastha%2C+Anne+Lauscher%2C+and+Iryna+Gurevych%2E+2023%2E+Exploring+jiu%2Djitsu+argumentation+for+writing+peer+review+rebuttals%2E)
23. Rafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D. Manning, and Chelsea Finn. 2024. [Direct preference optimization: Your language model is secretly a reward model](https://md.research.elamur.ai/free/en/papers/rafailov-2023-direct-preference-optimization-language-model/).
24. Zachary Robertson. 2023. Gpt4 is slightly helpful for peer-review assistance: A pilot study. [find](https://scholar.google.com/scholar?q=Zachary+Robertson%2E+2023%2E+Gpt4+is+slightly+helpful+for+peer%2Dreview+assistance%3A+A+pilot+study%2E)
25. Abdelrahman Sadallah, Tim Baumgärtner, Iryna Gurevych, and Ted Briscoe. 2025. The good, the bad and the constructive: Automatically measuring peer review’s utility for authors. [find](https://scholar.google.com/scholar?q=Abdelrahman+Sadallah%2C+Tim+Baumg%C3%A4rtner%2C+Iryna+Gurevych%2C+and+Ted+Briscoe%2E+2025%2E+The+good%2C+the+bad+and+the+constructive%3A+Automatically+measuring+peer+review%E2%80%99s+utility+for+authors%2E)
26. Hyungyu Shin, Jingyu Tang, Yoonjoo Lee, Nayoung Kim, Hyunseung Lim, Ji Yong Cho, Hwajung Hong, Moontae Lee, and Juho Kim. 2025. [Mind the blind spots: A focus-level evaluation framework for llm reviews](https://md.research.elamur.ai/free/en/papers/shin-2025-mind-blind-spots-focus-level/).
27. Cheng Tan, Dongxin Lyu, Siyuan Li, Zhangyang Gao, Jingxuan Wei, Siqi Ma, Zicheng Liu, and Stan Z. Li. 2024. Peer review as a multi-turn and long-context dialogue with role-based interactions. [find](https://scholar.google.com/scholar?q=Cheng+Tan%2C+Dongxin+Lyu%2C+Siyuan+Li%2C+Zhangyang+Gao%2C+Jingxuan+Wei%2C+Siqi+Ma%2C+Zicheng+Liu%2C+and+Stan+Z%2E+Li%2E+2024%2E+Peer+review+as+a+multi%2Dturn+and+long%2Dcontext+dialogue+with+role%2Dbased+interactions%2E)
28. Qwen Team. 2025. [Qwen3 technical report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). arXiv preprint arXiv:2505.09388.
29. Bin Wang, Chao Xu, Xiaomeng Zhao, Linke Ouyang, Fan Wu, Zhiyuan Zhao, Rui Xu, Kaiwen Liu, Yuan Qu, Fukai Shang, Bo Zhang, Liqun Wei, Zhihao Sui, Wei Li, Botian Shi, Yu Qiao, Dahua Lin, and Conghui He. 2024. Mineru: An open-source solution for precise document content extraction. [find](https://scholar.google.com/scholar?q=Bin+Wang%2C+Chao+Xu%2C+Xiaomeng+Zhao%2C+Linke+Ouyang%2C+Fan+Wu%2C+Zhiyuan+Zhao%2C+Rui+Xu%2C+Kaiwen+Liu%2C+Yuan+Qu%2C+Fukai+Shang%2C+Bo+Zhang%2C+Liqun+Wei%2C+Zhihao+Sui%2C+Wei+Li%2C+Botian+Shi%2C+Yu+Qiao%2C+Dahua+Lin%2C+and+Conghui+He%2E)
30. Yixuan Weng, Minjun Zhu, Guangsheng Bao, Hongbo Zhang, Jindong Wang, Yue Zhang, and Linyi Yang. 2025. Cycleresearcher: Improving automated research via automated review. [find](https://scholar.google.com/scholar?q=Yixuan+Weng%2C+Minjun+Zhu%2C+Guangsheng+Bao%2C+Hongbo+Zhang%2C+Jindong+Wang%2C+Yue+Zhang%2C+and+Linyi+Yang%2E+2025%2E+Cycleresearcher%3A+Improving+automated+research+via+automated+review%2E)
31. Po-Cheng Wu, An-Zi Yen, Hen-Hsen Huang, and Hsin-Hsi Chen. 2022. Incorporating peer reviews and rebuttal counter-arguments for meta-review generation. In Proceedings of the 31st ACM International Conference on Information & Knowledge Management, CIKM ’22, page 2189–2198, New York, NY, USA. Associati… [find](https://scholar.google.com/scholar?q=Po%2DCheng+Wu%2C+An%2DZi+Yen%2C+Hen%2DHsen+Huang%2C+and+Hsin%2DHsi+Chen%2E+2022%2E+Incorporating+peer+reviews+and+rebuttal+counter%2Darguments+for+meta%2Dreview+generation%2E+In+Proceedings+of+the+31st+ACM+International+Conf)
32. Weizhe Yuan, Pengfei Liu, and Graham Neubig. 2021. Can we automate scientific reviewing? [find](https://scholar.google.com/scholar?q=Weizhe+Yuan%2C+Pengfei+Liu%2C+and+Graham+Neubig%2E+2021%2E+Can+we+automate+scientific+reviewing%3F)
33. Daoze Zhang, Zhijian Bao, Sihang Du, Zhiyi Zhao, Kuangling Zhang, Dezheng Bao, and Yang Yang. 2025. Re 2 : A consistency-ensured dataset for full-stage peer review and multi-turn rebuttal discussions. [find](https://scholar.google.com/scholar?q=Daoze+Zhang%2C+Zhijian+Bao%2C+Sihang+Du%2C+Zhiyi+Zhao%2C+Kuangling+Zhang%2C+Dezheng+Bao%2C+and+Yang+Yang%2E+2025%2E+Re+2+%3A+A+consistency%2Densured+dataset+for+full%2Dstage+peer+review+and+multi%2Dturn+rebuttal+discussions%2E)
34. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric P. Xing, Hao Zhang, Joseph E. Gonzalez, and Ion Stoica. 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/).
35. Minjun Zhu, Yixuan Weng, Linyi Yang, and Yue Zhang. 2025. Deepreview: Improving llm-based paper review with human-like deep thinking process. [find](https://scholar.google.com/scholar?q=Minjun+Zhu%2C+Yixuan+Weng%2C+Linyi+Yang%2C+and+Yue+Zhang%2E+2025%2E+Deepreview%3A+Improving+llm%2Dbased+paper+review+with+human%2Dlike+deep+thinking+process%2E)

*Updated 1 November 2026*