Elamur Research⌕
☰
home / papers / 2026 / Unified Thinker: A General Reasoning Core for Image Generation

Unified Thinker: A General Reasoning Core for Image Generation

Paper · 2026 · Sashuai Zhou, Qiang Zhou, Jijin Hu et al.

Summary

Open-source image generators and editors still struggle with instructions that require real reasoning, such as solving a puzzle or inferring an implicit intent. Proprietary systems like Nano Banana handle these much better. The authors propose Unified Thinker, which separates a trainable MLLM "Thinker" from the image Generator, for example a diffusion backbone. The Thinker turns a request into a structured plan containing an intent summary, constraints and ordered sub-goals, ending in an enhanced prompt. The Generator then renders that prompt. Because the two parts are separate, reasoning can be upgraded without retraining the Generator.

Training has two stages:

  • Supervised fine-tuning on HieraReason-40K, a set of hierarchical reasoning traces synthesized with Gemini-3-Pro.
  • Dual-phase reinforcement learning with the Generator in the loop, where the reward measures how well the final image satisfies the constraints. This ties the plans to what the Generator can actually draw.

The approach was tested on reasoning and general text-to-image generation and image editing. It gives large gains on reasoning benchmarks and transfers to other generators such as BAGEL. This suggests that separating reasoning from rendering is a practical, reusable design.

Key points

  • Training data: HieraReason-40K (about 40K examples) draws on four sources covering text-to-image, general editing, reasoning generation and reasoning editing. Its traces are seeded with knowledge domains such as art and culture, logic, physics and chemistry, and natural sciences.
  • Thinker backbones: Qwen2.5-VL-7B and Qwen3-VL-8B. The main training generator is Qwen-Image-Edit.
  • RiseBench ablation: the overall score rises from 8.9 (baseline) to 16.4 with a Thinker, 20.2 after joint fine-tuning, and 24.2 after both Dual-RL stages. The 7B Thinker comes close to GPT-5 (26.9) and Gemini-2.5-Pro (25.2) used as planners, and beats Qwen3-VL-30B (23.1).
  • WiseBench (overall):
    • Qwen-Image-Edit: 0.62 → 0.73 (7B) and 0.74 (8B).
    • BAGEL: 0.52 → 0.70, showing transfer across generators.
  • Limitations: GEdit quality dips slightly when the Thinker is added (7.56 → 7.49) but ends at 7.67 after full training. Planning adds inference latency, and fine geometric edits and precise text rendering remain difficult.

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

🎓Fudan UniversityShanghai · China202 works · 22,509 citations🏢Alibaba GroupHangzhou · China201 works · 17,172 citations🎓Zhejiang UniversityHangzhou · China224 works · 13,587 citations🎓Nanjing UniversityNanjing · China99 works · 8,510 citations

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksComputer Vision and Pattern Recognition · 3,486 works

🧭 Topics 2Research subjects and languages the work is about.Reasoning · 1,685 worksVision & language · 811 works

Cite

Cite as Sashuai Zhou, Qiang Zhou, Jijin Hu, Hanqing Yang, Yue Cao, Junpeng Ma, Yinchao Ma, Jun Song, Tiezheng Ge, Cheng Yu, Bo Zheng and Zhou Zhao. 2026. Unified Thinker: A General Reasoning Core for Image Generation. 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. DOI 10.18653/v1/2026.acl-long.484

@inproceedings{zhou2026unified,
  title = {{Unified Thinker: A General Reasoning Core for Image Generation}},
  author = {Sashuai Zhou, Qiang Zhou, Jijin Hu, Hanqing Yang, Yue Cao, Junpeng Ma, Yinchao Ma, Jun Song, Tiezheng Ge, Cheng Yu, Bo Zheng and Zhou Zhao},
  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.484},
  url = {https://doi.org/10.18653/v1/2026.acl-long.484}
}

References

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

  1. 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. 2025a. Qwen3-vl technical report. Preprint, a…
  2. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, Humen Zhong, Yuanzhi Zhu, Mingkun Yang, Zhaohai Li, Jianqiang Wan, Pengfei Wang, Wei Ding, Zheren Fu, Yiheng Xu, and 8 others. 2025b. Qwen2.5-vl technical report. Preprint, arXiv:2502.1…
  3. Tim Brooks, Aleksander Holynski, and Alexei A Efros. 2023. Instructpix2pix: Learning to follow image editing instructions. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 18392–18402. find
  4. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, Luke Marris, Sam Petulla, Colin Gaffney, Asaf Aharoni, Nathan Lintz, Tiago Cardal Pais, Henrik Jacobsson, Idan Szpektor, Nan-Jiang Jiang, and 3416 other…
  5. Chaorui Deng, Deyao Zhu, Kunchang Li, Chenhui Gou, Feng Li, Zeyu Wang, Shu Zhong, Weihao Yu, Xiaonan Nie, Ziang Song, Guang Shi, and Haoqi Fan. 2025. Emerging properties in unified multimodal pretraining. arXiv preprint arXiv:2505.14683. — [arXiv:2505.14683](https://arxiv.org/abs/2505.14683) find
  6. Patrick Esser, Sumith Kulal, Andreas Blattmann, Rahim Entezari, Jonas Müller, Harry Saini, Yam Levi, Dominik Lorenz, Axel Sauer, Frederic Boesel, Dustin Podell, Tim Dockhorn, Zion English, and Robin Rombach. 2024. Scaling rectified flow transformers for high-resolution image synthesis. In Forty-firs… find
  7. Rongyao Fang, Aldrich Yu, Chengqi Duan, Linjiang Huang, Shuai Bai, Yuxuan Cai, Kun Wang, Si Liu, Xihui Liu, and Hongsheng Li. 2025. Flux-reason-6m & prism-bench: A million-scale text-to-image reasoning dataset and comprehensive benchmark. arXiv preprint arXiv:2509.09680. find
  8. Tsu-Jui Fu, Wenze Hsu, William Yang Wang, Shang-Hua Li, Scott Cohen, and Yang Wang. 2024. Guiding instruction-based image editing via multimodal large language models. In International Conference on Learning Representations (ICLR). find
  9. 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
  10. Ziyu Guo, Renrui Zhang, Chengzhuo Tong, Zhizheng Zhao, Peng Gao, Hongsheng Li, and Pheng-Ann Heng. 2025b. Can we generate images with cot? let’s verify and reinforce image generation step by step. CoRR, abs/2501.13926. — [arXiv:2501.13926](https://arxiv.org/abs/2501.13926) find
  11. Feng Han, Yibin Wang, Chenglin Li, Zheming Liang, Dianyi Wang, Yang Jiao, Zhipeng Wei, Chao Gong, Cheng Jin, Jingjing Chen, and Jiaqi Wang. 2025. Unireditbench: A unified reasoning-based image editing benchmark. CoRR, abs/2511.01295. — [arXiv:2511.01295](https://arxiv.org/abs/2511.01295) find
  12. Qingdong He, Xueqin Chen, Chaoyi Wang, Yanjie Pan, Xiaobin Hu, Zhenye Gan, Yabiao Wang, Chengjie Wang, Xiangtai Li, and Jiangning Zhang. 2025. Reasoning to edit: Hypothetical instruction-based image editing with visual reasoning. arXiv preprint arXiv:2507.01908. — [arXiv:2507.01908](https://arxiv.org/abs/2507.01908) find
  13. Jonathan Ho, Ajay Jain, and Pieter Abbeel. 2020. Denoising diffusion probabilistic models. Advances in neural information processing systems. find
  14. Wenxuan Huang, Shuang Chen, Zheyong Xie, Shaosheng Cao, Shixiang Tang, Yufan Shen, Qingyu Yin, Wenbo Hu, Xiaoman Wang, Yuntian Tang, Junbo Qiao, Yue Guo, Yao Hu, Zhenfei Yin, Philip Torr, Yu Cheng, Wanli Ouyang, and Shaohui Lin. 2025. Interleaving reasoning for better text-to-image generation. CoRR,… — [arXiv:2509.06945](https://arxiv.org/abs/2509.06945) find
  15. Yuzhou Huang, Liangbin Xie, Xintao Wang, Ziyang Yuan, Xiaodong Cun, Yixiao Ge, Jiantao Zhou, Chao Dong, Rui Huang, Ruimao Zhang, and Ying Shan. 2024. Smartedit: Exploring complex instruction-based image editing with multimodal large language models. In IEEE/CVF Conference on Computer Vision and Patt… find
  16. Aaron Hurst, Adam Lerer, Adam P. Goucher, Adam Perelman, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Welihinda, Alan Hayes, Alec Radford, Aleksander Madry, Alex Baker-Whitcomb, Alex Beutel, Alex Borzunov, Alex Carney, Alex Chow, Alex Kirillov, Alex Nichol, Alex Paino, and 79 others. 2024. Gpt-4o sy…
  17. Dongzhi Jiang, Ziyu Guo, Renrui Zhang, Zhuofan Zong, Hao Li, Le Zhuo, Shilin Yan, Pheng-Ann Heng, and Hongsheng Li. 2025. T2i-r1: Reinforcing image generation with collaborative semantic-level and token-level cot. arXiv preprint arXiv:2505.00703. — [arXiv:2505.00703](https://arxiv.org/abs/2505.00703) find
  18. Xuan Ju, Xian Liu, Xintao Wang, Yuxuan Bian, Ying Shan, and Qiang Xu. 2024. Brushnet: A plug-and-play image inpainting model with decomposed dual-branch diffusion. In European Conference on Computer Vision, pages 150–168. Springer. find
  19. Black Forest Labs. 2024. Flux.1: Text-to-image synthesis via flow matching. Technical Announcement. find
  20. Hongyu Li, Manyuan Zhang, Dian Zheng, Ziyu Guo, Yimeng Jia, Kaituo Feng, Hao Yu, Yexin Liu, Yan Feng, Peng Pei, Xunliang Cai, Linjiang Huang, Hongsheng Li, and Si Liu. 2025a. Editthinker: Unlocking iterative reasoning for any image editor. find
  21. Shufan Li, Konstantinos Kallidromitis, Akash Gokul, Arsh Koneru, Yusuke Kato, Kazuki Kozuka, and Aditya Grover. 2025b. Reflect-dit: Inference-time scaling for text-to-image diffusion transformers via in-context reflection. arXiv preprint arXiv:2503.12271. — [arXiv:2503.12271](https://arxiv.org/abs/2503.12271) find
  22. Yanghao Li, Rui Qian, Bowen Pan, Haotian Zhang, Haoshuo Huang, Bowen Zhang, Jialing Tong, Haoxuan You, Xianzhi Du, Zhe Gan, Hyunjik Kim, Chao Jia, Zhenbang Wang, Yinfei Yang, Mingfei Gao, Zi-Yi Dou, Wenze Hu, Chang Gao, Dongxu Li, and 8 others. 2025c. Manzano: A simple and scalable unified multimoda… — [arXiv:2509.16197](https://arxiv.org/abs/2509.16197) find
  23. Jiaqi Liao, Zhengyuan Yang, Linjie Li, Dianqi Li, Kevin Lin, Yu Cheng, and Lijuan Wang. 2025. Imagegen-cot: Enhancing text-to-image in-context learning with chain-of-thought reasoning. arXiv preprint arXiv:2503.19312. — [arXiv:2503.19312](https://arxiv.org/abs/2503.19312) find
  24. Bin Lin, Zongjian Li, Xinhua Cheng, Yuwei Niu, Yang Ye, Xianyi He, Shenghai Yuan, Wangbo Yu, Shaodong Wang, Yunyang Ge, Yatian Pang, and Li Yuan. 2025. Uniworld-v1: High-resolution semantic encoders for unified visual understanding and generation. CoRR, abs/2506.03147. — [arXiv:2506.03147](https://arxiv.org/abs/2506.03147) find
  25. Yaron Lipman, Ricky TQ Chen, Heli Ben-Hamu, Maximilian Nickel, and Matt Le. 2022. Flow matching for generative modeling. arXiv preprint arXiv:2210.02747.
  26. Jie Liu, Gongye Liu, Jiajun Liang, Yangguang Li, Jiaheng Liu, Xintao Wang, Pengfei Wan, Di Zhang, and Wanli Ouyang. 2025a. Flow-grpo: Training flow matching models via online rl. arXiv preprint arXiv:2505.05470. — [arXiv:2505.05470](https://arxiv.org/abs/2505.05470) find
  27. Shiyu Liu, Yucheng Han, Peng Xing, Fukun Yin, Rui Wang, Wei Cheng, Jiaqi Liao, Yingming Wang, Honghao Fu, Chunrui Han, Guopeng Li, Yuang Peng, Quan Sun, Jingwei Wu, Yan Cai, Zheng Ge, Ranchen Ming, Lei Xia, Xianfang Zeng, and 5 others. 2025b. Step1x-edit: A practical framework for general image edit…
  28. Shiyin Lu, Yang Li, Qing-Guo Chen, Zhao Xu, Weihua Luo, Kaifu Zhang, and Han-Jia Ye. 2024. Ovis: Structural embedding alignment for multimodal large language model. arXiv:2405.20797. — [arXiv:2405.20797](https://arxiv.org/abs/2405.20797) find
  29. Zhenxing Mi, Kuan-Chieh Wang, Guocheng Qian, Hanrong Ye, Runtao Liu, Sergey Tulyakov, Kfir Aberman, and Dan Xu. 2025. I think, therefore i diffuse: Enabling multimodal in-context reasoning in diffusion models. arXiv preprint arXiv:2502.10458. find
  30. Yuwei Niu, Munan Ning, Mengren Zheng, Weiyang Jin, Bin Lin, Peng Jin, Jiaqi Liao, Kunpeng Ning, Chaoran Feng, Bin Zhu, and Li Yuan. 2025. Wise: A world knowledge-informed semantic evaluation for text-to-image generation. arXiv preprint arXiv:2503.07265. — [arXiv:2503.07265](https://arxiv.org/abs/2503.07265) find
  31. Xichen Pan, Satya Narayan Shukla, Aashu Singh, Zhuokai Zhao, Shlok Kumar Mishra, Jialiang Wang, Zhiyang Xu, Jiuhai Chen, Kunpeng Li, Felix Juefei-Xu, Ji Hou, and Saining Xie. 2025. Transfer between modalities with metaqueries. Preprint, arXiv:2504.06256. — [arXiv:2504.06256](https://arxiv.org/abs/2504.06256) find
  32. William Peebles and Saining Xie. 2023. Scalable diffusion models with transformers. In Proceedings of the IEEE/CVF international conference on computer vision, pages 4195–4205.
  33. Yusu Qian, Eli Bocek-Rivele, Liangchen Song, Jialing Tong, Yinfei Yang, Jiasen Lu, Wenze Hu, and Zhe Gan. 2025. Pico-banana-400k: A large-scale dataset for text-guided image editing. arXiv preprint arXiv:2510.19808. — [arXiv:2510.19808](https://arxiv.org/abs/2510.19808) find
  34. Luozheng Qin, Jia Gong, Yuqing Sun, Tianjiao Li, Mengping Yang, Xiaomeng Yang, Chao Qu, Zhiyu Tan, and Hao Li. 2025. Uni-cot: Towards unified chain-of-thought reasoning across text and vision. arXiv preprint arXiv:2508.05606. find
  35. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D. Manning, Stefano Ermon, and Chelsea Finn. 2023. Direct preference optimization: Your language model is secretly a reward model. In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Sy…
  36. Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. 2022. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695.
  37. Quan Sun, Yufeng Cui, Xiaosong Zhang, Fan Zhang, Qiying Yu, Zhengxiong Luo, Yueze Wang, Yongming Rao, Jingjing Liu, Tiejun Huang, and Xinlong Wang. 2024. Generative multimodal models are in-context learners. Preprint, arXiv:2312.13286. — [arXiv:2312.13286](https://arxiv.org/abs/2312.13286) find
  38. Gemini Team, Rohan Anil, Sebastian Borgeaud, Jean-Baptiste Alayrac, Jiahui Yu, Radu Soricut, Johan Schalkwyk, Andrew M. Dai, Anja Hauth, Katie Millican, David Silver, Melvin Johnson, Ioannis Antonoglou, Julian Schrittwieser, Amelia Glaese, Jilin Chen, Emily Pitler, Timothy Lillicrap, Angeliki Lazari… — Gemini: A Family of Highly Capable Multimodal Models
  39. Yi Wang, Mushui Liu, Wanggui He, Longxiang Zhang, Ziwei Huang, Guanghao Zhang, Fangxun Shu, Tao Zhong, Dong She, Zhelun Yu, Haoyuan Li, Weilong Dai, Mingli Song, Jie Song, and Hao Jiang. 2025. MINT: multi-modal chain of thought in unified generative models for enhanced image generation. CoRR, abs/25… — [arXiv:2503.01298](https://arxiv.org/abs/2503.01298) find
  40. Chenfei Wu, Jiahao Li, Jingren Zhou, Junyang Lin, Kaiyuan Gao, Kun Yan, Sheng ming Yin, Shuai Bai, Xiao Xu, Yilei Chen, Yuxiang Chen, Zecheng Tang, Zekai Zhang, Zhengyi Wang, An Yang, Bowen Yu, Chen Cheng, Dayiheng Liu, Deqing Li, and 20 others. 2025a. Qwen-image technical report.
  41. Chengyue Wu, Xiaokang Chen, Zhiyu Wu, Yiyang Ma, Xingchao Liu, Zizheng Pan, Wen Liu, Zhenda Xie, Xingkai Yu, Chong Ruan, and Ping Luo. 2025b. Janus: Decoupling visual encoding for unified multimodal understanding and generation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR…
  42. Chenyuan Wu, Pengfei Zheng, Ruiran Yan, Shitao Xiao, Xin Luo, Yueze Wang, Wanli Li, Xiyan Jiang, Yexin Liu, Junjie Zhou, Ze Liu, Ziyi Xia, Chaofan Li, Haoge Deng, Jiahao Wang, Kun Luo, Bo Zhang, Defu Lian, Xinlong Wang, and 3 others. 2025c. Omnigen2: Exploration to advanced multimodal generation. Co… — [arXiv:2506.18871](https://arxiv.org/abs/2506.18871) find
  43. Junfeng Wu, Yi Jiang, Chuofan Ma, Yuliang Liu, Hengshuang Zhao, Zehuan Yuan, Song Bai, and Xiang Bai. 2025d. Liquid: Language models are scalable and unified multi-modal generators. International Journal of Computer Vision. find
  44. Size Wu, Zhonghua Wu, Zerui Gong, Qingyi Tao, Sheng Jin, Qinyue Li, Wei Li, and Chen Change Loy. 2025e. Openuni: A simple baseline for unified multimodal understanding and generation. Preprint, arXiv:2505.23661. — [arXiv:2505.23661](https://arxiv.org/abs/2505.23661) find
  45. Shitao Xiao, Yueze Wang, Junjie Zhou, Huaying Yuan, Xingrun Xing, Ruiran Yan, Chaofan Li, Shuting Wang, Tiejun Huang, and Zheng Liu. 2025. Omnigen: Unified image generation. pages 13294–13304. find
  46. Jinheng Xie, Zhenheng Yang, and Mike Zheng Shou. 2025. Show-o2: Improved native unified multimodal models. arXiv preprint arXiv:2506.15564. — [arXiv:2506.15564](https://arxiv.org/abs/2506.15564) find
  47. Xiaoda Yang, Jiayang Xu, Kaixuan Luan, Xinyu Zhan, Hongshun Qiu, Shijun Shi, Hao Li, Shuai Yang, Li Zhang, Checheng Yu, and 1 others. 2025. Omni-cam: Unified multimodal video generation via camera control. arXiv preprint arXiv:2504.02312. — [arXiv:2504.02312](https://arxiv.org/abs/2504.02312) find
  48. Jiaxin Ye, Gaoxiang Cong, Chenhui Wang, Xin-Cheng Wen, Zhaoyang Li, Boyuan Cao, and Hongming Shan. 2026. Hierarchical codec diffusion for video-to-speech generation. In IEEE Conf. Comput. Vis. Pattern Recog. find
  49. Fukun Yin, Shiyu Liu, Yucheng Han, Zhibo Wang, Peng Xing, Rui Wang, Wei Cheng, Yingming Wang, Aojie Li, Zixin Yin, Pengtao Chen, Xiangyu Zhang, Daxin Jiang, Xianfang Zeng, and Gang Yu. 2025. Rea-sonedit: Towards reasoning-enhanced image editing models. Preprint, arXiv:2511.22625. — [arXiv:2511.22625](https://arxiv.org/abs/2511.22625) find
  50. Qifan Yu, Wei Chow, Zhongqi Yue, Kaihang Pan, Yang Wu, Xiaoyang Wan, Juncheng Li, Siliang Tang, Hanwang Zhang, and Yueting Zhuang. 2025. Anyedit: Mastering unified high-quality image editing for any idea. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 26125–26135. find
  51. Dong Zhang, Lingfeng He, Rui Yan, Fei Shen, and Jinhui Tang. 2025. R-genie: Reasoning-guided generative image editing. arXiv preprint arXiv:2505.17768. — [arXiv:2505.17768](https://arxiv.org/abs/2505.17768) find
  52. Xiangyu Zhao, Peiyuan Zhang, Kexian Tang, Hao Li, Zicheng Zhang, Guangtao Zhai, Junchi Yan, Hua Yang, Xue Yang, and Haodong Duan. 2025. Envisioning beyond the pixels: Benchmarking reasoning-informed visual editing. arXiv preprint arXiv:2504.02826. — [arXiv:2504.02826](https://arxiv.org/abs/2504.02826) find
  53. Sashuai Zhou, Qiang Zhou, Junpeng Ma, Yue Cao, Ruofan Hu, Ziang Zhang, Xiaoda Yang, Zhibin Wang, Jun Song, Cheng Yu, Bo Zheng, and Zhou Zhao. 2026. Spatialreward: Verifiable spatial reward modeling for fine-grained spatial consistency in text-to-image generation. Preprint, arXiv:2603.22228. — [arXiv:2603.22228](https://arxiv.org/abs/2603.22228) find
  54. Junhao Zhuang, Yanhong Zeng, Wenran Liu, Chun Yuan, and Kai Chen. 2024. A task is worth one word: Learning with task prompts for high-quality versatile image inpainting. In European Conference on Computer Vision, pages 195–211. Springer. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/zhou-2026-unified-thinker-general-reasoning-core/