AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

DisRM: Reward Modeling as Discriminative Prediction

Runtao Liu, Jiahao Zhan, Yuxuan GUO, Yingqing He, Chen Wei, Alan Yuille, Qifeng Chen

ECCV 2026 · Image generation › Text-to-image diffusion · approach: Preference optimization for diffusion · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
select best of 10 samples with DisRMSD1.5 · JourneyDBFID: 72.06 -> 68.51Improved
fine-tune with DisRM-selected samples (SFT)SD1.5 · JourneyDBFID: 72.06 -> 64.98Improved
DPO with DisRM-built pairsSD1.5 · JourneyDBFID: 72.06 -> 63.61Improved
DPO with DisRM pairs instead of DiffusionDPO 1M labeled pairsSD1.5 · JourneyDBFID: 68.15 -> 63.61Improved
DisRM DPO on SDXLSDXL · JourneyDBFID: 62.83 -> 61.95Mixed
DisRM selection on SDXLSDXL · JourneyDBPickScore: 21.235 -> 21.311No clear effect
DPO with DisRM using safe PPDSD1.5 · safety COCOIP: 42 -> 18Improved
DPO with DisRM using safe PPDSDXL · safety COCOIP: 51 -> 17Improved
increase PPD size from 2K to 15K for selection (ablation)SD1.5 · safety COCOIP: 35 -> 34No clear effect
DPO with DisRM on VideoCrafter2VideoCrafter2 · VBenchVBench: 80.44 -> 81.38Improved
add rank-based bootstrapping instead of naive single checkpoint (ablation)DisRM · selection on JourneyDBFID: 14.48 -> 14.18Improved
annotate DPO pairs with DisRM instead of ImageRewardSD1.5 · 10k-pair DPOFID: 72.59 -> 67.42Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy