Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
diffusion-cot (Diffusion CoT)
[go: Go Back, main page]
Hugging Face's logo

Diffusion CoT
Team
non-profit
Activity Feed
\n
\n\nThis organization holds the artifacts for our research conducted on enabling reasoning in diffusion-based image synthesis models. Our first\neffort in this line of research is **ReflectionFlow**, where we introduce the first ever large-scale dataset, **GenRef**, suitable for\nreflection-tuning.\n\nBelow, we provide the links related to ReflectionFlow:\n\n* [ReflectionFlow paper](https://arxiv.org/abs/2504.16080)\n* [Projection website](https://diffusion-cot.github.io/reflection2perfection/)\n* [Models and datasets](https://huggingface.co/collections/diffusion-cot/reflectionflow-release-6803e14352b1b13a16aeda44)\n* [Code](https://github.com/Diffusion-CoT/ReflectionFlow)\n\nCitation\n\n```bibtex\nmisc{zhuo2025reflectionperfectionscalinginferencetime,\n title={From Reflection to Perfection: Scaling Inference-Time Optimization for Text-to-Image Diffusion Models via Reflection Tuning}, \n author={Le Zhuo and Liangbing Zhao and Sayak Paul and Yue Liao and Renrui Zhang and Yi Xin and Peng Gao and Mohamed Elhoseiny and Hongsheng Li},\n year={2025},\n eprint={2504.16080},\n archivePrefix={arXiv},\n primaryClass={cs.CV},\n url={https://arxiv.org/abs/2504.16080}, \n}\n```\n\nEnjoy 🤗","html":"
\n \n
\n\n

This organization holds the artifacts for our research conducted on enabling reasoning in diffusion-based image synthesis models. Our first\neffort in this line of research is ReflectionFlow, where we introduce the first ever large-scale dataset, GenRef, suitable for\nreflection-tuning.

\n

Below, we provide the links related to ReflectionFlow:

\n\n

Citation

\n
misc{zhuo2025reflectionperfectionscalinginferencetime,\n      title={From Reflection to Perfection: Scaling Inference-Time Optimization for Text-to-Image Diffusion Models via Reflection Tuning}, \n      author={Le Zhuo and Liangbing Zhao and Sayak Paul and Yue Liao and Renrui Zhang and Yi Xin and Peng Gao and Mohamed Elhoseiny and Hongsheng Li},\n      year={2025},\n      eprint={2504.16080},\n      archivePrefix={arXiv},\n      primaryClass={cs.CV},\n      url={https://arxiv.org/abs/2504.16080}, \n}\n
\n

Enjoy 🤗

\n","classNames":"hf-sanitized hf-sanitized-gNJLJGCWtfv-hHWMYQzrp"},"users":[{"_id":"5f7fbd813e94f16a85448745","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1649681653581-5f7fbd813e94f16a85448745.jpeg","isPro":true,"fullname":"Sayak Paul","user":"sayakpaul","type":"user"},{"_id":"6358a167f56b03ec9147074d","avatarUrl":"/avatars/e54ea7bf0c240cf76d538296efb3976c.svg","isPro":false,"fullname":"Le Zhuo","user":"JackyZhuo","type":"user"},{"_id":"63e5adeb46965e1161d54ad4","avatarUrl":"/avatars/65ce4eabd26601d669688f9cca646893.svg","isPro":false,"fullname":"zlb","user":"metazlb","type":"user"},{"_id":"639be86b59473c6ae02ef9c4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/639be86b59473c6ae02ef9c4/gw34RBCVZCOkcAA79xUr3.png","isPro":true,"fullname":"Jie Liu","user":"jieliu","type":"user"}],"userCount":4,"collections":[{"slug":"diffusion-cot/reflectionflow-release-6803e14352b1b13a16aeda44","title":"ReflectionFlow release","description":"https://diffusion-cot.github.io/reflection2perfection/","gating":false,"lastUpdated":"2025-04-23T09:46:17.905Z","owner":{"_id":"67bdb50c08abc7f641f215a7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f7fbd813e94f16a85448745/u7-k61BeAabAW9S4AUm94.png","fullname":"Diffusion CoT","name":"diffusion-cot","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"plan":"team","followerCount":20,"isUserFollowing":false},"items":[{"_id":"680846db9d64245d939afa83","position":0,"type":"paper","id":"2504.16080","title":"From Reflection to Perfection: Scaling Inference-Time Optimization for\n Text-to-Image Diffusion Models via Reflection Tuning","thumbnailUrl":"https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2504.16080.png","upvotes":15,"publishedAt":"2025-04-22T17:58:07.000Z","isUpvotedByUser":false},{"_id":"6803e153f8b0d34cab4f56f9","position":1,"type":"dataset","note":{"html":"Dataset used for reflection tuning.","text":"Dataset used for reflection tuning."},"author":"diffusion-cot","downloads":932,"gated":false,"id":"diffusion-cot/GenRef-wds","lastModified":"2025-04-24T19:22:11.000Z","datasetsServerInfo":{"viewer":"viewer-partial","numRows":1066717,"libraries":["datasets","webdataset","mlcroissant"],"formats":["webdataset"],"modalities":["image","text"]},"private":false,"repoType":"dataset","likes":14,"isLikedByUser":false,"isBenchmark":false},{"_id":"6803e14b10dede0a63cfdfaa","position":2,"type":"dataset","note":{"html":"Dataset used to fine-tune Qwen for reflection generation.","text":"Dataset used to fine-tune Qwen for reflection generation."},"author":"diffusion-cot","downloads":434,"gated":false,"id":"diffusion-cot/GenRef-CoT","lastModified":"2025-04-24T16:34:23.000Z","datasetsServerInfo":{"viewer":"viewer-partial","numRows":221092,"libraries":["datasets","webdataset","mlcroissant"],"formats":["webdataset"],"modalities":["image","text"]},"private":false,"repoType":"dataset","likes":3,"isLikedByUser":false,"isBenchmark":false},{"_id":"680847affc09ee563abb4540","position":3,"type":"model","note":{"html":"Reflection generation model based on Qwen.","text":"Reflection generation model based on Qwen."},"author":"diffusion-cot","authorData":{"_id":"67bdb50c08abc7f641f215a7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f7fbd813e94f16a85448745/u7-k61BeAabAW9S4AUm94.png","fullname":"Diffusion CoT","name":"diffusion-cot","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"plan":"team","followerCount":20,"isUserFollowing":false},"downloads":0,"gated":false,"id":"diffusion-cot/Reflection-Generator","availableInferenceProviders":[],"lastModified":"2025-04-23T15:51:03.000Z","likes":5,"private":false,"repoType":"model","isLikedByUser":false}],"position":1,"theme":"blue","private":false,"shareUrl":"https://hf.co/collections/diffusion-cot/reflectionflow-release","upvotes":13,"isUpvotedByUser":false}],"datasets":[{"author":"diffusion-cot","downloads":136,"gated":false,"id":"diffusion-cot/imgedit-simpler","lastModified":"2025-09-09T07:21:15.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":723776,"libraries":["datasets","dask","mlcroissant","polars"],"formats":["parquet"],"modalities":["image","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"diffusion-cot","downloads":66,"gated":false,"id":"diffusion-cot/echo-4o-instruction-following","lastModified":"2025-08-19T04:36:28.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":67958,"libraries":["datasets","dask","mlcroissant","polars"],"formats":["parquet"],"modalities":["image","text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"diffusion-cot","downloads":15,"gated":false,"id":"diffusion-cot/unified-test-data","lastModified":"2025-06-30T06:39:05.000Z","datasetsServerInfo":{"viewer":"viewer","numRows":174919,"libraries":["datasets","pandas","mlcroissant","polars"],"formats":["parquet"],"modalities":["text"]},"private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false},{"author":"diffusion-cot","downloads":932,"gated":false,"id":"diffusion-cot/GenRef-wds","lastModified":"2025-04-24T19:22:11.000Z","datasetsServerInfo":{"viewer":"viewer-partial","numRows":1066717,"libraries":["datasets","webdataset","mlcroissant"],"formats":["webdataset"],"modalities":["image","text"]},"private":false,"repoType":"dataset","likes":14,"isLikedByUser":false,"isBenchmark":false},{"author":"diffusion-cot","downloads":434,"gated":false,"id":"diffusion-cot/GenRef-CoT","lastModified":"2025-04-24T16:34:23.000Z","datasetsServerInfo":{"viewer":"viewer-partial","numRows":221092,"libraries":["datasets","webdataset","mlcroissant"],"formats":["webdataset"],"modalities":["image","text"]},"private":false,"repoType":"dataset","likes":3,"isLikedByUser":false,"isBenchmark":false},{"author":"diffusion-cot","downloads":1055,"gated":false,"id":"diffusion-cot/GenRef","lastModified":"2025-04-15T10:58:02.000Z","private":false,"repoType":"dataset","likes":0,"isLikedByUser":false,"isBenchmark":false}],"models":[{"author":"diffusion-cot","authorData":{"_id":"67bdb50c08abc7f641f215a7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f7fbd813e94f16a85448745/u7-k61BeAabAW9S4AUm94.png","fullname":"Diffusion CoT","name":"diffusion-cot","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"plan":"team","followerCount":20,"isUserFollowing":false},"downloads":0,"gated":false,"id":"diffusion-cot/kontext-stage1","availableInferenceProviders":[],"lastModified":"2025-08-12T06:43:30.000Z","likes":0,"private":false,"repoType":"model","isLikedByUser":false},{"author":"diffusion-cot","authorData":{"_id":"67bdb50c08abc7f641f215a7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f7fbd813e94f16a85448745/u7-k61BeAabAW9S4AUm94.png","fullname":"Diffusion CoT","name":"diffusion-cot","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"plan":"team","followerCount":20,"isUserFollowing":false},"downloads":0,"gated":false,"id":"diffusion-cot/FLUX-Corrector","availableInferenceProviders":[],"lastModified":"2025-04-23T15:52:01.000Z","likes":11,"private":false,"repoType":"model","isLikedByUser":false},{"author":"diffusion-cot","authorData":{"_id":"67bdb50c08abc7f641f215a7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f7fbd813e94f16a85448745/u7-k61BeAabAW9S4AUm94.png","fullname":"Diffusion CoT","name":"diffusion-cot","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"plan":"team","followerCount":20,"isUserFollowing":false},"downloads":0,"gated":false,"id":"diffusion-cot/Reflection-Generator","availableInferenceProviders":[],"lastModified":"2025-04-23T15:51:03.000Z","likes":5,"private":false,"repoType":"model","isLikedByUser":false},{"author":"diffusion-cot","authorData":{"_id":"67bdb50c08abc7f641f215a7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5f7fbd813e94f16a85448745/u7-k61BeAabAW9S4AUm94.png","fullname":"Diffusion CoT","name":"diffusion-cot","type":"org","isHf":false,"isHfAdmin":false,"isMod":false,"plan":"team","followerCount":20,"isUserFollowing":false},"downloads":0,"gated":false,"id":"diffusion-cot/Image-Verifier","availableInferenceProviders":[],"lastModified":"2025-04-23T15:49:17.000Z","likes":3,"private":false,"repoType":"model","isLikedByUser":false}],"paperPreviews":[],"spaces":[],"buckets":[],"numBuckets":0,"numDatasets":6,"numModels":4,"numSpaces":1,"lastOrgActivities":[{"time":"2026-02-20T08:43:36.837Z","user":"sayakpaul","userAvatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1649681653581-5f7fbd813e94f16a85448745.jpeg","type":"paper","paper":{"id":"2602.15449","title":"TAROT: Test-driven and Capability-adaptive Curriculum Reinforcement Fine-tuning for Code Generation with Large Language Models","publishedAt":"2026-02-17T09:29:18.000Z","upvotes":6,"isUpvotedByUser":true}},{"time":"2025-11-17T10:33:13.306Z","user":"JackyZhuo","userAvatarUrl":"","type":"paper","paper":{"id":"2504.04903","title":"Lumina-OmniLV: A Unified Multimodal Framework for General Low-Level\n Vision","publishedAt":"2025-04-07T10:22:00.000Z","upvotes":0,"isUpvotedByUser":false}},{"time":"2025-11-17T10:33:03.266Z","user":"JackyZhuo","userAvatarUrl":"","type":"paper","paper":{"id":"2510.05091","title":"Factuality Matters: When Image Generation and Editing Meet Structured\n Visuals","publishedAt":"2025-10-06T17:56:55.000Z","upvotes":20,"isUpvotedByUser":true}}],"acceptLanguages":["*"],"canReadRepos":false,"canReadSpaces":false,"blogPosts":[],"currentRepoPage":0,"filters":{},"paperView":false}">

AI & ML interests

diffusion

Recent Activity

sayakpaul  authored a paper about 11 hours ago
TAROT: Test-driven and Capability-adaptive Curriculum Reinforcement Fine-tuning for Code Generation with Large Language Models
JackyZhuo  authored a paper 3 months ago
Lumina-OmniLV: A Unified Multimodal Framework for General Low-Level Vision
JackyZhuo  authored a paper 3 months ago
Factuality Matters: When Image Generation and Editing Meet Structured Visuals
View all activity

Organization Card
Community About org cards

This organization holds the artifacts for our research conducted on enabling reasoning in diffusion-based image synthesis models. Our first effort in this line of research is ReflectionFlow, where we introduce the first ever large-scale dataset, GenRef, suitable for reflection-tuning.

Below, we provide the links related to ReflectionFlow:

Citation

misc{zhuo2025reflectionperfectionscalinginferencetime,
      title={From Reflection to Perfection: Scaling Inference-Time Optimization for Text-to-Image Diffusion Models via Reflection Tuning}, 
      author={Le Zhuo and Liangbing Zhao and Sayak Paul and Yue Liao and Renrui Zhang and Yi Xin and Peng Gao and Mohamed Elhoseiny and Hongsheng Li},
      year={2025},
      eprint={2504.16080},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2504.16080}, 
}

Enjoy 🤗

Collections 1