Papers › NoHumansRequired: Autonomous High-Quality Image Editing Triplet Mining

NoHumansRequired: Autonomous High-Quality Image Editing Triplet Mining

18 Jul 2025arXiv:2507.14119archive 2025-07-28

Maksim Kuprashevich, Grigorii Alekseenko, Irina Tolstykh, Georgii Fedorov, Bulat Suleimanov, Vladimir Dokholyan, Aleksandr Gordeev

Recent advances in generative modeling enable image editing assistants that follow natural language instructions without additional user input. Their supervised training requires millions of triplets: original image, instruction, edited image. Yet mining pixel-accurate examples is hard. Each edit must affect only prompt-specified regions, preserve stylistic coherence, respect physical plausibility, and retain visual appeal. The lack of robust automated edit-quality metrics hinders reliable automation at scale. We present an automated, modular pipeline that mines high-fidelity triplets across domains, resolutions, instruction complexities, and styles. Built on public generative models and running without human intervention, our system uses a task-tuned Gemini validator to score instruction adherence and aesthetics directly, removing any need for segmentation or grounding models. Inversion and compositional bootstrapping enlarge the mined set by approximately 2.2x, enabling large-scale high-fidelity training data. By automating the most repetitive annotation steps, the approach allows a new scale of training without human labeling effort. To democratize research in this resource-intensive area, we release NHR-Edit: an open dataset of 358k high-quality triplets. In the largest cross-dataset evaluation, it surpasses all public alternatives. We also release Bagel-NHR-Edit, an open-source fine-tuned Bagel model, which achieves state-of-the-art metrics in our experiments.

PaperPDF

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image EditingText-based Image Editing

1 archive task tag without a task page not shown.

Datasets

Introduced by this paper, per the archive.

NHR-Edit

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Editing GEdit-Bench-EN BAGEL-NHR-EDIT Overall 7.12 #1 of 3 Archive leaderboard report
Image Editing GEdit-Bench-EN BAGEL-NHR-EDIT Perceptual Quality 6.88 #1 of 3 Archive leaderboard report
Image Editing GEdit-Bench-EN BAGEL-NHR-EDIT Semantic Consistency 8.07 #1 of 3 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Action 3.95 #1 of 9 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Add 4.19 #1 of 9 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Adjust 3.55 #1 of 9 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Background 3.42 #1 of 9 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Extract 1.62 #1 of 9 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Hybrid 2.94 #1 of 9 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Overall 3.39 #1 of 9 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Remove 3.18 #1 of 9 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Replace 3.77 #1 of 9 Archive leaderboard report
Image Editing ImgEdit-Data BAGEL-NHR-EDIT Style 4.3 #1 of 9 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections