---
title: "PRIME — paper by Wang et al. (2026) — summary and ideas"
description: "RLVR (reinforcement learning with verifiable rewards) relies on verifiers that usually check only whether a model's final answer… 32 references."
authors: ["Xiangfeng Wang", "Hangyu Guo", "Yanlin Lai", "Mitt Huang", "Liang Zhao", "Chengyuan Yao", "Yinmin Zhang", "Qi Han", "Xiaoxiaoren", "Chun Yuan", "Tong Xu", "Zheng Ge", "Xiangyu Zhang", "Daxin Jiang"]
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.683/"
doi: "10.18653/v1/2026.acl-long.683"
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-prime-process-outcome-alignment-benchmark/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wang-2026-prime-process-outcome-alignment-benchmark/)

[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/39/) / PRIME: A Process-Outcome Alignment Benchmark for Verifiable Reasoning in Mathematics and Engineering

# PRIME: A Process-Outcome Alignment Benchmark for Verifiable Reasoning in Mathematics and Engineering

**Authors:** Xiangfeng Wang, [Hangyu Guo](https://md.research.elamur.ai/free/en/authors/hangyu-guo/), Yanlin Lai, Mitt Huang, [Liang Zhao](https://md.research.elamur.ai/free/en/authors/liang-zhao/), [Chengyuan Yao](https://md.research.elamur.ai/free/en/authors/chengyuan-yao/), [Yinmin Zhang](https://md.research.elamur.ai/free/en/authors/yinmin-zhang/), [Qi Han](https://md.research.elamur.ai/free/en/authors/qi-han/), Xiaoxiaoren, Chun Yuan, [Tong Xu](https://md.research.elamur.ai/free/en/authors/tong-xu/), Zheng Ge, [Xiangyu Zhang](https://md.research.elamur.ai/free/en/authors/xiangyu-zhang/) and [Daxin Jiang](https://md.research.elamur.ai/free/en/authors/daxin-jiang/) · **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.683/) · [PDF](https://aclanthology.org/2026.acl-long.683.pdf) · [DOI 10.18653/v1/2026.acl-long.683](https://doi.org/10.18653/v1/2026.acl-long.683) · [OpenAlex](https://openalex.org/W7166835393)

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

## Summary
RLVR (reinforcement learning with verifiable rewards) relies on verifiers that usually check only whether a model's final answer matches the reference. Such verifiers reward "lucky guesses," where a correct answer comes from flawed reasoning. In a human-annotated sample of responses to 2,500 STEM prompts, the authors found this in 16.98% of cases. To address this, they built PRIME, a benchmark that tests whether a verifier checks both answer correctness and whether the derivation logically supports the answer.

The construction pipeline had five stages:
- collecting college-level STEM problems;
- filtering them with models for verifiability and answer correctness;
- generating solutions with many different LRMs;
- keeping the cases that are hard to verify;
- having domain experts label each sample.

Many existing verifiers often miss derivation errors. Using verifiers selected via PRIME as process-aware reward models in RLVR clearly beat outcome-only verification. Verifier accuracy on PRIME also tracked downstream training gains almost linearly. This makes PRIME a practical tool for choosing reward models.

## Key points
- **Data:** PRIME contains 2,530 expert-annotated samples, distilled from a pool of over 7M textbook and exam problems. It covers 16 STEM categories and about 480 sub-disciplines, and 18 domain experts did the labeling.
- **Filtering:** GPT-OSS-120B checked whether each problem was verifiable. Reference answers were then cross-checked with Gemini-3-Pro, GPT-5 and Claude-Sonnet-4.5.
- **Training gains:** With process-aware verification, Qwen3-14B-Base improved over the outcome-only baseline by 8.29% on AIME24, 9.12% on AIME25 and 7.31% on Beyond-AIME (absolute).
- **Verifier comparison:** Five verifiers were compared: CompassVerifier-32B, Qwen3-4B, Qwen3-8B, GPT-OSS-20B and GPT-OSS-120B. With GPT-OSS-120B, the average score of Qwen3-14B-Base rose from 24.79% to 47.07%.
- **Correlation:** PRIME accuracy correlates linearly with downstream RLVR performance, with R² = 0.937 for Qwen3-8B-Base and 0.920 for Qwen3-14B-Base. The benchmark and code are released on GitHub and Hugging Face.

*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

- [Tsinghua University](https://md.research.elamur.ai/free/en/universities/tsinghua-university/) — University, Beijing, China · 333 works · 87,019 citations
- [University of Science and Technology of China](https://md.research.elamur.ai/free/en/universities/university-of-science-and-technology-of-china/) — University, Hefei, China · 108 works · 112,698 citations
- [StepFun AI](https://md.research.elamur.ai/free/en/companies/stepfun-ai/) — Company, Shanghai, China · 3 works · 131 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 · [🧮 Computational Theory and Mathematics](https://md.research.elamur.ai/free/en/shelves/computer-science/computational-theory-and-mathematics/) · 252 works

🧭 Topics (2): [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,426 works · [📏 Evaluation & benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) · 1,220 works

## Cite

Cite as: Xiangfeng Wang, Hangyu Guo, Yanlin Lai, Mitt Huang, Liang Zhao, Chengyuan Yao, Yinmin Zhang, Qi Han, Xiaoxiaoren, Chun Yuan, Tong Xu, Zheng Ge, Xiangyu Zhang and Daxin Jiang. 2026. PRIME: A Process-Outcome Alignment Benchmark for Verifiable Reasoning in Mathematics and Engineering. 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{wang2026prime,
  title = {{PRIME: A Process-Outcome Alignment Benchmark for Verifiable Reasoning in Mathematics and Engineering}},
  author = {Xiangfeng Wang, Hangyu Guo, Yanlin Lai, Mitt Huang, Liang Zhao, Chengyuan Yao, Yinmin Zhang, Qi Han, Xiaoxiaoren, Chun Yuan, Tong Xu, Zheng Ge, Xiangyu Zhang and Daxin Jiang},
  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.683},
  url = {https://doi.org/10.18653/v1/2026.acl-long.683}
}
```

- [← previous Preventing Safety Drift in Large Language Models via Coupled Weight and Activation Constraints](https://md.research.elamur.ai/free/en/papers/peng-2026-preventing-safety-drift-large-language/)
- [next → PrimeLine@DravidianLangTech 2026: Abusive Tamil Comment Detection Using MuRIL](https://md.research.elamur.ai/free/en/papers/v-2026-primeline-dravidianlangtech-2026-abusive-tamil/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/39/)

## References

32 references, 17 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Marah Abdin, Jyoti Aneja, Harkirat Behl, Sébastien Bubeck, Ronen Eldan, Suriya Gunasekar, Michael Harrison, Russell J Hewett, Mojan Javaheripi, Piero Kauffmann, and 1 others. 2024. [Phi-4 technical report](https://md.research.elamur.ai/free/en/papers/abdin-2024-phi-4-technical-report/). arXiv preprint arXiv:2412.08905.
2. Sandhini Agarwal, Lama Ahmad, Jason Ai, Sam Altman, Andy Applebaum, Edwin Arbus, Rahul K Arora, Yu Bai, Bowen Baker, Haiming Bao, and 1 others. 2025. [gpt-oss-120b & gpt-oss-20b model card](https://md.research.elamur.ai/free/en/papers/agarwal-2025-gpt-oss-120b-gpt-oss/). arXiv preprint arXiv:2508.10925.
3. Anthropic. 2025a. Claude Opus 4 and Claude Sonnet 4 System Card. System card, Anthropic. Accessed: 2026-01-05. [find](https://scholar.google.com/scholar?q=Anthropic%2E+2025a%2E+Claude+Opus+4+and+Claude+Sonnet+4+System+Card%2E+System+card%2C+Anthropic%2E+Accessed%3A+2026%2D01%2D05%2E)
4. Anthropic. 2025b. Claude Sonnet 4.5 System Card. System card, Anthropic. Accessed: 2026-01-05. [find](https://scholar.google.com/scholar?q=Anthropic%2E+2025b%2E+Claude+Sonnet+4%2E5+System+Card%2E+System+card%2C+Anthropic%2E+Accessed%3A+2026%2D01%2D05%2E)
5. ByteDance-Seed. 2025. Beyondaime: Advancing math reasoning evaluation beyond high school olympiads. [https://huggingface.co/ datasets/ByteDance-Seed/BeyondAIME](https://huggingface.co/datasets/ByteDance-Seed/ BeyondAIME). [find](https://scholar.google.com/scholar?q=ByteDance%2DSeed%2E+2025%2E+Beyondaime%3A+Advancing+math+reasoning+evaluation+beyond+high+school+olympiads%2E+%5Bhttps%3A%2F%2Fhuggingface%2Eco%2F+datasets%2FByteDance%2DSeed%2FBeyondAIME%5D%28https%3A%2F%2Fhuggingface%2Eco%2Fdatasets%2FByteDan)
6. ByteDance Seed Team. 2025. Seed 1.6: Multimodal models with adaptive deep thinking. [https://seed.bytedance.com/en/seed1_6](https://seed.bytedance.com/en/seed1_6). Technical Report. Accessed: 2026-01-05. [↗](https://seed.bytedance.com/en/seed1_6)
7. Ding Chen, Qingchen Yu, Pengyuan Wang, Wentao Zhang, Bo Tang, Feiyu Xiong, Xinchi Li, Minchuan Yang, and Zhiyu Li. 2025. xverify: Efficient answer verifier for reasoning model evaluations. CoRR, abs/2504.10481. — [arXiv:2504.10481](https://arxiv.org/abs/2504.10481) [find](https://scholar.google.com/scholar?q=Ding+Chen%2C+Qingchen+Yu%2C+Pengyuan+Wang%2C+Wentao+Zhang%2C+Bo+Tang%2C+Feiyu+Xiong%2C+Xinchi+Li%2C+Minchuan+Yang%2C+and+Zhiyu+Li%2E+2025%2E+xverify%3A+Efficient+answer+verifier+for+reasoning+model+evaluations%2E+CoRR%2C+abs%2F2)
8. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, and 1 others. 2025. [Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities](https://md.research.elamur.ai/free/en/papers/comanici-2025-gemini-2-5-pushing-frontier/). arX…
9. Xinrun Du, Yifan Yao, Kaijing Ma, Bingli Wang, Tianyu Zheng, King Zhu, Minghao Liu, Yiming Liang, Xiaolong Jin, Zhenlin Wei, and 1 others. 2025. Supergpqa: Scaling llm evaluation across 285 graduate disciplines. arXiv preprint arXiv:2502.14739. [find](https://scholar.google.com/scholar?q=Xinrun+Du%2C+Yifan+Yao%2C+Kaijing+Ma%2C+Bingli+Wang%2C+Tianyu+Zheng%2C+King+Zhu%2C+Minghao+Liu%2C+Yiming+Liang%2C+Xiaolong+Jin%2C+Zhenlin+Wei%2C+and+1+others%2E+2025%2E+Supergpqa%3A+Scaling+llm+evaluation+across+285+graduate+d)
10. Ruixiang Feng, Zhenwei An, Yuntao Wen, Ran Le, Yiming Jia, Chen Yang, Zongchao Chen, Lisi Chen, Shen Gao, Shuo Shang, and 1 others. 2025. Cosin-everifier: Tool-augmented answer verification for computation-oriented scientific questions. arXiv preprint arXiv:2512.01224. — [arXiv:2512.01224](https://arxiv.org/abs/2512.01224) [find](https://scholar.google.com/scholar?q=Ruixiang+Feng%2C+Zhenwei+An%2C+Yuntao+Wen%2C+Ran+Le%2C+Yiming+Jia%2C+Chen+Yang%2C+Zongchao+Chen%2C+Lisi+Chen%2C+Shen+Gao%2C+Shuo+Shang%2C+and+1+others%2E+2025%2E+Cosin%2Deverifier%3A+Tool%2Daugmented+answer+verification+for+comput)
11. Google DeepMind. 2025. Gemini: The Most Capable and General Model We’ve Ever Built. Accessed: 2026-01-05. [find](https://scholar.google.com/scholar?q=Google+DeepMind%2E+2025%2E+Gemini%3A+The+Most+Capable+and+General+Model+We%E2%80%99ve+Ever+Built%2E+Accessed%3A+2026%2D01%2D05%2E)
12. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, 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.
13. Hynek Kydlíček and Hugging Face Team. 2025. Math-Verify: A robust mathematical expression evaluation system. Accessed: 2026-01-05. [find](https://scholar.google.com/scholar?q=Hynek+Kydl%C3%ADc%CC%8Cek+and+Hugging+Face+Team%2E+2025%2E+Math%2DVerify%3A+A+robust+mathematical+expression+evaluation+system%2E+Accessed%3A+2026%2D01%2D05%2E)
14. Hunter Lightman, Vineet Kosaraju, Yura Burda, Harri Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe. 2023. [Let’s verify step by step](https://md.research.elamur.ai/free/en/papers/lightman-2023-let-s-verify-step-step/). arXiv preprint arXiv:2305.20050.
15. Aixin Liu, Aoxue Mei, Bangcai Lin, Bing Xue, Bingxuan Wang, Bingzheng Xu, Bochao Wu, Bowei Zhang, Chaofan Lin, Chen Dong, and 1 others. 2025a. [Deepseek-v3](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-v3-2-pushing/). 2: Pushing the frontier of open large language models. arXiv preprint arXiv:2512.02556.
16. Shudong Liu, Hongwei Liu, Junnan Liu, Linchen Xiao, Songyang Gao, Chengqi Lyu, Yuzhe Gu, Wenwei Zhang, Derek F. Wong, Songyang Zhang, and Kai Chen. 2025b. [Compassverifier: A unified and robust verifier for llms evaluation and outcome reward](https://md.research.elamur.ai/free/en/papers/liu-2025-compassverifier-unified-robust-verifier-llms/). CoRR, abs/2508.03686.
17. Dakota Mahan, Duy Van Phung, Rafael Rafailov, Chase Blagden, Nathan Lile, Louis Castricato, Jan-Philipp Fränken, Chelsea Finn, and Alon Albalak. 2024. Generative reward models. arXiv preprint arXiv:2410.12832. [find](https://scholar.google.com/scholar?q=Dakota+Mahan%2C+Duy+Van+Phung%2C+Rafael+Rafailov%2C+Chase+Blagden%2C+Nathan+Lile%2C+Louis+Castricato%2C+Jan%2DPhilipp+Fr%C3%A4nken%2C+Chelsea+Finn%2C+and+Alon+Albalak%2E+2024%2E+Generative+reward+models%2E+arXiv+preprint+arXiv%3A24)
18. OpenAI. 2025. GPT-5 System Card. System card, OpenAI. Accessed: 2026-01-05. [find](https://scholar.google.com/scholar?q=OpenAI%2E+2025%2E+GPT%2D5+System+Card%2E+System+card%2C+OpenAI%2E+Accessed%3A+2026%2D01%2D05%2E)
19. 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.
20. ByteDance Seed, Jiaze Chen, Tiantian Fan, Xin Liu, Lingjun Liu, Zhiqi Lin, Mingxuan Wang, Chengyi Wang, Xiangpeng Wei, Wenyuan Xu, and 1 others. 2025. Seed1. 5-thinking: Advancing superb reasoning models with reinforcement learning. arXiv preprint arXiv:2504.13914. — [Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning](https://md.research.elamur.ai/free/en/papers/bytedance-seed-2025-seed1-5-thinking-advancing/)
21. 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.
22. Yi Su, Dian Yu, Linfeng Song, Juntao Li, Haitao Mi, Zhaopeng Tu, Min Zhang, and Dong Yu. 2025. Crossing the reward bridge: Expanding RL with verifiable rewards across diverse domains. CoRR, abs/2503.23829. — [arXiv:2503.23829](https://arxiv.org/abs/2503.23829) [find](https://scholar.google.com/scholar?q=Yi+Su%2C+Dian+Yu%2C+Linfeng+Song%2C+Juntao+Li%2C+Haitao+Mi%2C+Zhaopeng+Tu%2C+Min+Zhang%2C+and+Dong+Yu%2E+2025%2E+Crossing+the+reward+bridge%3A+Expanding+RL+with+verifiable+rewards+across+diverse+domains%2E+CoRR%2C+abs%2F2503%2E2)
23. Kimi Team, Yifan Bai, Yiping Bao, Guanduo Chen, Jiahao Chen, Ningxin Chen, Ruijue Chen, Yanru Chen, Yuankun Chen, Yutian Chen, and 1 others. 2025. [Kimi k2: Open agentic intelligence](https://md.research.elamur.ai/free/en/papers/kimi-team-2025-kimi-k2-open-agentic/). arXiv preprint arXiv:2507.20534.
24. Jonathan Uesato, Nate Kushman, Ramana Kumar, Francis Song, Noah Siegel, Lisa Wang, Antonia Creswell, Geoffrey Irving, and Irina Higgins. 2022. [Solving math word problems with process-and outcome-based feedback](https://md.research.elamur.ai/free/en/papers/uesato-2022-solving-math-word-problems-process/). arXiv preprint arXiv:2211.14275.
25. Peiyi Wang, Lei Li, Zhihong Shao, Runxin Xu, Damai Dai, Yifei Li, Deli Chen, Yu Wu, and Zhifang Sui. 2024a. [Math-shepherd: Verify and reinforce llms step-by-step without human annotations](https://md.research.elamur.ai/free/en/papers/wang-2024-math-shepherd-verify-reinforce-llms/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Pape…
26. Yubo Wang, Xueguang Ma, Ge Zhang, Yuansheng Ni, Abhranil Chandra, Shiguang Guo, Weiming Ren, Aaran Arulraj, Xuan He, Ziyan Jiang, and 1 others. 2024b. [Mmlu-pro: A more robust and challenging multi-task language understanding benchmark](https://md.research.elamur.ai/free/en/papers/wang-2024-mmlu-pro-more-robust-challenging/). Advances in Neural Information Processing Systems, 37:95266–9529…
27. 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.
28. An Yang, Beichen Zhang, Binyuan Hui, Bofei Gao, Bowen Yu, Chengpeng Li, Dayiheng Liu, Jianhong Tu, Jingren Zhou, Junyang Lin, and 1 others. 2024. Qwen2. 5-math technical report: Toward mathematical expert model via self-improvement. arXiv preprint arXiv:2409.12122. — [Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement](https://md.research.elamur.ai/free/en/papers/yang-2024-qwen2-5-math-technical-report/)
29. Z.ai Team. 2025. GLM-4.6: Advanced Agentic, Reasoning and Coding Capabilities. Accessed: 2026-01-05. [find](https://scholar.google.com/scholar?q=Z%2Eai+Team%2E+2025%2E+GLM%2D4%2E6%3A+Advanced+Agentic%2C+Reasoning+and+Coding+Capabilities%2E+Accessed%3A+2026%2D01%2D05%2E)
30. Kaiyan Zhang, Yuxin Zuo, Bingxiang He, Youbang Sun, Runze Liu, Che Jiang, Yuchen Fan, Kai Tian, Guoli Jia, Pengfei Li, and 1 others. 2025. A survey of reinforcement learning for large reasoning models. arXiv preprint arXiv:2509.08827. — [arXiv:2509.08827](https://arxiv.org/abs/2509.08827) [find](https://scholar.google.com/scholar?q=Kaiyan+Zhang%2C+Yuxin+Zuo%2C+Bingxiang+He%2C+Youbang+Sun%2C+Runze+Liu%2C+Che+Jiang%2C+Yuchen+Fan%2C+Kai+Tian%2C+Guoli+Jia%2C+Pengfei+Li%2C+and+1+others%2E+2025%2E+A+survey+of+reinforcement+learning+for+large+reasoning+models)
31. Lunjun Zhang, Arian Hosseini, Hritik Bansal, Mehran Kazemi, Aviral Kumar, and Rishabh Agarwal. 2024. [Generative verifiers: Reward modeling as next-token prediction](https://md.research.elamur.ai/free/en/papers/zhang-2024-generative-verifiers-reward-modeling-as/), 2024. URL https://arxiv.org/abs/2408.15240, 1. [↗](https://arxiv.org/abs/2408.15240)
32. Shenghe Zheng, Chenyu Huang, Fangchen Yu, Junchi Yao, Jingqi Ye, Tao Chen, Yun Luo, Ning Ding, Lei Bai, Ganqu Cui, and Peng Ye. 2025. Sci-verifier: Scientific verifier with thinking. CoRR, abs/2509.24285. *Figure 6: Answer type distribution of P RIME .* CompassVerifier xVerify-9B-C Tencent-Qwen-7B-R… — [arXiv:2509.24285](https://arxiv.org/abs/2509.24285) [find](https://scholar.google.com/scholar?q=Shenghe+Zheng%2C+Chenyu+Huang%2C+Fangchen+Yu%2C+Junchi+Yao%2C+Jingqi+Ye%2C+Tao+Chen%2C+Yun+Luo%2C+Ning+Ding%2C+Lei+Bai%2C+Ganqu+Cui%2C+and+Peng+Ye%2E+2025%2E+Sci%2Dverifier%3A+Scientific+verifier+with+thinking%2E+CoRR%2C+abs%2F2509%2E24)

*Updated 1 November 2026*