---
title: "ImpRIF — paper by Yang et al. (2026) — summary and ideas"
description: "Large language models often fail on instructions whose real requirements are hidden. 48 references."
authors: ["Yuancheng Yang", "Lin Yang", "Xu Wang", "Chao Tong", "Haihua Yang"]
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.1796/"
doi: "10.18653/v1/2026.acl-long.1796"
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/yang-2026-imprif-stronger-implicit-reasoning-leads/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/yang-2026-imprif-stronger-implicit-reasoning-leads/)

[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/37/) / ImpRIF: Stronger Implicit Reasoning Leads to Better Complex Instruction Following

# ImpRIF: Stronger Implicit Reasoning Leads to Better Complex Instruction Following

**Authors:** Yuancheng Yang, [Lin Yang](https://md.research.elamur.ai/free/en/authors/lin-yang/), [Xu Wang](https://md.research.elamur.ai/free/en/authors/xu-wang/), Chao Tong and [Haihua Yang](https://md.research.elamur.ai/free/en/authors/haihua-yang/) · **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.1796/) · [PDF](https://aclanthology.org/2026.acl-long.1796.pdf) · [DOI 10.18653/v1/2026.acl-long.1796](https://doi.org/10.18653/v1/2026.acl-long.1796) · [OpenAlex](https://openalex.org/W7166801124)

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

## Summary
Large language models often fail on instructions whose real requirements are hidden. A constraint may only become concrete after the model recalls a fact, does some arithmetic, or checks a condition, for example a list whose length must match how many brains an octopus has. ImpRIF addresses this gap. The authors represent each such instruction as an explicit reasoning graph (ERG): a DAG whose nodes are conditional, mathematical, or knowledge steps that code can verify, and whose edges record which step depends on which.

From these graphs they synthesize large single- and multi-turn instruction datasets with adjustable difficulty. Each instruction comes with a reference answer whose chain-of-thought walks through the graph. Qwen3-4B, -8B and -32B are first fine-tuned on this data and then trained with GRPO using several reward signals.

On five complex instruction-following benchmarks, the trained models clearly outperform their base versions. The 32B model is reported to match or beat some larger or proprietary systems. This suggests that strengthening implicit reasoning can noticeably improve instruction adherence.

## Key points
- **Graph formalization:** ERG nodes fall into three types: conditional checks, math operations, and knowledge inference. The intermediate logic is hidden inside one natural-language instruction, and code or programmatic checks verify each node type.
- **Training pipeline:**
  - The first stage is full-parameter SFT on ERG-aligned CoT, using a `<think>`/`<answer>` format.
  - The second stage is GRPO with code-based rewards, LLM-judged rubric rewards, a thinking-process reward, and partial reward.
  - Combining the thinking and partial rewards worked best on most external benchmarks.
- **Thinking-mode ablation:** On a 7,777-sample subset, ERG-style CoT generally beat both direct answering and a generic structured template. The structured template did not help the 4B model.
- **Data complexity:** Constraints typically involve 3–7 nodes. Training on harder groups (Levels 3–5) gave larger in-domain gains, and adding complex samples to easy data still helped.
- **Limitations:**
  - No models beyond 32B were tested.
  - Seed constraints rely on pre-collected knowledge, which limits diversity.
  - GRPO was not compared against PPO, DPO, or other optimization methods.

*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

- [Beihang University](https://md.research.elamur.ai/free/en/universities/beihang-university/) — University, Beijing, China · 86 works · 9,097 citations

Also on the byline: [State Key Laboratory of Virtual Reality Technology and Systems](https://ror.org/0009eea46).

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works

🧭 Topics (1): [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,685 works

## Cite

Cite as: Yuancheng Yang, Lin Yang, Xu Wang, Chao Tong and Haihua Yang. 2026. ImpRIF: Stronger Implicit Reasoning Leads to Better Complex Instruction Following. 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{yang2026imprif,
  title = {{ImpRIF: Stronger Implicit Reasoning Leads to Better Complex Instruction Following}},
  author = {Yuancheng Yang, Lin Yang, Xu Wang, Chao Tong and Haihua Yang},
  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.1796},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1796}
}
```

- [← previous Importance of Prompt Optimisation for Error Detection in Medical Notes Using Language Models](https://md.research.elamur.ai/free/en/papers/myles-2026-importance-prompt-optimisation-error-detection/)
- [next → Improved Policy Optimization for Mixture-of-Experts Models: Importance Sampling and Rewarding from an Expert-Centric Perspective](https://md.research.elamur.ai/free/en/papers/qian-2026-improved-policy-optimization-mixture-experts/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/37/)

## References

48 references, 19 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. DeepSeek-AI, Aixin Liu, Aoxue Mei, Bangcai Lin, Bing Xue, Bingxuan Wang, Bingzheng Xu, Bochao Wu, Bowei Zhang, Chaofan Lin, Chen Dong, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenhao Xu, Chong Ruan, Damai Dai, Daya Guo, Dejian Yang, and 245 others. 2025. [DeepSeek-V3.2: Pushing the Frontier of Open…](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-v3-2-pushing/)
2. Kaustubh Deshpande, Ved Sirdeshmukh, Johannes Baptist Mols, Lifeng Jin, Ed-Yeremai Hernandez-Cardona, Dean Lee, Jeremy Kritz, Willow E. Primack, Summer Yue, and Chen Xing. 2025. MultiChallenge: A Realistic Multi-Turn Conversation Evaluation Benchmark Challenging to Frontier LLMs. In Findings of the… [find](https://scholar.google.com/scholar?q=Kaustubh+Deshpande%2C+Ved+Sirdeshmukh%2C+Johannes+Baptist+Mols%2C+Lifeng+Jin%2C+Ed%2DYeremai+Hernandez%2DCardona%2C+Dean+Lee%2C+Jeremy+Kritz%2C+Willow+E%2E+Primack%2C+Summer+Yue%2C+and+Chen+Xing%2E+2025%2E+MultiChallenge%3A+A+Real)
3. Shengyuan Ding, Shenxi Wu, Xiangyu Zhao, Yuhang Zang, Haodong Duan, Xiaoyi Dong, Pan Zhang, Yuhang Cao, Dahua Lin, and Jiaqi Wang. 2025. [find](https://scholar.google.com/scholar?q=Shengyuan+Ding%2C+Shenxi+Wu%2C+Xiangyu+Zhao%2C+Yuhang+Zang%2C+Haodong+Duan%2C+Xiaoyi+Dong%2C+Pan+Zhang%2C+Yuhang+Cao%2C+Dahua+Lin%2C+and+Jiaqi+Wang%2E+2025%2E)
4. MM-IFEngine: Towards Multimodal Instruction Following. In ICCV 2025. arXiv. [find](https://scholar.google.com/scholar?q=MM%2DIFEngine%3A+Towards+Multimodal+Instruction+Following%2E+In+ICCV+2025%2E+arXiv%2E)
5. Guanting Dong, Keming Lu, Chengpeng Li, Tingyu Xia, Bowen Yu, Chang Zhou, and Jingren Zhou. 2024. Self-play with Execution Feedback: Improving Instruction-following Capabilities of Large Language Models. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Guanting+Dong%2C+Keming+Lu%2C+Chengpeng+Li%2C+Tingyu+Xia%2C+Bowen+Yu%2C+Chang+Zhou%2C+and+Jingren+Zhou%2E+2024%2E+Self%2Dplay+with+Execution+Feedback%3A+Improving+Instruction%2Dfollowing+Capabilities+of+Large+Language+Mode)
6. Tingchen Fu, Jiawei Gu, Yafu Li, Xiaoye Qu, and Yu Cheng. 2025. [Scaling Reasoning, Losing Control: Evaluating Instruction Following in Large Reasoning Models](https://md.research.elamur.ai/free/en/papers/fu-2026-scaling-reasoning-losing-control-evaluating/). Preprint, arXiv:2505.14810.
7. Google. 2025a. Gemini 2.5 Pro. [find](https://scholar.google.com/scholar?q=Google%2E+2025a%2E+Gemini+2%2E5+Pro%2E)
8. Google. 2025b. Gemini 3 Pro. [find](https://scholar.google.com/scholar?q=Google%2E+2025b%2E+Gemini+3+Pro%2E)
9. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, Archi Mitra, Archie Sravankumar, Artem Korenev, Arthur Hinsvark, and 542 o… — [The Llama 3 Herd of Models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/)
10. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, and 175 others. 2025a. DeepSeek-R1 incentivizes reasoning in LLMs through rei… [find](https://scholar.google.com/scholar?q=Daya+Guo%2C+Dejian+Yang%2C+Haowei+Zhang%2C+Junxiao+Song%2C+Peiyi+Wang%2C+Qihao+Zhu%2C+Runxin+Xu%2C+Ruoyu+Zhang%2C+Shirong+Ma%2C+Xiao+Bi%2C+Xiaokang+Zhang%2C+Xingkai+Yu%2C+Yu+Wu%2C+Z%2E+F%2E+Wu%2C+Zhibin+Gou%2C+Zhihong+Shao%2C+Zhuoshu+Li)
11. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, and 175 others. 2025b. DeepSeek-R1 incentivizes reasoning in LLMs through rei… [find](https://scholar.google.com/scholar?q=Daya+Guo%2C+Dejian+Yang%2C+Haowei+Zhang%2C+Junxiao+Song%2C+Peiyi+Wang%2C+Qihao+Zhu%2C+Runxin+Xu%2C+Ruoyu+Zhang%2C+Shirong+Ma%2C+Xiao+Bi%2C+Xiaokang+Zhang%2C+Xingkai+Yu%2C+Yu+Wu%2C+Z%2E+F%2E+Wu%2C+Zhibin+Gou%2C+Zhihong+Shao%2C+Zhuoshu+Li)
12. Haoyu Han, Yaochen Xie, Hui Liu, Xianfeng Tang, Sreyashi Nag, William Headden, Yang Li, Chen Luo, Shuiwang Ji, Qi He, and Jiliang Tang. 2025. Reasoning with Graphs: Structuring Implicit Knowledge to Enhance LLMs Reasoning. In Findings of the Association for Computational Linguistics: ACL 2025, pages… [find](https://scholar.google.com/scholar?q=Haoyu+Han%2C+Yaochen+Xie%2C+Hui+Liu%2C+Xianfeng+Tang%2C+Sreyashi+Nag%2C+William+Headden%2C+Yang+Li%2C+Chen+Luo%2C+Shuiwang+Ji%2C+Qi+He%2C+and+Jiliang+Tang%2E+2025%2E+Reasoning+with+Graphs%3A+Structuring+Implicit+Knowledge+to+E)
13. Qianyu He, Jie Zeng, Qianxi He, Jiaqing Liang, and Yanghua Xiao. 2024a. From Complex to Simple: Enhancing Multi-Constraint Complex Instruction Following Ability of Large Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 10864–10882, Miami, Florida, USA.… [find](https://scholar.google.com/scholar?q=Qianyu+He%2C+Jie+Zeng%2C+Qianxi+He%2C+Jiaqing+Liang%2C+and+Yanghua+Xiao%2E+2024a%2E+From+Complex+to+Simple%3A+Enhancing+Multi%2DConstraint+Complex+Instruction+Following+Ability+of+Large+Language+Models%2E+In+Findings+o)
14. Qianyu He, Jie Zeng, Wenhao Huang, Lina Chen, Jin Xiao, Qianxi He, Xunzhe Zhou, Jiaqing Liang, and Yanghua Xiao. 2024b. Can Large Language Models Understand Real-World Complex Instructions? Proceedings of the AAAI Conference on Artificial Intelligence, 38(16):18188–18196. [find](https://scholar.google.com/scholar?q=Qianyu+He%2C+Jie+Zeng%2C+Wenhao+Huang%2C+Lina+Chen%2C+Jin+Xiao%2C+Qianxi+He%2C+Xunzhe+Zhou%2C+Jiaqing+Liang%2C+and+Yanghua+Xiao%2E+2024b%2E+Can+Large+Language+Models+Understand+Real%2DWorld+Complex+Instructions%3F+Proceeding)
15. Hui Huang, Jiaheng Liu, Yancheng He, Shilong Li, Bing Xu, Conghui Zhu, Muyun Yang, and Tiejun Zhao. 2025a. [MuSC: Improving Complex Instruction Following with Multi-granularity Self-Contrastive Training](https://md.research.elamur.ai/free/en/papers/huang-2025-musc-improving-complex-instruction-following/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volum…
16. Kaixuan Huang, Jiacheng Guo, Zihao Li, Xiang Ji, Jiawei Ge, Wenzhe Li, Yingqing Guo, Tianle Cai, Hui Yuan, Runzhe Wang, Yue Wu, Ming Yin, Shange Tang, Yangsibo Huang, Chi Jin, Xinyun Chen, Chiyuan Zhang, and Mengdi Wang. 2025b. MATH-perturb: Benchmarking LLMs’ math reasoning abilities against hard p… [find](https://scholar.google.com/scholar?q=Kaixuan+Huang%2C+Jiacheng+Guo%2C+Zihao+Li%2C+Xiang+Ji%2C+Jiawei+Ge%2C+Wenzhe+Li%2C+Yingqing+Guo%2C+Tianle+Cai%2C+Hui+Yuan%2C+Runzhe+Wang%2C+Yue+Wu%2C+Ming+Yin%2C+Shange+Tang%2C+Yangsibo+Huang%2C+Chi+Jin%2C+Xinyun+Chen%2C+Chiyuan+Zha)
17. Xiang Huang, Ting-En Lin, Feiteng Fang, Yuchuan Wu, Hangyu Li, Yuzhong Qu, Fei Huang, and Yongbin Li. 2025c. Reverse Preference Optimization for Complex Instruction Following. In Findings of the Association for Computational Linguistics: ACL 2025, pages 20700–20723, Vienna, Austria. Association for… [find](https://scholar.google.com/scholar?q=Xiang+Huang%2C+Ting%2DEn+Lin%2C+Feiteng+Fang%2C+Yuchuan+Wu%2C+Hangyu+Li%2C+Yuzhong+Qu%2C+Fei+Huang%2C+and+Yongbin+Li%2E+2025c%2E+Reverse+Preference+Optimization+for+Complex+Instruction+Following%2E+In+Findings+of+the+Assoc)
18. Yuxin Jiang, Yufei Wang, Xingshan Zeng, Wanjun Zhong, Liangyou Li, Fei Mi, Lifeng Shang, Xin Jiang, Qun Liu, and Wei Wang. 2024. [FollowBench: A Multi-level Fine-grained Constraints Following Benchmark for Large Language Models](https://md.research.elamur.ai/free/en/papers/jiang-2024-followbench-multi-level-fine-grained/). In Proceedings of the 62nd Annual Meeting of the Association for Computa…
19. Xiaomin Li, Zhou Yu, Zhiwei Zhang, Xupeng Chen, Ziji Zhang, Yingying Zhuang, Narayanan Sadagopan, and Anurag Beniwal. 2025. When Thinking Fails: The Pitfalls of Reasoning for Instruction-Following in LLMs. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Xiaomin+Li%2C+Zhou+Yu%2C+Zhiwei+Zhang%2C+Xupeng+Chen%2C+Ziji+Zhang%2C+Yingying+Zhuang%2C+Narayanan+Sadagopan%2C+and+Anurag+Beniwal%2E+2025%2E+When+Thinking+Fails%3A+The+Pitfalls+of+Reasoning+for+Instruction%2DFollowing+in+)
20. Mistral. 2025. Ministral 3. [find](https://scholar.google.com/scholar?q=Mistral%2E+2025%2E+Ministral+3%2E)
21. OpenAI. 2025a. ChatGPT (GPT-4.1). [find](https://scholar.google.com/scholar?q=OpenAI%2E+2025a%2E+ChatGPT+%28GPT%2D4%2E1%29%2E)
22. OpenAI. 2025b. ChatGPT (GPT-5). [find](https://scholar.google.com/scholar?q=OpenAI%2E+2025b%2E+ChatGPT+%28GPT%2D5%29%2E)
23. OpenAI. 2025c. ChatGPT (GPT-5.2). [find](https://scholar.google.com/scholar?q=OpenAI%2E+2025c%2E+ChatGPT+%28GPT%2D5%2E2%29%2E)
24. Mihir Parmar, Nisarg Patel, Neeraj Varshney, Mutsumi Nakamura, Man Luo, Santosh Mashetty, Arindam Mitra, and Chitta Baral. 2024. [LogicBench: Towards Systematic Evaluation of Logical Reasoning Ability of Large Language Models](https://md.research.elamur.ai/free/en/papers/parmar-2024-logicbench-systematic-evaluation-logical-reasoning/). In Proceedings of the 62nd Annual Meeting of the Association for Computati…
25. Chau Minh Pham, Simeng Sun, and Mohit Iyyer. 2024. [Suri: Multi-constraint Instruction Following in Long-form Text Generation](https://md.research.elamur.ai/free/en/papers/pham-2024-suri-multi-constraint-instruction-following/). In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 1722–1753, Miami, Florida, USA. Association for Computational Linguistics.
26. Yiwei Qin, Kaiqiang Song, Yebowen Hu, Wenlin Yao, Sangwoo Cho, Xiaoyang Wang, Xuansheng Wu, Fei Liu, Pengfei Liu, and Dong Yu. 2024. [InFoBench: Evaluating Instruction Following Ability in Large Language Models](https://md.research.elamur.ai/free/en/papers/qin-2024-infobench-evaluating-instruction-following-ability/). In Findings of the Association for Computational Linguistics: ACL 2024, pages 13025–13048…
27. Yulei Qin, Gang Li, Zongyi Li, Zihan Xu, Yuchen Shi, Zhekai Lin, Xiao Cui, Ke Li, and Xing Sun. 2025. Incentivizing Reasoning for Advanced Instruction-Following of Large Language Models. Preprint, arXiv:2506.01413. — [arXiv:2506.01413](https://arxiv.org/abs/2506.01413) [find](https://scholar.google.com/scholar?q=Yulei+Qin%2C+Gang+Li%2C+Zongyi+Li%2C+Zihan+Xu%2C+Yuchen+Shi%2C+Zhekai+Lin%2C+Xiao+Cui%2C+Ke+Li%2C+and+Xing+Sun%2E+2025%2E+Incentivizing+Reasoning+for+Advanced+Instruction%2DFollowing+of+Large+Language+Models%2E+Preprint%2C+arX)
28. Qwen, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, and 24 others. 2025. Qwen2.5 Technical Report. Preprint, arXiv:2412.15… [find](https://scholar.google.com/scholar?q=Qwen%2C+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+Huan+Lin%2C+Jian+Yang%2C+Jianhong+Tu%2C+Jianwei+Zhang%2C+Jianxin+Yang%2C+Jiaxi+Yan)
29. Qingyu Ren, Jie Zeng, Qianyu He, Jiaqing Liang, Yanghua Xiao, Weikang Zhou, Zeye Sun, and Fei Yu. 2025. Step-by-Step Mastery: Enhancing Soft Constraint Following Ability of Large Language Models. In Findings of the Association for Computational Linguistics: ACL 2025, pages 19581–19596, Vienna, Austr… [find](https://scholar.google.com/scholar?q=Qingyu+Ren%2C+Jie+Zeng%2C+Qianyu+He%2C+Jiaqing+Liang%2C+Yanghua+Xiao%2C+Weikang+Zhou%2C+Zeye+Sun%2C+and+Fei+Yu%2E+2025%2E+Step%2Dby%2DStep+Mastery%3A+Enhancing+Soft+Constraint+Following+Ability+of+Large+Language+Models%2E+In+F)
30. Zhihong Shao, Yuxiang Luo, Chengda Lu, Z. Z. Ren, Jiewen Hu, Tian Ye, Zhibin Gou, Shirong Ma, and Xiaokang Zhang. 2025. [DeepSeekMath-V2: Towards Self-Verifiable Mathematical Reasoning](https://md.research.elamur.ai/free/en/papers/shao-2025-deepseekmath-v2-self-verifiable/). Preprint, arXiv:2511.22570.
31. Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru Zhang, Yanghua Peng, Haibin Lin, and Chuan Wu. 2025. [HybridFlow: A flexible and efficient RLHF framework](https://md.research.elamur.ai/free/en/papers/sheng-2024-hybridflow-flexible-efficient-rlhf-framework/). In Proceedings of the Twentieth European Conference on Computer Systems, EuroSys ’25, pages 1279–1297, New York, NY, USA. Asso…
32. Tingyu Song, Guo Gan, Mingsheng Shang, and Yilun Zhao. 2025. IFIR: A Comprehensive Benchmark for Evaluating Instruction-Following in Expert-Domain Information Retrieval. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Hum… [find](https://scholar.google.com/scholar?q=Tingyu+Song%2C+Guo+Gan%2C+Mingsheng+Shang%2C+and+Yilun+Zhao%2E+2025%2E+IFIR%3A+A+Comprehensive+Benchmark+for+Evaluating+Instruction%2DFollowing+in+Expert%2DDomain+Information+Retrieval%2E+In+Proceedings+of+the+2025+Con)
33. Haoran Sun, Lixin Liu, Junjie Li, Fengyu Wang, Baohua Dong, Ran Lin, and Ruohui Huang. 2024. [Conifer: Improving Complex Constrained Instruction-Following Ability of Large Language Models](https://md.research.elamur.ai/free/en/papers/sun-2024-conifer-improving-complex-constrained-instruction/). Preprint, arXiv:2404.02823.
34. Kimi Team, Yifan Bai, Yiping Bao, Guanduo Chen, Jiahao Chen, Ningxin Chen, Ruijue Chen, Yanru Chen, Yuankun Chen, Yutian Chen, Zhuofu Chen, Jialei Cui, Hao Ding, Mengnan Dong, Angang Du, Chenzhuang Du, Dikang Du, Yulun Du, Yu Fan, and 150 others. 2025. [Kimi K2: Open Agentic Intelligence](https://md.research.elamur.ai/free/en/papers/kimi-team-2025-kimi-k2-open-agentic/). Preprint, a…
35. Chenyang Wang, Liang Wen, Shousheng Jia, Xiangzheng Zhang, and Liang Xu. 2025. Light-IF: Endowing LLMs with Generalizable Reasoning via Preview and Self-Checking for Complex Instruction Following. Preprint, arXiv:2508.03178. — [arXiv:2508.03178](https://arxiv.org/abs/2508.03178) [find](https://scholar.google.com/scholar?q=Chenyang+Wang%2C+Liang+Wen%2C+Shousheng+Jia%2C+Xiangzheng+Zhang%2C+and+Liang+Xu%2E+2025%2E+Light%2DIF%3A+Endowing+LLMs+with+Generalizable+Reasoning+via+Preview+and+Self%2DChecking+for+Complex+Instruction+Following%2E+Pre)
36. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. 2023. [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/). Preprint, arXiv:2201.11903.
37. Bosi Wen, Pei Ke, Xiaotao Gu, Lindong Wu, Hao Huang, Jinfeng Zhou, Wenchuang Li, Binxin Hu, Wendy Gao, Jiaxin Xu, Yiming Liu, Jie Tang, Hongning Wang, and Minlie Huang. 2025. Benchmarking complex instruction-following with multiple constraints composition. In Proceedings of the 38th International Co… [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+Yiming+Liu%2C+Jie+Tang%2C+Hongning+Wang%2C+and+Minlie+Huang%2E+2025%2E+Benchmarking+complex+inst)
38. Tianhao Wu, Janice Lan, Weizhe Yuan, Jiantao Jiao, Jason E. Weston, and Sainbayar Sukhbaatar. 2025a. Thinking LLMs: General Instruction Following with Thought Generation. In Forty-Second International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=Tianhao+Wu%2C+Janice+Lan%2C+Weizhe+Yuan%2C+Jiantao+Jiao%2C+Jason+E%2E+Weston%2C+and+Sainbayar+Sukhbaatar%2E+2025a%2E+Thinking+LLMs%3A+General+Instruction+Following+with+Thought+Generation%2E+In+Forty%2DSecond+International)
39. Xiaodong Wu, Minhao Wang, Yichen Liu, Xiaoming Shi, He Yan, Lu Xiangju, Junmin Zhu, and Wei Zhang. 2025b. [LIFBench: Evaluating the Instruction Following Performance and Stability of Large Language Models in Long-Context Scenarios](https://md.research.elamur.ai/free/en/papers/wu-2025-lifbench-evaluating-instruction-following-performance/). In Proceedings of the 63rd Annual Meeting of the Association for Comp…
40. Congying Xia, Chen Xing, Jiangshu Du, Xinyi Yang, Yihao Feng, Ran Xu, Wenpeng Yin, and Caiming Xiong. 2024. FOFO: A Benchmark to Evaluate LLMs’ Format-Following Capability. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 680–6… [find](https://scholar.google.com/scholar?q=Congying+Xia%2C+Chen+Xing%2C+Jiangshu+Du%2C+Xinyi+Yang%2C+Yihao+Feng%2C+Ran+Xu%2C+Wenpeng+Yin%2C+and+Caiming+Xiong%2E+2024%2E+FOFO%3A+A+Benchmark+to+Evaluate+LLMs%E2%80%99+Format%2DFollowing+Capability%2E+In+Proceedings+of+the+62nd+)
41. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, Chujie Zheng, Dayiheng Liu, Fan Zhou, Fei Huang, Feng Hu, Hao Ge, Haoran Wei, Huan Lin, Jialong Tang, and 41 others. 2025. [Qwen3 Technical Report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). Preprint, arXiv:2505.09388.
42. Lin Yang, Yuancheng Yang, Xu Wang, Changkun Liu, and Haihua Yang. 2026. MedMT-Bench: Can LLMs Memorize and Understand Long Multi-Turn Conversations in Medical Scenarios? Preprint, arXiv:2603.23519. — [arXiv:2603.23519](https://arxiv.org/abs/2603.23519) [find](https://scholar.google.com/scholar?q=Lin+Yang%2C+Yuancheng+Yang%2C+Xu+Wang%2C+Changkun+Liu%2C+and+Haihua+Yang%2E+2026%2E+MedMT%2DBench%3A+Can+LLMs+Memorize+and+Understand+Long+Multi%2DTurn+Conversations+in+Medical+Scenarios%3F+Preprint%2C+arXiv%3A2603%2E23519%2E+%E2%80%94+)
43. Junjie Ye, Caishuang Huang, Zhuohan Chen, Wenjie Fu, Chenyuan Yang, Leyi Yang, Yilong Wu, Peng Wang, Meng Zhou, Xiaolong Yang, Tao Gui, Qi Zhang, Zhongchao Shi, Jianping Fan, and Xuanjing Huang. 2025. [A Multi-Dimensional Constraint Framework for Evaluating and Improving Instruction Following in Larg…](https://md.research.elamur.ai/free/en/papers/ye-2026-muldimif-multi-dimensional-constraint-framework/)
44. Tao Zhang, ChengLIn Zhu, Yanjun Shen, Wenjing Luo, Yan Zhang, Hao Liang, Tao Zhang, Fan Yang, Mingan Lin, Yujing Qiao, Weipeng Chen, Bin Cui, Wentao Zhang, and Zenan Zhou. 2025a. [CFBench: A Comprehensive Constraints-Following Benchmark for LLMs](https://md.research.elamur.ai/free/en/papers/zhang-2025-cfbench-comprehensive-constraints-following-benchmark/). In Proceedings of the 63rd Annual Meeting of the Assoc…
45. Xinghua Zhang, Haiyang Yu, Cheng Fu, Fei Huang, and Yongbin Li. 2025b. [IOPO: Empowering LLMs with Complex Instruction Following via Input-Output Preference Optimization](https://md.research.elamur.ai/free/en/papers/zhang-2025-iopo-empowering-llms-complex-instruction/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 22185– 2…
46. Yifan Zhang and Team Math-AI. 2025. American invitational mathematics examination (AIME) 2025. [find](https://scholar.google.com/scholar?q=Yifan+Zhang+and+Team+Math%2DAI%2E+2025%2E+American+invitational+mathematics+examination+%28AIME%29+2025%2E)
47. Yaowei Zheng, Richong Zhang, Junhao Zhang, Yanhan Ye, Zheyan Luo, Zhangchi Feng, and Yongqiang Ma. 2024. [LlamaFactory: Unified efficient fine-tuning of 100+ language models](https://md.research.elamur.ai/free/en/papers/zheng-2024-llamafactory-unified-efficient-fine-tuning/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations),…
48. Tao Zou, Xinghua Zhang, Haiyang Yu, Minzheng Wang, Fei Huang, and Yongbin Li. 2025. EIF-BENCH: Extremely Complex Instruction Following Benchmark for Large Language Models. Preprint, arXiv:2506.08375. — [arXiv:2506.08375](https://arxiv.org/abs/2506.08375) [find](https://scholar.google.com/scholar?q=Tao+Zou%2C+Xinghua+Zhang%2C+Haiyang+Yu%2C+Minzheng+Wang%2C+Fei+Huang%2C+and+Yongbin+Li%2E+2025%2E+EIF%2DBENCH%3A+Extremely+Complex+Instruction+Following+Benchmark+for+Large+Language+Models%2E+Preprint%2C+arXiv%3A2506%2E08375%2E+)

*Updated 2 November 2026*