Hugging Face Daily Papers · · 5 min read

Coloring the Noise: Adversarial Sobolev Alignment for Faithful Image Super Resolution

Mirrored from Hugging Face Daily Papers for archival readability. Support the source by reading on the original site.

Excited to share our ICML 2026 paper, ASASR: Coloring the Noise for Faithful Image Super-Resolution.</p>\n<p>We believe faithful image super-resolution requires more than realistic textures: it requires spectral and structural alignment with natural images. ASASR explores this idea by coloring the noise and using adversarial Sobolev guidance to reduce hallucinations and preserve fine details.</p>\n","updatedAt":"2026-05-26T07:51:31.722Z","author":{"_id":"6a1547ddbdbb2f53b93ab213","avatarUrl":"/avatars/bf59ca5c69a9cdfecbb0f624bfe0562d.svg","fullname":"Hongbo Wang","name":"wafer-bob","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8269056081771851},"editors":["wafer-bob"],"editorAvatarUrls":["/avatars/bf59ca5c69a9cdfecbb0f624bfe0562d.svg"],"reactions":[],"isReport":false}},{"id":"6a15ff8006732f4fc18e29d9","author":{"_id":"65243980050781c16f234f1f","avatarUrl":"/avatars/743a009681d5d554c27e04300db9f267.svg","fullname":"Avi","name":"avahal","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":6,"isUserFollowing":false},"createdAt":"2026-05-26T20:16:00.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"the part that really sticks is how they recenter sr optimization around a Sobolev geometry by coloring the noise with a structured covariance. the adversarial module, grounded in the Riesz representation, produces worst-case Sobolev gradients to steer updates along the tangent space of plausible structures, which feels like a nontrivial shift from standard flow matching. i wonder how sensitive the gains are to the choice of the Sobolev exponent and the exact noise coloring, since natural image spectra can swing across datasets. the arxivlens breakdown helped me parse the method details and sanity-check the equations, worth a skim alongside the paper (https://arxivlens.com/PaperView/Details/coloring-the-noise-adversarial-sobolev-alignment-for-faithful-image-super-resolution-4691-db59c1cc).","html":"<p>the part that really sticks is how they recenter sr optimization around a Sobolev geometry by coloring the noise with a structured covariance. the adversarial module, grounded in the Riesz representation, produces worst-case Sobolev gradients to steer updates along the tangent space of plausible structures, which feels like a nontrivial shift from standard flow matching. i wonder how sensitive the gains are to the choice of the Sobolev exponent and the exact noise coloring, since natural image spectra can swing across datasets. the arxivlens breakdown helped me parse the method details and sanity-check the equations, worth a skim alongside the paper (<a href=\"https://arxivlens.com/PaperView/Details/coloring-the-noise-adversarial-sobolev-alignment-for-faithful-image-super-resolution-4691-db59c1cc\" rel=\"nofollow\">https://arxivlens.com/PaperView/Details/coloring-the-noise-adversarial-sobolev-alignment-for-faithful-image-super-resolution-4691-db59c1cc</a>).</p>\n","updatedAt":"2026-05-26T20:16:00.489Z","author":{"_id":"65243980050781c16f234f1f","avatarUrl":"/avatars/743a009681d5d554c27e04300db9f267.svg","fullname":"Avi","name":"avahal","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":6,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8544911742210388},"editors":["avahal"],"editorAvatarUrls":["/avatars/743a009681d5d554c27e04300db9f267.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.23264","authors":[{"_id":"6a15492fb57a1823d5708c5b","user":{"_id":"6a1547ddbdbb2f53b93ab213","avatarUrl":"/avatars/bf59ca5c69a9cdfecbb0f624bfe0562d.svg","isPro":false,"fullname":"Hongbo Wang","user":"wafer-bob","type":"user","name":"wafer-bob"},"name":"Hongbo Wang","status":"claimed_verified","statusLastChangedAt":"2026-05-26T07:47:38.499Z","hidden":false},{"_id":"6a15492fb57a1823d5708c5c","name":"Huaibo Huang","hidden":false},{"_id":"6a15492fb57a1823d5708c5d","name":"Pin Wang","hidden":false},{"_id":"6a15492fb57a1823d5708c5e","name":"Jinhua Hao","hidden":false},{"_id":"6a15492fb57a1823d5708c5f","name":"Chao Zhou","hidden":false},{"_id":"6a15492fb57a1823d5708c60","name":"Ran He","hidden":false}],"publishedAt":"2026-05-22T00:00:00.000Z","submittedOnDailyAt":"2026-05-26T00:00:00.000Z","title":"Coloring the Noise: Adversarial Sobolev Alignment for Faithful Image Super Resolution","submittedOnDailyBy":{"_id":"6a1547ddbdbb2f53b93ab213","avatarUrl":"/avatars/bf59ca5c69a9cdfecbb0f624bfe0562d.svg","isPro":false,"fullname":"Hongbo Wang","user":"wafer-bob","type":"user","name":"wafer-bob"},"summary":"Generative priors in Image Super-Resolution (SR) often compromise faithful restoration, we attribute this limitation to a fundamental spectral misalignment between isotropic objectives and the intrinsic natural image manifold. While Direct Preference Optimization offers a path to alignment, its reliance on spectrally flat Gaussian noise fails to distinguish authentic high-frequency details from hallucinations. To bridge this geometric gap, we propose ASASR, a theoretically grounded framework that recasts the generative flow into a Sobolev-induced Riemannian geometry by explicitly coloring the noise transition kernel to mirror natural spectral decay. Driving this geometric alignment, we integrate a parametric adversary grounded in the Riesz Representation Theorem, which synthesizes targeted negative samples equivalent to worst-case Sobolev gradients to direct optimization along the tangent space of plausible structural failures. Extensive evaluations demonstrate that ASASR outperforms leading generative baselines, particularly in preserving spectral consistency and structural fidelity, offering a robust solution that effectively mitigates artifacts.","upvotes":3,"discussionId":"6a15492fb57a1823d5708c61","githubRepo":"https://github.com/wafer-bob/ASASR","githubRepoAddedBy":"user","ai_summary":"ASASR addresses spectral misalignment in image super-resolution by leveraging Riemannian geometry and adversarial training to improve structural fidelity and reduce artifacts.","ai_keywords":["generative priors","image super-resolution","spectral misalignment","isotropic objectives","natural image manifold","Direct Preference Optimization","spectrally flat Gaussian noise","Sobolev-induced Riemannian geometry","noise transition kernel","natural spectral decay","parametric adversary","Riesz Representation Theorem","worst-case Sobolev gradients","tangent space","structural fidelity","artifacts"],"githubStars":71,"organization":{"_id":"640a887796aae649741a586f","name":"CASIA","fullname":"Chinese Academic of Science Institute of Automation","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1678411888885-6388984e8a5dbe2f3dc5afee.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6a1547ddbdbb2f53b93ab213","avatarUrl":"/avatars/bf59ca5c69a9cdfecbb0f624bfe0562d.svg","isPro":false,"fullname":"Hongbo Wang","user":"wafer-bob","type":"user"},{"_id":"695f8d5d9e778cf056e56017","avatarUrl":"/avatars/a1ec00149303e4c69e43821d2ee43218.svg","isPro":false,"fullname":"Yusong Lin","user":"x1aoche","type":"user"},{"_id":"65518b9e685ba4c13dac65dc","avatarUrl":"/avatars/c631c84b076ed1b975583662a05550e2.svg","isPro":false,"fullname":"Dai Fang","user":"DaiFang1023","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"640a887796aae649741a586f","name":"CASIA","fullname":"Chinese Academic of Science Institute of Automation","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1678411888885-6388984e8a5dbe2f3dc5afee.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.23264.md"}">
Papers
arxiv:2605.23264

Coloring the Noise: Adversarial Sobolev Alignment for Faithful Image Super Resolution

Published on May 22
· Submitted by
Hongbo Wang
on May 26
Authors:
,
,
,
,

Abstract

ASASR addresses spectral misalignment in image super-resolution by leveraging Riemannian geometry and adversarial training to improve structural fidelity and reduce artifacts.

AI-generated summary

Generative priors in Image Super-Resolution (SR) often compromise faithful restoration, we attribute this limitation to a fundamental spectral misalignment between isotropic objectives and the intrinsic natural image manifold. While Direct Preference Optimization offers a path to alignment, its reliance on spectrally flat Gaussian noise fails to distinguish authentic high-frequency details from hallucinations. To bridge this geometric gap, we propose ASASR, a theoretically grounded framework that recasts the generative flow into a Sobolev-induced Riemannian geometry by explicitly coloring the noise transition kernel to mirror natural spectral decay. Driving this geometric alignment, we integrate a parametric adversary grounded in the Riesz Representation Theorem, which synthesizes targeted negative samples equivalent to worst-case Sobolev gradients to direct optimization along the tangent space of plausible structural failures. Extensive evaluations demonstrate that ASASR outperforms leading generative baselines, particularly in preserving spectral consistency and structural fidelity, offering a robust solution that effectively mitigates artifacts.

Community

Paper author Paper submitter about 17 hours ago

Excited to share our ICML 2026 paper, ASASR: Coloring the Noise for Faithful Image Super-Resolution.

We believe faithful image super-resolution requires more than realistic textures: it requires spectral and structural alignment with natural images. ASASR explores this idea by coloring the noise and using adversarial Sobolev guidance to reduce hallucinations and preserve fine details.

the part that really sticks is how they recenter sr optimization around a Sobolev geometry by coloring the noise with a structured covariance. the adversarial module, grounded in the Riesz representation, produces worst-case Sobolev gradients to steer updates along the tangent space of plausible structures, which feels like a nontrivial shift from standard flow matching. i wonder how sensitive the gains are to the choice of the Sobolev exponent and the exact noise coloring, since natural image spectra can swing across datasets. the arxivlens breakdown helped me parse the method details and sanity-check the equations, worth a skim alongside the paper (https://arxivlens.com/PaperView/Details/coloring-the-noise-adversarial-sobolev-alignment-for-faithful-image-super-resolution-4691-db59c1cc).

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images

· Sign up or log in to comment

Get this paper in your agent:

hf papers read 2605.23264
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2605.23264 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2605.23264 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2605.23264 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

Discussion (0)

Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.

Sign in →

No comments yet. Sign in and be the first to say something.

More from Hugging Face Daily Papers