---
title: "Data Mixing Agent — paper by Yang et al. (2026)"
description: "The authors propose Data Mixing Agent, a small learned policy that sets domain proportions step by… 48 references, cited by 1 paper here."
authors: ["Kailai Yang", "Xiao Liu", "Lei Ji", "Hao Li", "Xiao Liang", "Zhiwei Liu", "Yeyun Gong", "Peng Cheng", "Mao 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.427/"
doi: "10.18653/v1/2026.acl-long.427"
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-data-mixing-agent-learning-re/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/yang-2026-data-mixing-agent-learning-re/)

[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/11/) / Data Mixing Agent: Learning to Re-weight Domains for Continual Pre-training

# Data Mixing Agent: Learning to Re-weight Domains for Continual Pre-training

**Authors:** [Kailai Yang](https://md.research.elamur.ai/free/en/authors/kailai-yang/), [Xiao Liu](https://md.research.elamur.ai/free/en/authors/xiao-liu/), [Lei Ji](https://md.research.elamur.ai/free/en/authors/lei-ji/), [Hao Li](https://md.research.elamur.ai/free/en/authors/hao-li/), [Xiao Liang](https://md.research.elamur.ai/free/en/authors/xiao-liang/), [Zhiwei Liu](https://md.research.elamur.ai/free/en/authors/zhiwei-liu/), [Yeyun Gong](https://md.research.elamur.ai/free/en/authors/yeyun-gong/), [Peng Cheng](https://md.research.elamur.ai/free/en/authors/peng-cheng/) and [Mao Yang](https://md.research.elamur.ai/free/en/authors/mao-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.427/) · [PDF](https://aclanthology.org/2026.acl-long.427.pdf) · [DOI 10.18653/v1/2026.acl-long.427](https://doi.org/10.18653/v1/2026.acl-long.427) · [OpenAlex](https://openalex.org/W4417495159)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [1](https://md.research.elamur.ai/free/en/papers/yang-2026-data-mixing-agent-learning-re/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
Continual pre-training of an LLM on a narrow target field, such as math, often causes catastrophic forgetting of general skills. Mixing source and target data by domain helps, but existing reweighting schemes depend on hand-crafted heuristics. The authors propose Data Mixing Agent, a small learned policy that sets domain proportions step by step during training. Domain reweighting is framed as a Markov Decision Process. Domains come from the Nvidia domain classifier, which assigns each field's data to 26 categories. The training data for the agent is built as follows:

- Hundreds of random mixing trajectories over DCLM (general data) and the Dolmino-mix-1124 math split are sampled.
- Small proxy models are trained on each trajectory.
- Each checkpoint is scored in a cheap evaluation environment using per-token log probability.

The agent is then trained on these trajectories and scores with Conservative Q-Learning (CQL). Compared with static and dynamic reweighting baselines, it gives better balance between general and math benchmarks. It also transfers to new source data, models, domain spaces, and code generation without retraining. This suggests data-mixing heuristics can be learned once and reused.

## Key points
- **Agent training data:** 384 sampled trajectories, generated by an inductive scoring procedure that favours smooth, diverse, target-leaning distributions.
- **Main result:** an average gain of 3.02% across 8 general and 4 math benchmarks over RegMix and a dynamic baseline (Xia et al., 2023).
- **Generalization without retraining:** the agent transfers to 3 unseen source fields, 4 target models, 2 domain spaces, and the code generation field.
- **Data efficiency:** with a 21B-token budget, DataAgent RL stopped early at 19.92B tokens and DataAgent SFT at 18.86B, saving up to 2.14B tokens while still beating RegMix.
- **Start-state estimation:** when pre-training data is unavailable, the start state can be estimated from a model's own generated text. This was validated on five Pythia models via KL divergence.
- **Stated limitation:** the agent may perform poorly when the target field differs strongly from the source field (e.g., medical).

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 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

## Cite

Cite as: Kailai Yang, Xiao Liu, Lei Ji, Hao Li, Xiao Liang, Zhiwei Liu, Yeyun Gong, Peng Cheng and Mao Yang. 2026. Data Mixing Agent: Learning to Re-weight Domains for Continual Pre-training. 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{yang2026data,
  title = {{Data Mixing Agent: Learning to Re-weight Domains for Continual Pre-training}},
  author = {Kailai Yang, Xiao Liu, Lei Ji, Hao Li, Xiao Liang, Zhiwei Liu, Yeyun Gong, Peng Cheng and Mao 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.427},
  url = {https://doi.org/10.18653/v1/2026.acl-long.427}
}
```

- [← previous Data Asgardians at BEA 2026 Shared Task 1: A Hybrid Transformer–Feature Ensemble for L1-Aware English Vocabulary Difficulty Prediction](https://md.research.elamur.ai/free/en/papers/pineda-2026-data-asgardians-bea-2026-shared/)
- [next → DataBees at SemEval-2026 Task 9: Detecting Multilingual, Multicultural and Multievent Online Polarization](https://md.research.elamur.ai/free/en/papers/sriram-2026-databees-semeval-2026-task-9/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/11/)

## References

48 references, 25 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Leo Gao, Stella Biderman, Sid Black, Laurence Golding, Travis Hoppe, Charles Foster, Jason Phang, Horace He, Anish Thite, Noa Nabeshima, et al. 2020. The pile: An 800gb dataset of diverse text for language modeling. arXiv preprint arXiv:2101.00027. [find](https://scholar.google.com/scholar?q=Leo+Gao%2C+Stella+Biderman%2C+Sid+Black%2C+Laurence+Golding%2C+Travis+Hoppe%2C+Charles+Foster%2C+Jason+Phang%2C+Horace+He%2C+Anish+Thite%2C+Noa+Nabeshima%2C+et+al%2E+2020%2E+The+pile%3A+An+800gb+dataset+of+diverse+text+for+lan)
2. Aida Amini, Saadia Gabriel, Peter Lin, Rik Koncel-Kedziorski, Yejin Choi, and Hannaneh Hajishirzi. 2019. [Mathqa: Towards interpretable math word problem solving with operation-based formalisms](https://md.research.elamur.ai/free/en/papers/amini-2019-mathqa-interpretable-math-word-problem/). arXiv preprint arXiv:1905.13319.
3. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, et al. 2024. [The llama 3 herd of models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/). arXiv e-prints, pages arXiv–2407.
4. Jacob Austin, Augustus Odena, Maxwell Nye, Maarten Bosma, Henryk Michalewski, David Dohan, Ellen Jiang, Carrie Cai, Michael Terry, Quoc Le, et al. 2021. [Program synthesis with large language models](https://md.research.elamur.ai/free/en/papers/austin-2021-program-synthesis-large-language-models/). arXiv preprint arXiv:2108.07732.
5. Daya Guo, Qihao Zhu, Dejian Yang, Zhenda Xie, Kai Dong, Wentao Zhang, Guanting Chen, Xiao Bi, Yu Wu, YK Li, et al. 2024. Deepseek-coder: When the large language model meets programming– the rise of code intelligence. arXiv preprint arXiv:2401.14196. [find](https://scholar.google.com/scholar?q=Daya+Guo%2C+Qihao+Zhu%2C+Dejian+Yang%2C+Zhenda+Xie%2C+Kai+Dong%2C+Wentao+Zhang%2C+Guanting+Chen%2C+Xiao+Bi%2C+Yu+Wu%2C+YK+Li%2C+et+al%2E+2024%2E+Deepseek%2Dcoder%3A+When+the+large+language+model+meets+programming%E2%80%93+the+rise+of+co)
6. Stella Biderman, Hailey Schoelkopf, Quentin Gregory Anthony, Herbie Bradley, Kyle O’Brien, Eric Hallahan, Mohammad Aflah Khan, Shivanshu Purohit, USVSN Sai Prashanth, Edward Raff, et al. 2023. Pythia: A suite for analyzing large language models across training and scaling. In International Conferenc… [find](https://scholar.google.com/scholar?q=Stella+Biderman%2C+Hailey+Schoelkopf%2C+Quentin+Gregory+Anthony%2C+Herbie+Bradley%2C+Kyle+O%E2%80%99Brien%2C+Eric+Hallahan%2C+Mohammad+Aflah+Khan%2C+Shivanshu+Purohit%2C+USVSN+Sai+Prashanth%2C+Edward+Raff%2C+et+al%2E+2023%2E+Pythia%3A)
7. Yonatan Bisk, Rowan Zellers, Jianfeng Gao, Yejin Choi, et al. 2020. [Piqa: Reasoning about physical commonsense in natural language](https://md.research.elamur.ai/free/en/papers/bisk-2019-piqa-reasoning-about-physical-commonsense/). In Proceedings of the AAAI conference on artificial intelligence, volume 34, pages 7432–7439.
8. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde De Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, et al. 2021. [Evaluating large language models trained on code](https://md.research.elamur.ai/free/en/papers/chen-2021-evaluating-large-language-models-trained/). arXiv preprint arXiv:2107.03374.
9. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2020. [Measuring massive multitask language understanding](https://md.research.elamur.ai/free/en/papers/hendrycks-2020-measuring-massive-multitask-language-understanding/). arXiv preprint arXiv:2009.03300.
10. 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.
11. Jordan Hoffmann, Sebastian Borgeaud, Arthur Mensch, Elena Buchatskaya, Trevor Cai, Eliza Rutherford, Diego de Las Casas, Lisa Anne Hendricks, Johannes Welbl, Aidan Clark, et al. 2022. [Training compute-optimal large language models](https://md.research.elamur.ai/free/en/papers/hoffmann-2022-training-compute-optimal-large-language/). arXiv preprint arXiv:2203.15556.
12. Jian Hu, Xibin Wu, Zilin Zhu, Weixun Wang, Dehao Zhang, Yu Cao, et al. 2024. [Openrlhf: An easy-to-use, scalable and high-performance rlhf framework](https://md.research.elamur.ai/free/en/papers/hu-2024-openrlhf-easy-use-scalable/). arXiv preprint arXiv:2405.11143.
13. Binyuan Hui, Jian Yang, Zeyu Cui, Jiaxi Yang, Dayiheng Liu, Lei Zhang, Tianyu Liu, Jiajun Zhang, Bowen Yu, Keming Lu, et al. 2024. Qwen2. 5-coder technical report. arXiv preprint arXiv:2409.12186. — [Qwen2.5-Coder Technical Report](https://md.research.elamur.ai/free/en/papers/hui-2024-qwen2-5-coder-technical-report/)
14. Hamish Ivison, Yizhong Wang, Jiacheng Liu, Zeqiu Wu, Valentina Pyatkin, Nathan Lambert, Noah A Smith, Yejin Choi, and Hanna Hajishirzi. 2024. [Unpacking dpo and ppo: Disentangling best practices for learning from preference feedback](https://md.research.elamur.ai/free/en/papers/ivison-2024-unpacking-dpo-ppo-disentangling-best/). Advances in neural information processing systems, 37:36602–36633.
15. Aviral Kumar, Aurick Zhou, George Tucker, and Sergey Levine. 2020. Conservative q-learning for offline reinforcement learning. Advances in neural information processing systems, 33:1179–1191. [find](https://scholar.google.com/scholar?q=Aviral+Kumar%2C+Aurick+Zhou%2C+George+Tucker%2C+and+Sergey+Levine%2E+2020%2E+Conservative+q%2Dlearning+for+offline+reinforcement+learning%2E+Advances+in+neural+information+processing+systems%2C+33%3A1179%E2%80%931191%2E)
16. Aitor Lewkowycz, Anders Andreassen, David Dohan, Ethan Dyer, Henryk Michalewski, Vinay Ramasesh, Ambrose Slone, Cem Anil, Imanol Schlag, Theo Gutman-Solo, et al. 2022. [Solving quantitative reasoning problems with language models](https://md.research.elamur.ai/free/en/papers/lewkowycz-2022-solving-quantitative-reasoning-problems-language/). Advances in Neural Information Processing Systems, 35:3843– 3857.
17. Hao Li, Bowen Deng, Chang Xu, Zhiyuan Feng, Viktor Schlegel, Yu-Hao Huang, Yizheng Sun, Jingyuan Sun, Kailai Yang, Yiyao Yu, et al. 2025. Mira: Medical time series foundation model for real-world health data. arXiv preprint arXiv:2506.07584. — [arXiv:2506.07584](https://arxiv.org/abs/2506.07584) [find](https://scholar.google.com/scholar?q=Hao+Li%2C+Bowen+Deng%2C+Chang+Xu%2C+Zhiyuan+Feng%2C+Viktor+Schlegel%2C+Yu%2DHao+Huang%2C+Yizheng+Sun%2C+Jingyuan+Sun%2C+Kailai+Yang%2C+Yiyao+Yu%2C+et+al%2E+2025%2E+Mira%3A+Medical+time+series+foundation+model+for+real%2Dworld+heal)
18. Jeffrey Li, Alex Fang, Georgios Smyrnis, Maor Ivgi, Matt Jordan, Samir Yitzhak Gadre, Hritik Bansal, Etash Guha, Sedrick Scott Keh, Kushal Arora, et al. 2024. [Datacomp-lm: In search of the next generation of training sets for language models](https://md.research.elamur.ai/free/en/papers/li-2024-datacomp-lm-search-next-generation/). Advances in Neural Information Processing Systems, 37:142…
19. Zhenghao Lin, Zihao Tang, Xiao Liu, Yeyun Gong, Yi Cheng, Qi Chen, Hang Li, Ying Xin, Ziyue Yang, Kailai Yang, et al. 2025. Sigma: Differential rescaling of query, key and value for efficient language models. arXiv preprint arXiv:2501.13629. — [arXiv:2501.13629](https://arxiv.org/abs/2501.13629) [find](https://scholar.google.com/scholar?q=Zhenghao+Lin%2C+Zihao+Tang%2C+Xiao+Liu%2C+Yeyun+Gong%2C+Yi+Cheng%2C+Qi+Chen%2C+Hang+Li%2C+Ying+Xin%2C+Ziyue+Yang%2C+Kailai+Yang%2C+et+al%2E+2025%2E+Sigma%3A+Differential+rescaling+of+query%2C+key+and+value+for+efficient+language)
20. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, et al. 2024a. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437. [find](https://scholar.google.com/scholar?q=Aixin+Liu%2C+Bei+Feng%2C+Bing+Xue%2C+Bingxuan+Wang%2C+Bochao+Wu%2C+Chengda+Lu%2C+Chenggang+Zhao%2C+Chengqi+Deng%2C+Chenyu+Zhang%2C+Chong+Ruan%2C+et+al%2E+2024a%2E+Deepseek%2Dv3+technical+report%2E+arXiv+preprint+arXiv%3A2412%2E19437)
21. Jian Liu, Leyang Cui, Hanmeng Liu, Dandan Huang, Yile Wang, and Yue Zhang. 2020. [Logiqa: A challenge dataset for machine reading comprehension with logical reasoning](https://md.research.elamur.ai/free/en/papers/liu-2020-logiqa-challenge-dataset-machine-reading/). arXiv preprint arXiv:2007.08124.
22. Qian Liu, Xiaosen Zheng, Niklas Muennighoff, Guangtao Zeng, Longxu Dou, Tianyu Pang, Jing Jiang, and Min Lin. 2024b. [Regmix: Data mixture as regression for language model pre-training](https://md.research.elamur.ai/free/en/papers/liu-2024-regmix-data-mixture-regression-language/). arXiv preprint arXiv:2407.01492.
23. Yun Luo, Zhen Yang, Fandong Meng, Yafu Li, Jie Zhou, and Yue Zhang. 2023. An empirical study of catastrophic forgetting in large language models during continual fine-tuning. arXiv preprint arXiv:2308.08747. [find](https://scholar.google.com/scholar?q=Yun+Luo%2C+Zhen+Yang%2C+Fandong+Meng%2C+Yafu+Li%2C+Jie+Zhou%2C+and+Yue+Zhang%2E+2023%2E+An+empirical+study+of+catastrophic+forgetting+in+large+language+models+during+continual+fine%2Dtuning%2E+arXiv+preprint+arXiv%3A2308)
24. Zheheng Luo, Xin Zhang, Xiao Liu, Haoling Li, Yeyun Gong, Chen Qi, and Peng Cheng. 2024. Veloci-tune: A velocity-based dynamic domain reweighting method for continual pre-training. arXiv preprint arXiv:2411.14318. — [arXiv:2411.14318](https://arxiv.org/abs/2411.14318) [find](https://scholar.google.com/scholar?q=Zheheng+Luo%2C+Xin+Zhang%2C+Xiao+Liu%2C+Haoling+Li%2C+Yeyun+Gong%2C+Chen+Qi%2C+and+Peng+Cheng%2E+2024%2E+Veloci%2Dtune%3A+A+velocity%2Dbased+dynamic+domain+reweighting+method+for+continual+pre%2Dtraining%2E+arXiv+preprint+arXi)
25. Todor Mihaylov, Peter Clark, Tushar Khot, and Ashish Sabharwal. 2018. Can a suit of armor conduct electricity? a new dataset for open book question answering. arXiv preprint arXiv:1809.02789. [find](https://scholar.google.com/scholar?q=Todor+Mihaylov%2C+Peter+Clark%2C+Tushar+Khot%2C+and+Ashish+Sabharwal%2E+2018%2E+Can+a+suit+of+armor+conduct+electricity%3F+a+new+dataset+for+open+book+question+answering%2E+arXiv+preprint+arXiv%3A1809%2E02789%2E)
26. Team OLMo, Pete Walsh, Luca Soldaini, Dirk Groeneveld, Kyle Lo, Shane Arora, Akshita Bhagia, Yuling Gu, Shengyi Huang, Matt Jordan, et al. 2024. 2 olmo 2 furious. arXiv preprint arXiv:2501.00656. [find](https://scholar.google.com/scholar?q=Team+OLMo%2C+Pete+Walsh%2C+Luca+Soldaini%2C+Dirk+Groeneveld%2C+Kyle+Lo%2C+Shane+Arora%2C+Akshita+Bhagia%2C+Yuling+Gu%2C+Shengyi+Huang%2C+Matt+Jordan%2C+et+al%2E+2024%2E+2+olmo+2+furious%2E+arXiv+preprint+arXiv%3A2501%2E00656%2E)
27. Guilherme Penedo, Hynek Kydlíček, Anton Lozhkov, Margaret Mitchell, Colin A Raffel, Leandro Von Werra, Thomas Wolf, et al. 2024. [The fineweb datasets: Decanting the web for the finest text data at scale](https://md.research.elamur.ai/free/en/papers/penedo-2024-fineweb-datasets-decanting-web-finest/). Advances in Neural Information Processing Systems, 37:30811–30849.
28. Alexey Rukhovich, Alexander Podolskiy, and Irina Piontkovskaya. 2025. Commute your domains: Trajectory optimality criterion for multi-domain learning. arXiv preprint arXiv:2501.15556. — [arXiv:2501.15556](https://arxiv.org/abs/2501.15556) [find](https://scholar.google.com/scholar?q=Alexey+Rukhovich%2C+Alexander+Podolskiy%2C+and+Irina+Piontkovskaya%2E+2025%2E+Commute+your+domains%3A+Trajectory+optimality+criterion+for+multi%2Ddomain+learning%2E+arXiv+preprint+arXiv%3A2501%2E15556%2E+%E2%80%94+%5BarXiv%3A2501%2E15)
29. Shiori Sagawa, Pang Wei Koh, Tatsunori B Hashimoto, and Percy Liang. 2019. Distributionally robust neural networks for group shifts: On the importance of regularization for worst-case generalization. arXiv preprint arXiv:1911.08731. [find](https://scholar.google.com/scholar?q=Shiori+Sagawa%2C+Pang+Wei+Koh%2C+Tatsunori+B+Hashimoto%2C+and+Percy+Liang%2E+2019%2E+Distributionally+robust+neural+networks+for+group+shifts%3A+On+the+importance+of+regularization+for+worst%2Dcase+generalization%2E+)
30. Keisuke Sakaguchi, Ronan Le Bras, Chandra Bhagavatula, and Yejin Choi. 2021. Winogrande: An adversarial winograd schema challenge at scale. Communications of the ACM, 64(9):99–106. [find](https://scholar.google.com/scholar?q=Keisuke+Sakaguchi%2C+Ronan+Le+Bras%2C+Chandra+Bhagavatula%2C+and+Yejin+Choi%2E+2021%2E+Winogrande%3A+An+adversarial+winograd+schema+challenge+at+scale%2E+Communications+of+the+ACM%2C+64%289%29%3A99%E2%80%93106%2E)
31. Takuma Seno and Michita Imai. 2022. d3rlpy: An offline deep reinforcement learning library. Journal of Machine Learning Research, 23(315):1–20. [find](https://scholar.google.com/scholar?q=Takuma+Seno+and+Michita+Imai%2E+2022%2E+d3rlpy%3A+An+offline+deep+reinforcement+learning+library%2E+Journal+of+Machine+Learning+Research%2C+23%28315%29%3A1%E2%80%9320%2E)
32. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, YK Li, Y Wu, et al. 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.
33. Zhiqiang Shen, Tianhua Tao, Liqun Ma, Willie Neiswanger, Zhengzhong Liu, Hongyi Wang, Bowen Tan, Joel Hestness, Natalia Vassilieva, Daria Soboleva, et al. 2023. [Slimpajama-dc: Understanding data combinations for llm training](https://md.research.elamur.ai/free/en/papers/shen-2023-slimpajama-dc-understanding-data-combinations/). arXiv preprint arXiv:2309.10818.
34. Haizhou Shi, Zihao Xu, Hengyi Wang, Weiyi Qin, Wenyuan Wang, Yibin Wang, Zifeng Wang, Sayna Ebrahimi, and Hao Wang. 2024. Continual learning of large language models: A comprehensive survey. ACM Computing Surveys. [find](https://scholar.google.com/scholar?q=Haizhou+Shi%2C+Zihao+Xu%2C+Hengyi+Wang%2C+Weiyi+Qin%2C+Wenyuan+Wang%2C+Yibin+Wang%2C+Zifeng+Wang%2C+Sayna+Ebrahimi%2C+and+Hao+Wang%2E+2024%2E+Continual+learning+of+large+language+models%3A+A+comprehensive+survey%2E+ACM+Compu)
35. Mohammad Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper, and Bryan Catanzaro. 2019. Megatron-lm: Training multi-billion parameter language models using model parallelism. arXiv preprint arXiv:1909.08053. [find](https://scholar.google.com/scholar?q=Mohammad+Shoeybi%2C+Mostofa+Patwary%2C+Raul+Puri%2C+Patrick+LeGresley%2C+Jared+Casper%2C+and+Bryan+Catanzaro%2E+2019%2E+Megatron%2Dlm%3A+Training+multi%2Dbillion+parameter+language+models+using+model+parallelism%2E+arXiv+p)
36. Dan Su, Kezhi Kong, Ying Lin, Joseph Jennings, Brandon Norick, Markus Kliegl, Mostofa Patwary, Mohammad Shoeybi, and Bryan Catanzaro. 2024. [Nemotron-cc: Transforming common crawl into a refined long-horizon pretraining dataset](https://md.research.elamur.ai/free/en/papers/su-2025-nemotron-cc-transforming-common-crawl/). arXiv preprint arXiv:2412.02595.
37. Sang Michael Xie, Hieu Pham, Xuanyi Dong, Nan Du, Hanxiao Liu, Yifeng Lu, Percy S Liang, Quoc V Le, Tengyu Ma, and Adams Wei Yu. 2023. Doremi: Optimizing data mixtures speeds up language model pretraining. Advances in Neural Information Processing Systems, 36:69798–69818. [find](https://scholar.google.com/scholar?q=Sang+Michael+Xie%2C+Hieu+Pham%2C+Xuanyi+Dong%2C+Nan+Du%2C+Hanxiao+Liu%2C+Yifeng+Lu%2C+Percy+S+Liang%2C+Quoc+V+Le%2C+Tengyu+Ma%2C+and+Adams+Wei+Yu%2E+2023%2E+Doremi%3A+Optimizing+data+mixtures+speeds+up+language+model+pretrai)
38. 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.
39. Richard S Sutton, David McAllester, Satinder Singh, and Yishay Mansour. 1999. Policy gradient methods for reinforcement learning with function approximation. Advances in neural information processing systems, 12. [find](https://scholar.google.com/scholar?q=Richard+S+Sutton%2C+David+McAllester%2C+Satinder+Singh%2C+and+Yishay+Mansour%2E+1999%2E+Policy+gradient+methods+for+reinforcement+learning+with+function+approximation%2E+Advances+in+neural+information+processing+)
40. An Yang, Beichen Zhang, Binyuan Hui, Bofei Gao, Bowen Yu, Chengpeng Li, Dayiheng Liu, Jianhong Tu, Jingren Zhou, Junyang Lin, et al. 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/)
41. Tao Tu, Shekoofeh Azizi, Danny Driess, Mike Schaekermann, Mohamed Amin, Pi-Chuan Chang, Andrew Carroll, Charles Lau, Ryutaro Tanno, Ira Ktena, et al. 2024. Towards generalist biomedical ai. Nejm Ai, 1(3):AIoa2300138. [find](https://scholar.google.com/scholar?q=Tao+Tu%2C+Shekoofeh+Azizi%2C+Danny+Driess%2C+Mike+Schaekermann%2C+Mohamed+Amin%2C+Pi%2DChuan+Chang%2C+Andrew+Carroll%2C+Charles+Lau%2C+Ryutaro+Tanno%2C+Ira+Ktena%2C+et+al%2E+2024%2E+Towards+generalist+biomedical+ai%2E+Nejm+Ai%2C+1)
42. Jiasheng Ye, Peiju Liu, Tianxiang Sun, Jun Zhan, Yunhua Zhou, and Xipeng Qiu. 2024. Data mixing laws: Optimizing data mixtures by predicting language modeling performance. arXiv preprint arXiv:2403.16952. [find](https://scholar.google.com/scholar?q=Jiasheng+Ye%2C+Peiju+Liu%2C+Tianxiang+Sun%2C+Jun+Zhan%2C+Yunhua+Zhou%2C+and+Xipeng+Qiu%2E+2024%2E+Data+mixing+laws%3A+Optimizing+data+mixtures+by+predicting+language+modeling+performance%2E+arXiv+preprint+arXiv%3A2403%2E16)
43. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. [Attention is all you need](https://md.research.elamur.ai/free/en/papers/vaswani-2017-attention-all-need/). Advances in neural information processing systems, 30.
44. Longhui Yu, Weisen Jiang, Han Shi, Jincheng Yu, Zhengying Liu, Yu Zhang, James T Kwok, Zhenguo Li, Adrian Weller, and Weiyang Liu. 2023. [Metamath: Bootstrap your own mathematical questions for large language models](https://md.research.elamur.ai/free/en/papers/yu-2023-metamath-bootstrap-your-own-mathematical/). arXiv preprint arXiv:2309.12284.
45. Ke Wang, Houxing Ren, Aojun Zhou, Zimu Lu, Sichun Luo, Weikang Shi, Renrui Zhang, Linqi Song, Mingjie Zhan, and Hongsheng Li. 2023. [Mathcoder: Seamless code integration in llms for enhanced mathematical reasoning](https://md.research.elamur.ai/free/en/papers/wang-2023-mathcoder-seamless-code-integration-llms/). arXiv preprint arXiv:2310.03731.
46. Johannes Welbl, Nelson F Liu, and Matt Gardner. 2017. [Crowdsourcing multiple choice science questions](https://md.research.elamur.ai/free/en/papers/welbl-2017-crowdsourcing-multiple-choice-science-questions/). arXiv preprint arXiv:1707.06209.
47. Rowan Zellers, Ari Holtzman, Yonatan Bisk, Ali Farhadi, and Yejin Choi. 2019. Hellaswag: Can a machine really finish your sentence? arXiv preprint arXiv:1905.07830. [find](https://scholar.google.com/scholar?q=Rowan+Zellers%2C+Ari+Holtzman%2C+Yonatan+Bisk%2C+Ali+Farhadi%2C+and+Yejin+Choi%2E+2019%2E+Hellaswag%3A+Can+a+machine+really+finish+your+sentence%3F+arXiv+preprint+arXiv%3A1905%2E07830%2E)
48. Xiyuan Zhang, Ranak Roy Chowdhury, Rajesh K Gupta, and Jingbo Shang. 2024. Large language models for time series: A survey. arXiv preprint arXiv:2402.01801. — [arXiv:2402.01801](https://arxiv.org/abs/2402.01801) [find](https://scholar.google.com/scholar?q=Xiyuan+Zhang%2C+Ranak+Roy+Chowdhury%2C+Rajesh+K+Gupta%2C+and+Jingbo+Shang%2E+2024%2E+Large+language+models+for+time+series%3A+A+survey%2E+arXiv+preprint+arXiv%3A2402%2E01801%2E+%E2%80%94+%5BarXiv%3A2402%2E01801%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F)

*Updated 31 October 2026*