Skip to content
← Back to explorer

EditReward: A Human-Aligned Reward Model for Instruction-Guided Image Editing

Keming Wu, Sicong Jiang, Max Ku, Ping Nie, Minghao Liu, Wenhu Chen · Sep 30, 2025 · Citations: 0

General Llm As Judge Pairwise Preference
Open arXiv Find Implementation RSS feed Shortlist (0)

Data freshness

Extraction: Fresh

Check recency before relying on this page for active eval decisions. Use stale pages as context and verify against current hub results.

Metadata refreshed

Feb 28, 2026, 6:55 AM

Recent

Extraction refreshed

Mar 8, 2026, 2:53 AM

Fresh

Extraction source

Persisted extraction

Confidence 0.75

Abstract

Recently, we have witnessed great progress in image editing with natural language instructions. Several closed-source models like GPT-Image-1, Seedream, and Google-Nano-Banana have shown highly promising progress. However, the open-source models are still lagging. The main bottleneck is the lack of a reliable reward model to scale up high-quality synthetic training data. To address this critical bottleneck, we built EditReward, trained with our new large-scale human preference dataset, meticulously annotated by trained experts following a rigorous protocol containing over 200K preference pairs. EditReward demonstrates superior alignment with human preferences in instruction-guided image editing tasks. Experiments show that EditReward achieves state-of-the-art human correlation on established benchmarks such as GenAI-Bench, AURORA-Bench, ImagenHub, and our new EditReward-Bench, outperforming a wide range of VLM-as-judge models. Furthermore, we use EditReward to select a high-quality subset from the existing noisy ShareGPT-4o-Image dataset. We train Step1X-Edit on the selected subset, which shows significant improvement over training on the full set. This demonstrates EditReward's ability to serve as a reward model to scale up high-quality training data for image editing. Furthermore, its strong alignment suggests potential for advanced applications like reinforcement learning-based post-training and test-time scaling of image editing models. EditReward with its training dataset will be released to help the community build more high-quality image editing training datasets.

HFEPX Relevance Assessment

This paper has useful evaluation signal, but protocol completeness is partial; pair it with related papers before deciding implementation strategy.

Best use

Secondary protocol comparison source

Use if you need

A secondary eval reference to pair with stronger protocol papers.

Main weakness

No major weakness surfaced.

Trust level

High

Eval-Fit Score

67/100 • Medium

Useful as a secondary reference; validate protocol details against neighboring papers.

Human Feedback Signal

Detected

Evaluation Signal

Detected

HFEPX Fit

Moderate-confidence candidate

Extraction confidence: High

If you are doing eval pipeline work, start here:

Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub Tool-Use Eval Hub

Field Provenance & Confidence

Each key protocol field shows extraction state, confidence band, and data source so you can decide whether to trust it directly or validate from full text.

Human Feedback Types

strong

Pairwise Preference

Confidence: High Source: Persisted extraction evidenced

Directly usable for protocol triage.

Evidence snippet: Recently, we have witnessed great progress in image editing with natural language instructions.

Evaluation Modes

strong

Llm As Judge

Confidence: High Source: Persisted extraction evidenced

Includes extracted eval setup.

Evidence snippet: Recently, we have witnessed great progress in image editing with natural language instructions.

Quality Controls

missing

Not reported

Confidence: Low Source: Persisted extraction missing

No explicit QC controls found.

Evidence snippet: Recently, we have witnessed great progress in image editing with natural language instructions.

Benchmarks / Datasets

strong

Genai Bench, Aurora Bench, Editreward Bench

Confidence: High Source: Persisted extraction evidenced

Useful for quick benchmark comparison.

Evidence snippet: Experiments show that EditReward achieves state-of-the-art human correlation on established benchmarks such as GenAI-Bench, AURORA-Bench, ImagenHub, and our new EditReward-Bench, outperforming a wide range of VLM-as-judge models.

Reported Metrics

missing

Not extracted

Confidence: Low Source: Persisted extraction missing

No metric anchors detected.

Evidence snippet: Recently, we have witnessed great progress in image editing with natural language instructions.

Rater Population

strong

Domain Experts

Confidence: High Source: Persisted extraction evidenced

Helpful for staffing comparability.

Evidence snippet: To address this critical bottleneck, we built EditReward, trained with our new large-scale human preference dataset, meticulously annotated by trained experts following a rigorous protocol containing over 200K preference pairs.

Human Data Lens

  • Uses human feedback: Yes
  • Feedback types: Pairwise Preference
  • Rater population: Domain Experts
  • Unit of annotation: Unknown
  • Expertise required: General
  • Extraction source: Persisted extraction

Evaluation Lens

  • Evaluation modes: Llm As Judge
  • Agentic eval: None
  • Quality controls: Not reported
  • Confidence: 0.75
  • Flags: runtime_fallback_extraction

Protocol And Measurement Signals

Benchmarks / Datasets

Genai-BenchAurora-BenchEditreward-Bench

Reported Metrics

No metric terms were extracted from the available abstract.

Research Brief

Deterministic synthesis

To address this critical bottleneck, we built EditReward, trained with our new large-scale human preference dataset, meticulously annotated by trained experts following a rigorous protocol containing over 200K preference pairs. HFEPX signals include Pairwise Preference, Llm As Judge with confidence 0.75. Updated from current HFEPX corpus.

Generated Mar 8, 2026, 2:53 AM · Grounded in abstract + metadata only

Key Takeaways

  • To address this critical bottleneck, we built EditReward, trained with our new large-scale human preference dataset, meticulously annotated by trained experts following a rigorous…
  • EditReward demonstrates superior alignment with human preferences in instruction-guided image editing tasks.

Researcher Actions

  • Compare its human-feedback setup against pairwise and rubric hubs.
  • Cross-check benchmark overlap: Genai-Bench, Aurora-Bench, Editreward-Bench.
  • Verify metric definitions before comparing against your eval pipeline.

Caveats

  • Generated from title, abstract, and extracted metadata only; full-paper implementation details are not parsed.
  • Extraction confidence is probabilistic and should be validated for critical decisions.

Research Summary

Contribution Summary

  • To address this critical bottleneck, we built EditReward, trained with our new large-scale human preference dataset, meticulously annotated by trained experts following a rigorous protocol containing over 200K preference pairs.
  • EditReward demonstrates superior alignment with human preferences in instruction-guided image editing tasks.
  • Experiments show that EditReward achieves state-of-the-art human correlation on established benchmarks such as GenAI-Bench, AURORA-Bench, ImagenHub, and our new EditReward-Bench, outperforming a wide range of VLM-as-judge models.

Why It Matters For Eval

  • To address this critical bottleneck, we built EditReward, trained with our new large-scale human preference dataset, meticulously annotated by trained experts following a rigorous protocol containing over 200K preference pairs.
  • EditReward demonstrates superior alignment with human preferences in instruction-guided image editing tasks.

Researcher Checklist

  • Pass: Human feedback protocol is explicit

    Detected: Pairwise Preference

  • Pass: Evaluation mode is explicit

    Detected: Llm As Judge

  • Gap: Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Pass: Benchmark or dataset anchors are present

    Detected: Genai-Bench, Aurora-Bench, Editreward-Bench

  • Gap: Metric reporting is present

    No metric terms extracted.

Related Papers

Papers are ranked by protocol overlap, extraction signal alignment, and semantic proximity.

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.

Post a Job Get a Quote