https://github.com/miraflow/DistilDIRE

\n","updatedAt":"2024-07-25T22:46:56.578Z","author":{"_id":"60f1abe7544c2adfd699860c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674929746905-60f1abe7544c2adfd699860c.jpeg","fullname":"AK","name":"akhaliq","type":"user","isPro":false,"isHf":true,"isHfAdmin":false,"isMod":false,"followerCount":8244}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.5971387624740601},"editors":["akhaliq"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674929746905-60f1abe7544c2adfd699860c.jpeg"],"reactions":[{"reaction":"❤️","users":["aerinkim","yevvonlim"],"count":2},{"reaction":"🔥","users":["aerinkim","yevvonlim"],"count":2},{"reaction":"👍","users":["aerinkim","yevvonlim"],"count":2}],"isReport":false}},{"id":"66a2fc7ca74b20aad76bc131","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":264},"createdAt":"2024-07-26T01:31:40.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"This is an automated message from the [Librarian Bot](https://huggingface.co/librarian-bots). I found the following papers similar to this paper. \n\nThe following papers were recommended by the Semantic Scholar API \n\n* [Accelerating Diffusion for SAR-to-Optical Image Translation via Adversarial Consistency Distillation](https://huggingface.co/papers/2407.06095) (2024)\n* [Adv-KD: Adversarial Knowledge Distillation for Faster Diffusion Sampling](https://huggingface.co/papers/2405.20675) (2024)\n* [Latent Denoising Diffusion GAN: Faster Sampling, Higher Image Quality](https://huggingface.co/papers/2406.11713) (2024)\n* [RIGID: A Training-free and Model-Agnostic Framework for Robust AI-Generated Image Detection](https://huggingface.co/papers/2405.20112) (2024)\n* [Real-Time Deepfake Detection in the Real-World](https://huggingface.co/papers/2406.09398) (2024)\n\n\n Please give a thumbs up to this comment if you found it helpful!\n\n If you want recommendations for any Paper on Hugging Face checkout [this](https://huggingface.co/spaces/librarian-bots/recommend_similar_papers) Space\n\n You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: `@librarian-bot recommend`","html":"

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

\n

The following papers were recommended by the Semantic Scholar API

\n

\n
• Accelerating Diffusion for SAR-to-Optical Image Translation via Adversarial Consistency Distillation (2024)
\n
• Adv-KD: Adversarial Knowledge Distillation for Faster Diffusion Sampling (2024)
\n
• Latent Denoising Diffusion GAN: Faster Sampling, Higher Image Quality (2024)
\n
• RIGID: A Training-free and Model-Agnostic Framework for Robust AI-Generated Image Detection (2024)
\n
• Real-Time Deepfake Detection in the Real-World (2024)
\n

\n

Please give a thumbs up to this comment if you found it helpful!

\n

If you want recommendations for any Paper on Hugging Face checkout this Space

\n

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2024-07-26T01:31:40.402Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":264}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7235991358757019},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[{"reaction":"👍","users":["aerinkim"],"count":1}],"isReport":false}},{"id":"66a3fd4df5da61319f693583","author":{"_id":"630629d9660f01f150a110d3","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/630629d9660f01f150a110d3/-9_cmjec1aEdX_1BKosCk.png","fullname":"aerin kim","name":"aerinkim","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":2},"createdAt":"2024-07-26T19:47:25.000Z","type":"comment","data":{"edited":true,"hidden":false,"latest":{"raw":"Using the distilled version of \"reconstruction then compare\" method, Distil-DIRE (Diffusion Reconstruction Error) significantly speeds up the detection while maintaining accuracy.\n\nTry our deepfake detector:\nhttps://detect.truemedia.org ✅\n\n\n","html":"

Using the distilled version of \"reconstruction then compare\" method, Distil-DIRE (Diffusion Reconstruction Error) significantly speeds up the detection while maintaining accuracy.

\n

Try our deepfake detector:
https://detect.truemedia.org ✅

\n

\n","updatedAt":"2024-07-29T17:37:34.634Z","author":{"_id":"630629d9660f01f150a110d3","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/630629d9660f01f150a110d3/-9_cmjec1aEdX_1BKosCk.png","fullname":"aerin kim","name":"aerinkim","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":2}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.5373315215110779},"editors":["aerinkim"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/630629d9660f01f150a110d3/-9_cmjec1aEdX_1BKosCk.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2406.00856","authors":[{"_id":"66a2cb4884301ad74809dbac","user":{"_id":"64c224cad1ca220e304248bb","avatarUrl":"/avatars/2e7ff92cec355db17a83dfc1cac6c2e8.svg","isPro":false,"fullname":"Yewon Lim","user":"yevvonlim","type":"user"},"name":"Yewon Lim","status":"claimed_verified","statusLastChangedAt":"2024-10-14T08:33:55.245Z","hidden":false},{"_id":"66a2cb4884301ad74809dbad","name":"Changyeon Lee","hidden":false},{"_id":"66a2cb4884301ad74809dbae","user":{"_id":"630629d9660f01f150a110d3","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/630629d9660f01f150a110d3/-9_cmjec1aEdX_1BKosCk.png","isPro":false,"fullname":"aerin kim","user":"aerinkim","type":"user"},"name":"Aerin Kim","status":"claimed_verified","statusLastChangedAt":"2024-07-26T07:40:21.978Z","hidden":false},{"_id":"66a2cb4884301ad74809dbaf","name":"Oren Etzioni","hidden":false}],"publishedAt":"2024-06-02T20:22:38.000Z","submittedOnDailyAt":"2024-07-25T21:16:56.571Z","title":"DistilDIRE: A Small, Fast, Cheap and Lightweight Diffusion Synthesized\n Deepfake Detection","submittedOnDailyBy":{"_id":"60f1abe7544c2adfd699860c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674929746905-60f1abe7544c2adfd699860c.jpeg","isPro":false,"fullname":"AK","user":"akhaliq","type":"user"},"summary":"A dramatic influx of diffusion-generated images has marked recent years,\nposing unique challenges to current detection technologies. While the task of\nidentifying these images falls under binary classification, a seemingly\nstraightforward category, the computational load is significant when employing\nthe \"reconstruction then compare\" technique. This approach, known as DIRE\n(Diffusion Reconstruction Error), not only identifies diffusion-generated\nimages but also detects those produced by GANs, highlighting the technique's\nbroad applicability. To address the computational challenges and improve\nefficiency, we propose distilling the knowledge embedded in diffusion models to\ndevelop rapid deepfake detection models. Our approach, aimed at creating a\nsmall, fast, cheap, and lightweight diffusion synthesized deepfake detector,\nmaintains robust performance while significantly reducing operational demands.\nMaintaining performance, our experimental results indicate an inference speed\n3.2 times faster than the existing DIRE framework. This advance not only\nenhances the practicality of deploying these systems in real-world settings but\nalso paves the way for future research endeavors that seek to leverage\ndiffusion model knowledge.","upvotes":12,"discussionId":"66a2cb4984301ad74809dbe7","ai_summary":"Knowledge distillation from diffusion models is used to develop a more efficient and lightweight deepfake detector compared to the DIRE framework.","ai_keywords":["diffusion-generated images","binary classification","reconstruction then compare","DIRE","diffusion models","deepfake detection","knowledge distillation","rapid detection"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"630629d9660f01f150a110d3","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/630629d9660f01f150a110d3/-9_cmjec1aEdX_1BKosCk.png","isPro":false,"fullname":"aerin kim","user":"aerinkim","type":"user"},{"_id":"628ac0cfc204ff147827e56b","avatarUrl":"/avatars/dcf7c089fe5530daf3e79d542481e43c.svg","isPro":false,"fullname":"Sejin Paik","user":"sjp","type":"user"},{"_id":"64c224cad1ca220e304248bb","avatarUrl":"/avatars/2e7ff92cec355db17a83dfc1cac6c2e8.svg","isPro":false,"fullname":"Yewon Lim","user":"yevvonlim","type":"user"},{"_id":"64f9e36dc3fec440b20701cf","avatarUrl":"/avatars/6abd29ff49151f22976d6a2a3372f401.svg","isPro":false,"fullname":"Changyeon Lee","user":"cycycy1","type":"user"},{"_id":"6576cc45addd9129ffbc56f2","avatarUrl":"/avatars/9742ecbaee9dcc487abdd2d310f48052.svg","isPro":false,"fullname":"Arnab Karmakar","user":"arnabk1","type":"user"},{"_id":"66897e8457fd09c47d11dc28","avatarUrl":"/avatars/ca8fb71afe84c638f448d9c17f023e29.svg","isPro":false,"fullname":"Anton Brooks","user":"ANTONBROOKS","type":"user"},{"_id":"6689821cbcbb6192a42f6a34","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6689821cbcbb6192a42f6a34/blB7dpsf5H3uh4lRhpgkT.jpeg","isPro":false,"fullname":"Luciano Prutt","user":"thebluehedgehog","type":"user"},{"_id":"64ba972e90dfdda6ab7fbf54","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64ba972e90dfdda6ab7fbf54/631OMPITxvxH6u4sGhoHa.jpeg","isPro":false,"fullname":"Santosh Patapati","user":"Soontosh","type":"user"},{"_id":"6538119803519fddb4a17e10","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6538119803519fddb4a17e10/ffJMkdx-rM7VvLTCM6ri_.jpeg","isPro":false,"fullname":"samusenps","user":"samusenps","type":"user"},{"_id":"641b754d1911d3be6745cce9","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/641b754d1911d3be6745cce9/DxjZG1XT4H3ZHF7qHxWxk.jpeg","isPro":true,"fullname":"atayloraerospace","user":"Taylor658","type":"user"},{"_id":"64169a99bce2fed80ab86122","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1679202958868-noauth.jpeg","isPro":false,"fullname":"Sigrid Jin","user":"sigridjineth","type":"user"},{"_id":"663ccbff3a74a20189d4aa2e","avatarUrl":"/avatars/83a54455e0157480f65c498cd9057cf2.svg","isPro":false,"fullname":"Nguyen Van Thanh","user":"NguyenVanThanhHust","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">