Pith. sign in

REVIEW 10 cited by

MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark

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 2402.04788 v3 pith:5D6NBOWV submitted 2024-02-07 cs.CL cs.AIcs.CV

classification cs.CLcs.AIcs.CV
keywords mllmsmultimodalassessingbatchbenchmarkchallengescomparisondiverse
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Multimodal Large Language Models (MLLMs) have gained significant attention recently, showing remarkable potential in artificial general intelligence. However, assessing the utility of MLLMs presents considerable challenges, primarily due to the absence of multimodal benchmarks that align with human preferences. Drawing inspiration from the concept of LLM-as-a-Judge within LLMs, this paper introduces a novel benchmark, termed MLLM-as-a-Judge, to assess the ability of MLLMs in assisting judges across diverse modalities, encompassing three distinct tasks: Scoring Evaluation, Pair Comparison, and Batch Ranking. Our study reveals that, while MLLMs demonstrate remarkable human-like discernment in Pair Comparison, there is a significant divergence from human preferences in Scoring Evaluation and Batch Ranking. Furthermore, a closer examination reveals persistent challenges in the judgment capacities of LLMs, including diverse biases, hallucinatory responses, and inconsistencies in judgment, even in advanced models such as GPT-4V. These findings emphasize the pressing need for enhancements and further research efforts to be undertaken before regarding MLLMs as fully reliable evaluators. In light of this, we advocate for additional efforts dedicated to supporting the continuous development within the domain of MLLM functioning as judges. The code and dataset are publicly available at our project homepage: \url{https://mllm-judge.github.io/}.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 10 Pith papers

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

  1. MentalThink: Shaping Thoughts in Mental SVG World

    cs.AI 2026-07 conditional novelty 7.0 of 10

    MLLMs that generate and render SVG sketches as multi-turn intermediate reasoning steps reach 55.1% on VSIBench and 76.0% on MindCube, far above the Qwen2.5-VL-7B backbone.

  2. OSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models

    cs.AI 2026-07 accept novelty 6.0 of 10

    VLM judges of CUA trajectories are systematically lenient; OSReward measures this with human gold, and OS-Shepherd open models close most of the reliability gap cheaply.

  3. Chain-of-Models: Cross-Model Auditing for Bias-Robust LLM Judges

    cs.CL 2026-05 conditional novelty 6.0 of 10

    Per-bias selection of a cross-family LLM auditor lifts biased-judgment accuracy from 0.805/0.824 baselines to 0.884.

  4. Adaptive Repetition for Mitigating Position Bias in LLM-Based Ranking

    cs.LG 2025-07 conditional novelty 6.0 of 10

    An adaptive early-stopping rule for repeated LLM judgments cuts position-bias mitigation cost by roughly 80 percent while keeping the consensus result.

  5. IntuiTF: MLLM-Guided Transfer Function Optimization for Direct Volume Rendering

    cs.GR 2025-06 conditional novelty 6.0 of 10

    IntuiTF is a framework that guides evolutionary transfer function optimization with MLLM-based human-aligned visual evaluation, achieving up to 0.85 agreement with human experts on medical volume renderings.

  6. Fast or Slow? Integrating Fast Intuition and Deliberate Thinking for Enhancing Visual Question Answering

    cs.CL 2025-06 conditional novelty 6.0 of 10

    FOCUS improves VQA accuracy by routing easy questions through fast zero-shot answering and hard questions through question-conditioned image segmentation before the final answer.

  7. HiGarment: Cross-modal Harmony Based Diffusion Model for Flat Sketch to Realistic Garment Image

    cs.CV 2025-05 conditional novelty 6.0 of 10

    HiGarment generates realistic garment images from a flat sketch and text prompt by retrieving fabric references and dynamically weighting sketch versus text cues, outperforming five existing controllable generation me...

  8. Chart-to-Experience: Benchmarking Multimodal LLMs for Predicting Experiential Impact of Charts

    cs.HC 2025-05 conditional novelty 6.0 of 10

    Multimodal LLMs underperform humans at directly rating charts' experiential impact, but they are substantially better at pairwise comparisons, especially when the human ratings differ clearly.

  9. LLM-as-a-Verifier: A General-Purpose Verification Framework

    cs.AI 2026-07 conditional novelty 5.0 of 10

    Expecting over scoring-token logits yields continuous, scalable verification that improves agent trajectory selection and dense RL rewards across coding, robotics, and medical benchmarks.

  10. $I^2G$: Generating Instructional Illustrations via Text-Conditioned Diffusion

    cs.CL 2025-05 reject novelty 5.0 of 10

    A pairwise-conditioned diffusion model generates instructional illustrations from procedural text and is finetuned with a text-image alignment reward.

Pith tools