---
title: "Render-of-Thought — paper by Wang et al. (2026)"
description: "Chain-of-Thought (CoT) reasoning improves LLM accuracy, but its long outputs slow inference and use… 39 references, cited by 1 paper here."
authors: ["Yifan Wang", "Shiyu Li", "Peiming Li", "Xiaochen Yang", "Zheng Wei", "Yang Tang"]
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.2097/"
doi: "10.18653/v1/2026.acl-long.2097"
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-render-thought-rendering-textual-chain/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wang-2026-render-thought-rendering-textual-chain/)

[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/16/) / Render-of-Thought: Rendering Textual Chain-of-Thought as Images for Visual Latent Reasoning

# Render-of-Thought: Rendering Textual Chain-of-Thought as Images for Visual Latent Reasoning

**Authors:** [Yifan Wang](https://md.research.elamur.ai/free/en/authors/yifan-wang/), [Shiyu Li](https://md.research.elamur.ai/free/en/authors/shiyu-li/), Peiming Li, Xiaochen Yang, Zheng Wei and [Yang Tang](https://md.research.elamur.ai/free/en/authors/yang-tang/) · **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.2097/) · [PDF](https://aclanthology.org/2026.acl-long.2097.pdf) · [DOI 10.18653/v1/2026.acl-long.2097](https://doi.org/10.18653/v1/2026.acl-long.2097) · [OpenAlex](https://openalex.org/W7166858751)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · 1 in our library, of the 149,439 papers whose bibliographies we read

## Summary
Chain-of-Thought (CoT) reasoning improves LLM accuracy, but its long outputs slow inference and use a lot of memory. Existing latent-reasoning methods compress thoughts into hidden vectors and only supervise the final answer, so the intermediate reasoning cannot be inspected. Render-of-Thought (RoT) addresses both problems. It draws each textual reasoning chain as a single-line image and passes it through the frozen vision encoder of an existing VLM to obtain target embeddings. A small projection head on the LLM learns to predict these embeddings, so the model reasons in a continuous visual latent space that stays anchored to the rendered text.

Training has two stages. First, only the head is trained, using an MSE loss against the encoder outputs plus a loss on the end marker and the answer. Then the model is fine-tuned to generate the latent trajectory autoregressively. No rendering happens at inference.

On math and logic benchmarks with Qwen3-VL-4B-Instruct, RoT cuts reasoning tokens by 3–4× and speeds up inference compared with explicit CoT, while keeping accuracy competitive. It needs no extra pre-training.

## Key points
- **Rendering setup:** CoT text is rendered as one line, black on white, with dynamic width. By default, images are 32 px tall with 4 px of padding and a 20 px font, which gives left-to-right patch order matching the text sequence.
- **Architecture and losses:** The Visual Projection Head is a two-layer MLP with SwiGLU activation. In Stage I the LLM backbone and vision encoder are frozen, and the loss combines MSE alignment with cross-entropy on `<|img_end|>` and the answer.
- **Models and data:** The models used are Qwen3-VL-2B/4B-Instruct and LLaVa-V1.6-Mistral-7B. The datasets are GSM8k-Aug, GSM8k, GSM-Hard, SVAMP, MultiArith and MATH; code is released at github.com/TencentBAC/RoT.
- **Token budget:** The best latent budget depends on the task: 32 tokens for GSM8k-Aug and 64 for MATH. Later latent tokens become nearly identical, which suggests the core reasoning is encoded early.
- **Limitations:** Evaluation covers only English math and logic tasks, so other languages, including Ukrainian, are untested. Dynamic stopping via special tokens is unstable, and rendering adds training cost.

*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
- [Peking University](https://md.research.elamur.ai/free/en/universities/peking-university/) — University, Beijing, China · 244 works · 39,359 citations
- [Tsinghua Shenzhen International Graduate School](https://md.research.elamur.ai/free/en/universities/tsinghua-shenzhen-international-graduate-school/) — University, Shenzhen, China · 28 works · 3,035 citations
- [University of Glasgow](https://md.research.elamur.ai/free/en/universities/university-of-glasgow/) — University, Glasgow, United Kingdom · 11 works · 635 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 · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 2,869 works

🧭 Topics (1): [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,426 works

## Cite

Cite as: Yifan Wang, Shiyu Li, Peiming Li, Xiaochen Yang, Zheng Wei and Yang Tang. 2026. Render-of-Thought: Rendering Textual Chain-of-Thought as Images for Visual Latent Reasoning. 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{wang2026render,
  title = {{Render-of-Thought: Rendering Textual Chain-of-Thought as Images for Visual Latent Reasoning}},
  author = {Yifan Wang, Shiyu Li, Peiming Li, Xiaochen Yang, Zheng Wei and Yang Tang},
  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.2097},
  url = {https://doi.org/10.18653/v1/2026.acl-long.2097}
}
```

- [← previous REGLAT at SemEval-2026 Task 9: Enhancing Arabic Online Polarization Detection Using AraBERT and Synonym Replacement Augmentation](https://md.research.elamur.ai/free/en/papers/fetouh-2026-reglat-semeval-2026-task-9/)
- [next → RepoGenesis: Benchmarking End-to-End Microservice Generation from Readme to Repository](https://md.research.elamur.ai/free/en/papers/peng-2026-repogenesis-benchmarking-end-end-microservice/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/16/)

## References

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

1. Pranjal Aggarwal and Sean Welleck. 2025. [L1: Controlling how long a reasoning model thinks with reinforcement learning](https://md.research.elamur.ai/free/en/papers/aggarwal-2025-l1-controlling-long-reasoning-model/). arXiv preprint arXiv:2503.04697.
2. Shuai Bai, Yuxuan Cai, Ruizhe Chen, Keqin Chen, Xionghui Chen, Zesen Cheng, Lianghao Deng, Wei Ding, Chang Gao, Chunjiang Ge, Wenbin Ge, Zhifang Guo, Qidong Huang, Jie Huang, Fei Huang, Binyuan Hui, Shutong Jiang, Zhaohai Li, Mingsheng Li, and 45 others. 2025. [Qwen3-vl technical report](https://md.research.elamur.ai/free/en/papers/bai-2025-qwen3-vl-technical-report/). Preprint, ar…
3. Zigeng Chen, Xinyin Ma, Gongfan Fang, Ruonan Yu, and Xinchao Wang. 2025. Verithinker: Learning to verify makes reasoning model efficient. arXiv preprint arXiv:2505.17941. — [arXiv:2505.17941](https://arxiv.org/abs/2505.17941) [find](https://scholar.google.com/scholar?q=Zigeng+Chen%2C+Xinyin+Ma%2C+Gongfan+Fang%2C+Ruonan+Yu%2C+and+Xinchao+Wang%2E+2025%2E+Verithinker%3A+Learning+to+verify+makes+reasoning+model+efficient%2E+arXiv+preprint+arXiv%3A2505%2E17941%2E+%E2%80%94+%5BarXiv%3A2505%2E17941%5D%28https%3A%2F%2F)
4. Jiale Cheng, Yusen Liu, Xinyu Zhang, Yulin Fei, Wenyi Hong, Ruiliang Lyu, Weihan Wang, Zhe Su, Xiaotao Gu, Xiao Liu, and 1 others. 2025. [Glyph: Scaling context windows via visual-text compression](https://md.research.elamur.ai/free/en/papers/cheng-2026-glyph-scaling-context-windows-visual/). arXiv preprint arXiv:2510.17800.
5. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias 45244 9 Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, and 1 others. 2021. [Training verifiers to solve math word problems](https://md.research.elamur.ai/free/en/papers/cobbe-2021-training-verifiers-solve-math-word/). arXiv preprint arXiv:2110.14168.
6. Yuntian Deng, Kiran Prasad, Roland Fernandez, Paul Smolensky, Vishrav Chaudhary, and Stuart Shieber. 2023. Implicit chain of thought reasoning via knowledge distillation. arXiv preprint arXiv:2311.01460. [find](https://scholar.google.com/scholar?q=Yuntian+Deng%2C+Kiran+Prasad%2C+Roland+Fernandez%2C+Paul+Smolensky%2C+Vishrav+Chaudhary%2C+and+Stuart+Shieber%2E+2023%2E+Implicit+chain+of+thought+reasoning+via+knowledge+distillation%2E+arXiv+preprint+arXiv%3A2311%2E014)
7. Luyu Gao, Aman Madaan, Shuyan Zhou, Uri Alon, Pengfei Liu, Yiming Yang, Jamie Callan, and Graham Neubig. 2023. Pal: Program-aided language models. In International Conference on Machine Learning, pages 10764–10799. PMLR. [find](https://scholar.google.com/scholar?q=Luyu+Gao%2C+Aman+Madaan%2C+Shuyan+Zhou%2C+Uri+Alon%2C+Pengfei+Liu%2C+Yiming+Yang%2C+Jamie+Callan%2C+and+Graham+Neubig%2E+2023%2E+Pal%3A+Program%2Daided+language+models%2E+In+International+Conference+on+Machine+Learning%2C+page)
8. Tingxu Han, Zhenting Wang, Chunrong Fang, Shiyu Zhao, Shiqing Ma, and Zhenyu Chen. 2025. [Token-budget-aware llm reasoning](https://md.research.elamur.ai/free/en/papers/han-2024-token-budget-aware-llm-reasoning/). In Findings of the Association for Computational Linguistics: ACL 2025, pages 24842–24855.
9. Shibo Hao, Sainbayar Sukhbaatar, DiJia Su, Xian Li, Zhiting Hu, Jason Weston, and Yuandong Tian. 2024. Training large language models to reason in a continuous latent space. arXiv preprint arXiv:2412.06769. [find](https://scholar.google.com/scholar?q=Shibo+Hao%2C+Sainbayar+Sukhbaatar%2C+DiJia+Su%2C+Xian+Li%2C+Zhiting+Hu%2C+Jason+Weston%2C+and+Yuandong+Tian%2E+2024%2E+Training+large+language+models+to+reason+in+a+continuous+latent+space%2E+arXiv+preprint+arXiv%3A2412%2E)
10. D Hendrycks. 2016. Gaussian error linear units (gelus). arXiv preprint arXiv:1606.08415. [find](https://scholar.google.com/scholar?q=D+Hendrycks%2E+2016%2E+Gaussian+error+linear+units+%28gelus%29%2E+arXiv+preprint+arXiv%3A1606%2E08415%2E)
11. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. 2024. [Measuring mathematical problem solving with the math dataset](https://md.research.elamur.ai/free/en/papers/hendrycks-2021-measuring-mathematical-problem-solving-math/), 2021. URL https://arxiv.org/abs/2103.03874, 2. [↗](https://arxiv.org/abs/2103.03874)
12. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen, and 1 others. 2022. [Lora: Low-rank adaptation of large language models](https://md.research.elamur.ai/free/en/papers/hu-2021-lora-low-rank-adaptation-large/). ICLR, 1(2):3.
13. Yu Kang, Xianghui Sun, Liangyu Chen, and Wei Zou. 2025. C3ot: Generating shorter chain-of-thought without compromising effectiveness. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 24312–24320. [find](https://scholar.google.com/scholar?q=Yu+Kang%2C+Xianghui+Sun%2C+Liangyu+Chen%2C+and+Wei+Zou%2E+2025%2E+C3ot%3A+Generating+shorter+chain%2Dof%2Dthought+without+compromising+effectiveness%2E+In+Proceedings+of+the+AAAI+Conference+on+Artificial+Intelligence%2C+)
14. Bangzheng Li, Ximeng Sun, Jiang Liu, Ze Wang, Jialian Wu, Xiaodong Yu, Hao Chen, Emad Barsoum, Muhao Chen, and Zicheng Liu. 2025a. Latent visual reasoning. arXiv preprint arXiv:2509.24251. — [arXiv:2509.24251](https://arxiv.org/abs/2509.24251) [find](https://scholar.google.com/scholar?q=Bangzheng+Li%2C+Ximeng+Sun%2C+Jiang+Liu%2C+Ze+Wang%2C+Jialian+Wu%2C+Xiaodong+Yu%2C+Hao+Chen%2C+Emad+Barsoum%2C+Muhao+Chen%2C+and+Zicheng+Liu%2E+2025a%2E+Latent+visual+reasoning%2E+arXiv+preprint+arXiv%3A2509%2E24251%2E+%E2%80%94+%5BarXiv%3A25)
15. Yanhong Li, Zixuan Lan, and Jiawei Zhou. 2025b. Text or pixels? evaluating efficiency and understanding of llms with visual text inputs. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 10564–10578. [find](https://scholar.google.com/scholar?q=Yanhong+Li%2C+Zixuan+Lan%2C+and+Jiawei+Zhou%2E+2025b%2E+Text+or+pixels%3F+evaluating+efficiency+and+understanding+of+llms+with+visual+text+inputs%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+E)
16. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023. [Visual instruction tuning](https://md.research.elamur.ai/free/en/papers/liu-2023-visual-instruction-tuning/). Advances in neural information processing systems, 36:34892– 34916.
17. Jiayu Liu, Zhenya Huang, Anya Sims, Enhong Chen, Yee Whye Teh, and Ning Miao. 2025. Marcos: Deep thinking by markov chain of continuous thoughts. arXiv preprint arXiv:2509.25020. — [arXiv:2509.25020](https://arxiv.org/abs/2509.25020) [find](https://scholar.google.com/scholar?q=Jiayu+Liu%2C+Zhenya+Huang%2C+Anya+Sims%2C+Enhong+Chen%2C+Yee+Whye+Teh%2C+and+Ning+Miao%2E+2025%2E+Marcos%3A+Deep+thinking+by+markov+chain+of+continuous+thoughts%2E+arXiv+preprint+arXiv%3A2509%2E25020%2E+%E2%80%94+%5BarXiv%3A2509%2E25020%5D%28)
18. Ilya Loshchilov and Frank Hutter. 2017. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101. [find](https://scholar.google.com/scholar?q=Ilya+Loshchilov+and+Frank+Hutter%2E+2017%2E+Decoupled+weight+decay+regularization%2E+arXiv+preprint+arXiv%3A1711%2E05101%2E)
19. Yujie Lu, Xiujun Li, Tsu-Jui Fu, Miguel Eckstein, and William Yang Wang. 2024. From text to pixel: Advancing long-context understanding in mllms. arXiv preprint arXiv:2405.14213. — [arXiv:2405.14213](https://arxiv.org/abs/2405.14213) [find](https://scholar.google.com/scholar?q=Yujie+Lu%2C+Xiujun+Li%2C+Tsu%2DJui+Fu%2C+Miguel+Eckstein%2C+and+William+Yang+Wang%2E+2024%2E+From+text+to+pixel%3A+Advancing+long%2Dcontext+understanding+in+mllms%2E+arXiv+preprint+arXiv%3A2405%2E14213%2E+%E2%80%94+%5BarXiv%3A2405%2E14213%5D%28)
20. Haotian Luo, Li Shen, Haiying He, Yibo Wang, Shiwei Liu, Wei Li, Naiqiang Tan, Xiaochun Cao, and Dacheng Tao. 2025. O1-pruner: Length-harmonizing fine-tuning for o1-like reasoning pruning. arXiv preprint arXiv:2501.12570. [find](https://scholar.google.com/scholar?q=Haotian+Luo%2C+Li+Shen%2C+Haiying+He%2C+Yibo+Wang%2C+Shiwei+Liu%2C+Wei+Li%2C+Naiqiang+Tan%2C+Xiaochun+Cao%2C+and+Dacheng+Tao%2E+2025%2E+O1%2Dpruner%3A+Length%2Dharmonizing+fine%2Dtuning+for+o1%2Dlike+reasoning+pruning%2E+arXiv+prepr)
21. Zhiheng Lyu, Xueguang Ma, and Wenhu Chen. 2025. Pixelworld: Towards perceiving everything as pixels. arXiv preprint arXiv:2501.19339. — [arXiv:2501.19339](https://arxiv.org/abs/2501.19339) [find](https://scholar.google.com/scholar?q=Zhiheng+Lyu%2C+Xueguang+Ma%2C+and+Wenhu+Chen%2E+2025%2E+Pixelworld%3A+Towards+perceiving+everything+as+pixels%2E+arXiv+preprint+arXiv%3A2501%2E19339%2E+%E2%80%94+%5BarXiv%3A2501%2E19339%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2501%2E19339%29)
22. Arkil Patel, Satwik Bhattamishra, and Navin Goyal. 2021. [Are nlp models really able to solve simple math word problems?](https://md.research.elamur.ai/free/en/papers/patel-2021-are-nlp-models-really-able/) arXiv preprint arXiv:2103.07191.
23. Jeff Rasley, Samyam Rajbhandari, Olatunji Ruwase, and Yuxiong He. 2020. Deepspeed: System optimizations enable training deep learning models with over 100 billion parameters. In Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, pages 3505–3506. [find](https://scholar.google.com/scholar?q=Jeff+Rasley%2C+Samyam+Rajbhandari%2C+Olatunji+Ruwase%2C+and+Yuxiong+He%2E+2020%2E+Deepspeed%3A+System+optimizations+enable+training+deep+learning+models+with+over+100+billion+parameters%2E+In+Proceedings+of+the+26t)
24. 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.
25. Subhro Roy and Dan Roth. 2015. [Solving general arithmetic word problems](https://md.research.elamur.ai/free/en/papers/roy-2015-solving-general-arithmetic-word-problems/). In Proceedings of the 2015 conference on empirical methods in natural language processing, pages 1743–1752.
26. Noam Shazeer. 2020. Glu variants improve transformer. arXiv preprint arXiv:2002.05202. [find](https://scholar.google.com/scholar?q=Noam+Shazeer%2E+2020%2E+Glu+variants+improve+transformer%2E+arXiv+preprint+arXiv%3A2002%2E05202%2E)
27. Xuan Shen, Yizhou Wang, Xiangxi Shi, Yanzhi Wang, Pu Zhao, and Jiuxiang Gu. 2025a. Efficient reasoning with hidden thinking. arXiv preprint arXiv:2501.19201. — [arXiv:2501.19201](https://arxiv.org/abs/2501.19201) [find](https://scholar.google.com/scholar?q=Xuan+Shen%2C+Yizhou+Wang%2C+Xiangxi+Shi%2C+Yanzhi+Wang%2C+Pu+Zhao%2C+and+Jiuxiang+Gu%2E+2025a%2E+Efficient+reasoning+with+hidden+thinking%2E+arXiv+preprint+arXiv%3A2501%2E19201%2E+%E2%80%94+%5BarXiv%3A2501%2E19201%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs)
28. Zhenyi Shen, Hanqi Yan, Linhai Zhang, Zhanghao Hu, Yali Du, and Yulan He. 2025b. [Codi: Compressing chain-of-thought into continuous space via self-distillation](https://md.research.elamur.ai/free/en/papers/shen-2025-codi-compressing-chain-thought-into/). arXiv preprint arXiv:2502.21074.
29. Wenhui Tan, Jiaze Li, Jianzhong Ju, Zhenbo Luo, Jian Luan, and Ruihua Song. 2025. [Think silently, think fast: Dynamic latent compression of llm reasoning chains](https://md.research.elamur.ai/free/en/papers/tan-2025-think-silently-think-fast-dynamic/). arXiv preprint arXiv:2505.16552.
30. Yibo Wang, Haotian Luo, Huanjin Yao, Tiansheng Huang, Haiying He, Rui Liu, Naiqiang Tan, Jiaxing Huang, Xiaochun Cao, Dacheng Tao, and 1 others. 2025. R1-compress: Long chain-of-thought compression via chunk compression and search. arXiv preprint arXiv:2505.16838. 45245 10 — [arXiv:2505.16838](https://arxiv.org/abs/2505.16838) [find](https://scholar.google.com/scholar?q=Yibo+Wang%2C+Haotian+Luo%2C+Huanjin+Yao%2C+Tiansheng+Huang%2C+Haiying+He%2C+Rui+Liu%2C+Naiqiang+Tan%2C+Jiaxing+Huang%2C+Xiaochun+Cao%2C+Dacheng+Tao%2C+and+1+others%2E+2025%2E+R1%2Dcompress%3A+Long+chain%2Dof%2Dthought+compression+vi)
31. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. [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/). Advances in neural information processing systems, 35:24824– 24837.
32. Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pier-ric Cistac, Tim Rault, Remi Louf, Morgan Funtowicz, and 1 others. 2020. [Transformers: State-of-the-art natural language processing](https://md.research.elamur.ai/free/en/papers/wolf-2020-transformers-state-art-natural-language/). In Proceedings of the 2020 conference on empirical methods in natural lan…
33. Heming Xia, Chak Tou Leong, Wenjie Wang, Yongqi Li, and Wenjie Li. 2025. [Tokenskip: Controllable chain-of-thought compression in llms](https://md.research.elamur.ai/free/en/papers/xia-2025-tokenskip-controllable-chain-thought-compression/). arXiv preprint arXiv:2502.12067.
34. Violet Xiang, Charlie Snell, Kanishk Gandhi, Alon Albalak, Anikait Singh, Chase Blagden, Duy Phung, Rafael Rafailov, Nathan Lile, Dakota Mahan, and 1 others. 2025. Towards system 2 reasoning in llms: Learning how to think with meta chain-of-thought. arXiv preprint arXiv:2501.04682. [find](https://scholar.google.com/scholar?q=Violet+Xiang%2C+Charlie+Snell%2C+Kanishk+Gandhi%2C+Alon+Albalak%2C+Anikait+Singh%2C+Chase+Blagden%2C+Duy+Phung%2C+Rafael+Rafailov%2C+Nathan+Lile%2C+Dakota+Mahan%2C+and+1+others%2E+2025%2E+Towards+system+2+reasoning+in+llms%3A+)
35. Yige Xu, Xu Guo, Zhiwei Zeng, and Chunyan Miao. 2025. [Softcot: Soft chain-of-thought for efficient reasoning with llms](https://md.research.elamur.ai/free/en/papers/xu-2025-softcot-soft-chain-thought-efficient/). arXiv preprint arXiv:2502.12134.
36. 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.
37. Zhenrui Yue, Bowen Jin, Huimin Zeng, Honglei Zhuang, Zhen Qin, Jinsung Yoon, Lanyu Shang, Jiawei Han, and Dong Wang. 2025. Hybrid latent reasoning via reinforcement learning. arXiv preprint arXiv:2505.18454. — [arXiv:2505.18454](https://arxiv.org/abs/2505.18454) [find](https://scholar.google.com/scholar?q=Zhenrui+Yue%2C+Bowen+Jin%2C+Huimin+Zeng%2C+Honglei+Zhuang%2C+Zhen+Qin%2C+Jinsung+Yoon%2C+Lanyu+Shang%2C+Jiawei+Han%2C+and+Dong+Wang%2E+2025%2E+Hybrid+latent+reasoning+via+reinforcement+learning%2E+arXiv+preprint+arXiv%3A2505)
38. Jintian Zhang, Yuqi Zhu, Mengshu Sun, Yujie Luo, Shuofei Qiao, Lun Du, Da Zheng, Huajun Chen, and Ningyu Zhang. 2025. [Lightthinker: Thinking step-by-step compression](https://md.research.elamur.ai/free/en/papers/zhang-2025-lightthinker-thinking-step-step-compression/). arXiv preprint arXiv:2502.15589. 45246 11
39. Render-of-Thought: Rendering Textual Chain-of-Thought as Images for Visual Latent Reasoning [find](https://scholar.google.com/scholar?q=Render%2Dof%2DThought%3A+Rendering+Textual+Chain%2Dof%2DThought+as+Images+for+Visual+Latent+Reasoning)

*Updated 1 November 2026*