Pith. sign in

REVIEW 8 cited by

MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2407.04842 v1 pith:EUHJMUW3 submitted 2024-07-05 cs.CV cs.CLcs.LG

MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?

classification cs.CV cs.CLcs.LG
keywords feedbackmodelsjudgesmultimodalvlmsmj-benchbiasgeneration
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

While text-to-image models like DALLE-3 and Stable Diffusion are rapidly proliferating, they often encounter challenges such as hallucination, bias, and the production of unsafe, low-quality output. To effectively address these issues, it is crucial to align these models with desired behaviors based on feedback from a multimodal judge. Despite their significance, current multimodal judges frequently undergo inadequate evaluation of their capabilities and limitations, potentially leading to misalignment and unsafe fine-tuning outcomes. To address this issue, we introduce MJ-Bench, a novel benchmark which incorporates a comprehensive preference dataset to evaluate multimodal judges in providing feedback for image generation models across four key perspectives: alignment, safety, image quality, and bias. Specifically, we evaluate a large variety of multimodal judges including smaller-sized CLIP-based scoring models, open-source VLMs (e.g. LLaVA family), and close-source VLMs (e.g. GPT-4o, Claude 3) on each decomposed subcategory of our preference dataset. Experiments reveal that close-source VLMs generally provide better feedback, with GPT-4o outperforming other judges in average. Compared with open-source VLMs, smaller-sized scoring models can provide better feedback regarding text-image alignment and image quality, while VLMs provide more accurate feedback regarding safety and generation bias due to their stronger reasoning capabilities. Further studies in feedback scale reveal that VLM judges can generally provide more accurate and stable feedback in natural language (Likert-scale) than numerical scales. Notably, human evaluations on end-to-end fine-tuned models using separate feedback from these multimodal judges provide similar conclusions, further confirming the effectiveness of MJ-Bench. All data, code, models are available at https://huggingface.co/MJ-Bench.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Z-Reward: Beyond Scalar Rewards by Internalizing Reasoning into Score Distributions

    cs.CV 2026-06 conditional novelty 6.0

    A teacher-student reward model learns reasoning-conditioned score distributions for text-to-image images, yielding ~89% preference accuracy and a 41% net human-preference gain when used for generator optimization.

  2. Z-Reward: Beyond Scalar Rewards by Internalizing Reasoning into Score Distributions

    cs.CV 2026-06 unverdicted novelty 6.0

    Z-Reward trains a 27B reasoning teacher VLM on score distributions via GDSO and distills it via RISD into a 9B student, reaching 89.6% and 88.6% human preference accuracy with 41.3% optimization gain over SFT baseline.

  3. DiRotQ: Rotation-Aware Quantization for 4-bit Diffusion Transformers

    cs.CV 2026-05 unverdicted novelty 6.0

    DiRotQ uses PCA-based rotation-aware activation quantization combined with GPTQ to achieve better FID and PSNR in 4-bit diffusion transformers than prior methods like SVDQuant.

  4. RewardBench 2: Advancing Reward Model Evaluation

    cs.CL 2025-06 unverdicted novelty 6.0

    RewardBench 2 is a new benchmark that supplies challenging fresh human prompts for reward model evaluation, yielding lower average scores but higher correlation with downstream best-of-N sampling and RLHF training per...

  5. Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps

    cs.CV 2025-01 conditional novelty 6.0

    Diffusion models improve generation quality via inference-time search over noise candidates guided by verifiers and algorithms, yielding gains beyond denoising step scaling on class- and text-conditioned benchmarks.

  6. VisionReward: Fine-Grained Multi-Dimensional Human Preference Learning for Image and Video Generation

    cs.CV 2024-12 unverdicted novelty 6.0

    VisionReward learns multi-dimensional human preferences for image and video generation via hierarchical assessment and linear weighting, outperforming VideoScore by 17.2% in prediction accuracy and yielding 31.6% high...

  7. Multi-Task Reinforcement Learning for Enhanced Multimodal LLM-as-a-Judge

    cs.CL 2026-03 unverdicted novelty 4.0

    MT-RL-Judge uses multi-task RL to train MLLM judges that outperform baselines in consistency, human preference correlation, and out-of-distribution generalization.

  8. Reinforcement Learning from Human Feedback

    cs.LG 2025-04 unverdicted novelty 2.0

    The book introduces the origins, mathematical setup, and optimization stages of RLHF including reward modeling, reinforcement learning, rejection sampling, and direct alignment algorithms.