Vision-Language Model-as-a-Judge

Ask about this topic

Overview

Vision-language model-as-a-judge studies how multimodal models can evaluate visual outputs and tasks, including image and video generation, image editing, and agent actions. Common approaches use pairwise comparisons, explicit rubrics, step-by-step visual verification, or learned reward models trained on human preferences. Benchmarks compare judgments with human ratings and test whether evaluators notice specific errors, follow instructions, and remain consistent across tasks and languages. A central open challenge is that strong average agreement can coexist with grounding failures, position and authority biases, uncertainty, and weak performance on unfamiliar or fine-grained criteria.

Latest papers

CardsList

59 papers15 papers in the last four weeks, up 88% on the four weeks before. 0.1% of all new papers.

Jun 18, 2026

Judging to Improve: A De-biased VLM-as-3D-Judge Protocol for Single-Image 3D Generation

A companion study established a de-biased, cross-model VLM-as-3D-judge that reliably ranks single-image-to-3D mesh quality where cheap geometry and CLIP proxies fall short. This paper asks: can that judge's preferences specialize a strong open generator, TRELLIS, on one asset class (furniture), cheaply and without human labels? Taking the judge from ranking to optimization is where the work lives. Pushing a VLM judge into the training and evaluation loop exposes failure modes ranking never triggered, so our contribution is an optimization-grade hardening of the judge: a training judge (Qwen2.5-VL-7B) held distinct from an evaluation judge (InternVL3-8B) to break circularity; position-bias correction; and fixes for three failure modes (image overload, geometry-hiding splat renders, and reference-free judging that rewards clean-but-wrong outputs), with calibration evidence (clear-gap win-rate 0.83-1.0; base-vs-base ~0.5). Using this protocol as an independent evaluator, and working only from public models and data with lightweight parameter-efficient adaptation, we find our methods match the strong base rather than exceed it. Independent base samples carry essentially no learnable preference (0.94 order-flip rate), so signal must be engineered by quality-contrastive construction. Across six adaptation methods, two input regimes, and a severity sweep, the most targeted - conditioner repair under severe degradation - reaches parity (0.50) with the base, while no method clears the >=65% win-rate target. The result is mechanistic: clean inputs saturate the judge, flow-DIT fine-tuning washes out through the sampler, and conditioning repair is the locus that moves geometry. Win-rates are directional at n=8 objects. Matching a strong public-data base with cheap adaptation is itself informative: exceeding it needs more than lightweight PEFT on public data, and the judge protocol is reusable.

Ali Asaria, Tony Salomone, Deep Gandhi

Transformer Lab

VLM Evaluation3D Asset GenerationVision-Language Model-as-a-Judge
AskPDF
Jun 15, 2026

TimeVista: Exploring and Exploiting Vision-Language Models as Judges for Time Series Forecasting

High-quality time series forecasting is pivotal for real-world decision-making. However, traditional point-wise metrics often fail to reveal complex temporal patterns and align poorly with human intuitive preferences. While the ''LLM-as-a-Judge'' paradigm has revolutionized text evaluation by providing flexible, human-aligned judgment, its application to time series remains largely unexplored. In this paper, we leverage Vision-Language Models (VLMs) as judges for time series forecasting, harnessing their ability to comprehend time series plots grounded in textual information. Specifically, we propose a novel framework integrating micro- and macro-level judgments informed by contextual information to evaluate time series forecasting. To this end, we introduce TimeVista, a comprehensive VLM-as-a-Judge benchmark comprising 5563 time series samples paired with detailed evaluation rubrics. Extensive meta-evaluations demonstrate that VLMs are highly reliable judges, achieving significantly higher consistency with human preferences than conventional metrics. Building upon our benchmark, we comprehensively assess recent Time Series Foundation Models (TSFMs) under the VLM-as-a-Judge paradigm. Our results demonstrate that VLMs serve as robust and interpretable judges, providing a comprehensive, human-aligned standard for evaluating time series models.

Zhi Chen, Yuxuan Wang, Jialong Wu +5

School of Software, BNRist, Tsinghua University, Beijing 100084, China

VLM EvaluationForecasting BenchmarksVision-Language Model-as-a-JudgeTime-Series Foundation Models
AskPDF
May 31, 2026

TECCI: Tricky Edits of Collected and Curated Images

Despite tremendous recent progress, current text-guided image editing methods still struggle with many aspects of editing involving instruction following, minimally editing the source image, and ensuring high visual quality. These problems are especially apparent when the requested edit is challenging, such as those that involve position, motion, viewpoint, scale and creative edits. To systematically test generative image editors, we propose a novel image editing benchmark -- TECCI: Tricky Edits of Collected and Curated Images. TECCI consists of a completely new set of images we are releasing. The images in TECCI span 7 image categories. The images and these categories were curated intentionally to target weaknesses of existing methods. The edit instructions in TECCI are automatically generated by Gemini, covering 5 edit types per source image. We also curated a set of 530 images for which we created challenging manually written edit instructions. Overall, TECCI contains 7550 pairs of images and edit instructions. We conduct human evaluations of five leading image editing models on TECCI. Humans judge outputs along three dimensions: 1) instruction following, 2) minimality of the edits, and 3) visual quality. To scale-up the evaluation, we also build an auto-rater using Gemini that achieves 74.7% accuracy in matching human evaluations. Our evaluations reveal that: 1) none of the models exceed a 22% overall success rate, demonstrating the challenging nature of TECCI, 2) Nano Banana Pro is the best performing model overall, 3) models perform significantly better at instruction following compared to minimal edits and visual quality, 4) models struggle with editing architecture and nature images which require strong understanding of spatial layout and intricate visual details. 5) reasoning and creative edits are the most difficult, whereas color and appearance edits are the easiest.

Aishwarya Agrawal, Roy Hirsch, Yasumasa Onoe +2

Google Research · Google DeepMind

VLM EvaluationImage Editing EvaluationText-Guided Image EditingVision-Language Model-as-a-Judge
AskPDF
May 29, 2026

A Visually Impaired Assistance Benchmark for VLM-as-a-Judge Evaluation

AI-based Visually Impaired Assistance (VIA) remains challenging, largely due to the high cost of human evaluation. The VLM-as-a-Judge paradigm may offer a promising alternative, although it has mostly been studied in general domains. We therefore ask whether such judges can be trusted for VIA tasks. To investigate this question, we introduce VIABLE (Visually Impaired Assistance Benchmark for VLM-as-a-Judge Evaluation), the first benchmark for VLM-as-a-Judge evaluation in VIA. VIABLE contains over 300K judgment samples across three scenarios and introduces an Effectiveness--Impartiality--Stability framework with a 12-mode failure taxonomy. Based on VIABLE, our systematic study of seven judges across different model scales shows that existing models are largely unreliable across all evaluation axes. The strongest judge, GPT-5.4, achieves only 52.6% single-failure diagnostic accuracy, yet exhibits the highest self-preference rate at 94.2%; while open-source judges are strongly biased and adversarially fragile. To address these issues, we propose VIA-Judge-Agent, a model-agnostic inference-time harness that augments judges with visual evidence extraction and a taxonomy-guided workflow. It enables positive improvements in diagnostic accuracy and downstream VIA responses more preferred by BLV users. Data and code are available at: https://github.com/YiyiyiZhao/VIABLE

Yi Zhao, Siqi Wang, Zhe Hu +2

Department of Computing, The Hong Kong Polytechnic University

VLM EvaluationLLM-as-a-JudgeLanguage Model Bias EvaluationVision-Language Model-as-a-JudgeVisually Impaired Navigation AssistanceAssistive Technology for Blind and Low-Vision People
AskPDF
May 10, 2026

DeltaRubric: Generative Multimodal Reward Modeling via Joint Planning and Verification

Aligning Multimodal Large Language Models (MLLMs) requires reliable reward models, yet existing single-step evaluators can suffer from lazy judging, exploiting language priors over fine-grained visual verification. While rubric-based evaluation mitigates these biases in text-only settings, extending it to multimodal tasks is bottlenecked by the complexity of visual reasoning. The critical differences between responses often depend on instance-specific visual details. Robust evaluation requires dynamically synthesizing rubrics that isolate spatial and factual discrepancies. To address this, we introduce DeltaRubric\textbf{DeltaRubric}, an approach that reformulates multimodal preference evaluation as a plan-and-execute process within a single MLLM. DeltaRubric operates in two steps: acting first as a Disagreement Planner\textit{Disagreement Planner}, the model generates a neutral, instance-specific verification checklist. Transitioning into a Checklist Verifier\textit{Checklist Verifier}, it executes these self-generated checks against the image and question to produce the final grounded judgment. We formulate DeltaRubric as a multi-role reinforcement learning problem, jointly optimizing planning and verification capabilities. Validated on Qwen3-VL 4B and 8B Instruct models, DeltaRubric achieves solid empirical gains. For instance, On VL-RewardBench, it improves base model overall accuracy by +22.6\textbf{+22.6} (4B) and +18.8\textbf{+18.8} (8B) points, largely outperforming standard no-rubric baselines. The results demonstrate that decomposing evaluation into structured, verifiable steps leads to more reliable and generalizable multimodal reward modeling.

Rui Liu, Dian Yu, Zhenwen Liang +6

1Tencent Hunyuan · University of Maryland, College Park · University of North Carolina, Chapel Hill

Multimodal UnderstandingMultimodal Reward ModelingMultimodal Large Language ModelsRubric-Based EvaluationMultimodal Foundation ModelsVision-Language Model-as-a-Judge+3
AskPDF
May 5, 2026

WorldJen: An End-to-End Multi-Dimensional Benchmark for Generative Video Models

Evaluating generative video models remains an open problem. Reference-based metrics such as Structural Similarity Index Measure (SSIM) and Peak Signal to Noise Ratio (PSNR) reward pixel fidelity over semantic correctness, while Frechet Video Distance (FVD) favors distributional textures over physical plausibility. Binary Visual Question Answering (VQA) based benchmarks like VBench~2.0 are prone to yes-bias and rely on low-resolution auditors that miss temporal failures. Moreover, their prompts target a single dimension at a time, multiplying the number of videos required while still not guaranteeing reliable results. WorldJen addresses these limitations directly. Binary VQA is replaced with Likert-scale questionnaires graded by a VLM that receives frames at native video resolution. Video generation costs are addressed by using adversarially curated prompts that are designed to exercise up to 16 quality dimensions simultaneously. The framework is built around two interlocking contributions. First, A blind human preference study is conducted, accumulating (2,696 pairwise annotations from 7 annotators with 100% pair coverage over 50 of the curated prompts ×\times 6 state-of-the-art video models. A mean inter-annotator agreement of 66.9% is achieved and the study establishes a human ground-truth Bradley-Terry (BT) rating with a three-tier structure. Second, A VLM-as-a-judge evaluation engine using prompt-specific, dimension-specific Likert questionnaires (10 questions per dimension, 47,160 scored responses) judges the videos and reproduces the human-established three-tier BT rating structure independently. The VLM achieves a Spearman ρ^=1.000, p=0.0014\hatρ=1.000,~p=0.0014 that is interpreted as tier agreement with the human results. Six focused ablation studies validate the robustness of the VLM evaluation framework. Project page: https://moonmath.ai/worldjen/

Karthik Inbasekar, Guy Rom, Omer Shlomovits

Human Preference EvaluationVideo Quality AssessmentVision-Language Model-as-a-JudgeVideo Generation Evaluation
AskPDF
Apr 28, 2026

VLM Judges Can Rank but Cannot Score: Task-Dependent Uncertainty in Multimodal Evaluation

Vision-language models (VLMs) are increasingly used as automated judges for multimodal systems, yet their scores provide no indication of reliability. We study this problem through conformal prediction, a distribution-free framework that converts a judge's point score into a calibrated prediction interval using only score-token log-probabilities, with no retraining. We present the first systematic analysis of conformal prediction for VLM-as-a-Judge across 3 judges and 14 visual task categories. Our results show that evaluation uncertainty is strongly task-dependent: intervals cover ~40% of the score range for aesthetics and natural images but expand to ~70% for chart and mathematical reasoning, yielding a quantitative reliability map for multimodal evaluation. We further identify a failure mode not captured by standard evaluation metrics, ranking-scoring decoupling, where judges achieve high ranking correlation while producing wide, uninformative intervals, correctly ordering responses but failing to assign reliable absolute scores. Finally, we show that interval width is driven primarily by task difficulty and annotation quality, i.e., the same judge and method yield 4.5x narrower intervals on a clean, multi-annotator captioning benchmark. Code: https://github.com/divake/VLM-Judge-Uncertainty

Divake Kumar, Sina Tayebati, Devashri Naik +2

University of Illinois at Chicago · 2AI Labs at Capital One

VLM EvaluationVision-Language ModelsLLM-as-a-JudgeConformal PredictionVision-Language Model-as-a-JudgeMultimodal Model Evaluation
AskPDF
Apr 27, 2026

SycoPhantasy: Quantifying Sycophancy and Hallucination in Small Open Weight VLMs for Vision-Language Scoring of Fantasy Characters

Vision-language models (VLMs) are increasingly deployed as evaluators in tasks requiring nuanced image understanding, yet their reliability in scoring alignment between images and text descriptions remains underexplored. We investigate whether small, open-weight VLMs exhibit sycophantic behavior when evaluating image-text alignment: assigning high scores without grounding their judgments in visual evidence. To quantify this phenomenon, we introduce the Bluffing Coefficient (\bc), a metric that measures the mismatch between a model's score and its evidence recall. We evaluate six open-weight VLMs ranging from 450M to 8B parameters on a benchmark of 173,810 AI-generated character portraits paired with detailed textual descriptions. Our analysis reveals a significant inverse correlation between model size and sycophancy rate (r=−0.96r = -0.96, p=0.002p = 0.002), with smaller models exhibiting substantially higher rates of unjustified high scores. The smallest model tested (LFM2-VL, 450M) produced sycophantic evaluations in 22.3% of cases, compared to 6.0% for the largest (LLaVA-1.6, 7B). These findings have direct implications for the deployment of small, open-weight VLMs as automated evaluators within attribute-rich, synthetic image evaluation tasks, where the gap between assigned scores and cited visual evidence is both measurable and consequential.

Arya Shah, Deepali Mishra, Chaklam Silpasuwanchai

IIT Gandhinagar Gandhinagar, India · IIT Kanpur Kanpur, India · Asian Institute of Technology Bangkok, Thailand

VLM EvaluationVision-Language ModelsVision-Language AlignmentVision-Language Model-as-a-JudgeSycophancy in VLMs
AskPDF
Apr 23, 2026

Seeing Isn't Believing: Uncovering Blind Spots in Evaluator Vision-Language Models

Large Vision-Language Models (VLMs) are increasingly used to evaluate outputs of other models, for image-to-text (I2T) tasks such as visual question answering, and text-to-image (T2I) generation tasks. Despite this growing reliance, the reliability of these Evaluator VLMs remains under explored. In this work, we systematically evaluate the reliability of Evaluator VLMs across both I2T and T2I tasks. We introduce targeted perturbations that degrade output quality along key error dimensions, including object hallucinations, spatial reasoning, factual grounding, and visual fidelity. These perturbations test whether Evaluator VLMs can reliably account for these quality degrading errors in their evaluations. Using a comprehensive benchmark of over 4000 perturbed instances spanning 40 perturbation dimensions, we evaluate 4 prominent VLMs using single-answer scoring, pairwise comparison, and reference-guided paradigms. Our findings reveal that current VLM evaluators exhibit substantial blind spots: they often fail to detect perturbed outputs - in some cases exceeding 50%, struggle particularly with fine-grained compositional and spatial errors, and are often insensitive to hallucinated content that contradicts the input image. Pairwise comparison proves more reliable, though failure rates persist. These results highlight the unreliable nature of current Evaluator VLMs and urge caution in their deployment for benchmarking and development decisions. Code and data have been made publicly available.

Mohammed Safi Ur Rahman Khan, Sanjay Suryanarayanan, Tushar Anand +1

♠Nilekani Centre at AI4Bharat ♢Indian Institute of Technology Madras · ♠Nilekani Centre at AI4Bharat · ♣BITS Pilani, Hyderabad

VLM EvaluationVLM RobustnessLanguage Model Generation EvaluationVision-Language Model-as-a-JudgeT2I Generation Evaluation
AskPDF
Apr 21, 2026

Lost in Translation: Do LVLM Judges Generalize Across Languages?

Automatic evaluators such as reward models play a central role in the alignment and evaluation of large vision-language models (LVLMs). Despite their growing importance, these evaluators are almost exclusively assessed on English-centric benchmarks, leaving open the question of how well these evaluators generalize across languages. To answer this question, we introduce MM-JudgeBench, the first large-scale benchmark for multilingual and multimodal judge model evaluation, which includes over 60K pairwise preference instances spanning 25 typologically diverse languages. MM-JudgeBench integrates two complementary subsets: a general vision-language preference evaluation subset extending VL-RewardBench, and a chart-centric visual-text reasoning subset derived from OpenCQA, enabling systematic analysis of reward models (i.e., LVLM judges) across diverse settings. We additionally release a multilingual training set derived from MM-RewardBench, disjoint from our evaluation data, to support domain adaptation. By evaluating 22 LVLMs (15 open-source, 7 proprietary), we uncover substantial cross-lingual performance variance in our proposed benchmark. Our analysis further shows that model size and architecture are poor predictors of multilingual robustness, and that even state-of-the-art LVLM judges exhibit inconsistent behavior across languages. Together, these findings expose fundamental limitations of current reward modeling and underscore the necessity of multilingual, multimodal benchmarks for developing reliable automated evaluators.

Md Tahmid Rahman Laskar, Mohammed Saidul Islam, Mir Tafseer Nayeem +5

York University · University of Alberta · Nanyang Technological University +1

Multimodal Model EvaluationMultilingual VLM EvaluationMultimodal Reward ModelingVLM EvaluationVision-Language Model-as-a-JudgeCross-Lingual Consistency in Language Models+1
AskPDF
Apr 20, 2026

MM-JudgeBias: A Benchmark for Evaluating Compositional Biases in MLLM-as-a-Judge

Multimodal Large Language Models (MLLMs) have been increasingly used as automatic evaluators-a paradigm known as MLLM-as-a-Judge. However, their reliability and vulnerabilities to biases remain underexplored. We find that many MLLM judges fail to reliably integrate key visual or textual cues, yielding unreliable evaluations when evidence is missing or mismatched, and exhibiting instability under semantically irrelevant perturbations. To address this, we systematically define Compositional Bias in MLLM-as-a-Judge systems and introduce MM-JudgeBias, a benchmark for evaluating it. MM-JudgeBias introduces controlled perturbations across Query, Image, and Response, and evaluates model behavior via two complementary metrics: Bias-Deviation (BD) for sensitivity and Bias-Conformity (BC) for stability. Our dataset of over 1,800 curated and refined multimodal samples, drawn from 29 source benchmarks, enables a fine-grained diagnosis of nine bias types across diverse tasks and domains. Experiments on 26 state-of-the-art MLLMs reveal systematic modality neglect and asymmetric evaluation tendencies, underscoring the need for more reliable judges.

Sua Lee, Sanghee Park, Jinbae Im

Seoul National University · NAVER Cloud AI · KAIST AI

LLM-as-a-JudgeMultimodal Large Language ModelsBias in LLM-as-a-JudgeVision-Language Model-as-a-JudgeMultimodal Model Evaluation
AskPDF
Apr 20, 2026

When Vision-Language Models Judge Without Seeing: Exposing Informativeness Bias

The reliability of VLM-as-a-Judge is critical for the automatic evaluation of vision-language models (VLMs). Despite recent progress, our analysis reveals that VLM-as-a-Judge often pays limited attention to the image when making decisions. Instead, they often blindly favor the more informative answer, even when they can recognize it conflicts with the image content. We call this problem informativeness bias, which significantly undermines judge reliability. To address it, we propose BIRCH (Balanced Informativeness and CoRrectness with a Truthful AnCHor), a judging paradigm that first corrects inconsistencies with the image content in candidate answers, and then compares the answers against this corrected version. This shifts the judge's focus from informativeness to image-grounded correctness. Experiments on multiple models and benchmarks show that BIRCH reduces informativeness bias by up to 17%, resulting in performance gains of up to 9.8%. Our work reveals an overlooked but fundamental flaw in current VLM-as-a-Judge systems and highlights the need for more principled designs.

Xiaohan Zou, Roshan Sridhar, Mohammadtaher Safarzadeh +1

The Pennsylvania State University · Oracle AI

VLM EvaluationVision-Language AlignmentVision-Language Model-as-a-Judge
AskPDF
Apr 14, 2026

Redefining Quality Criteria and Distance-Aware Score Modeling for Image Editing Assessment

Recent advances in image editing have heightened the need for reliable Image Editing Quality Assessment (IEQA). Unlike traditional methods, IEQA requires complex reasoning over multimodal inputs and multi-dimensional assessments. Existing MLLM-based approaches often rely on human heuristic prompting, leading to two key limitations: rigid metric prompting and distance-agnostic score modeling. These issues hinder alignment with implicit human criteria and fail to capture the continuous structure of score spaces. To address this, we propose Define-and-Score Image Editing Quality Assessment (DS-IEQA), a unified framework that jointly learns evaluation criteria and score representations. Specifically, we introduce Feedback-Driven Metric Prompt Optimization (FDMPO) to automatically refine metric definitions via probabilistic feedback. Furthermore, we propose Token-Decoupled Distance Regression Loss (TDRL), which decouples numerical tokens from language modeling to explicitly model score continuity through expected distance minimization. Extensive experiments show our method's superior performance; it ranks 4th in the 2026 NTIRE X-AIGC Quality Assessment Track 2 without any additional training data.

Xinjie Zhang, Qiang Li, Xiaowen Ma +3

Northwestern Polytechnical University, China · Shenzhen Research Institute of Northwestern Polytechnical University, China

Image Editing EvaluationImage Quality AssessmentVision-Language Model-as-a-JudgeAutomatic Prompt Optimization
AskPDF
Apr 9, 2026

Imperfect Visual Verification for Code Edition : A Case Study on TikZ

LLMs have significantly advanced code generation, enabling the synthesis of functional programs. While recent systems achieve strong performance on many coding benchmarks, tasks involving programs such as TikZ that generate visual artifacts remain challenging, in particular on visual code customization. Unlike generation from scratch, customization requires localized, semantics-preserving edits: the model must locate relevant code, modify it according to the instruction, and preserve the remaining structure and rendering. Approaches based on post-hoc iterative refinement/correction where a verifier provides feedback to guide corrections, have shown promise. However, in the case of programs with a visual outcome such as in TikZ, where correctness is harder or likely impossible to formalize and evaluate automatically, deterministic verifiers do not exist. Hence, developers can only rely on imperfect verifiers. In this paper, we conduct an empirical study to answer:to what extent can iterative refinement remain effective when the verifier itself is unreliable?} We use TikZ as a focused case study that isolates the core difficulties of the problem (weak code structure, fine-grained visual semantics, and difficult feature localization) in a controlled and challenging setting. We define visual code customization as an iterative editing problem with an imperfect oracle, and introduce a framework for analyzing such iterative refinements. We conduct a large-scale study and evaluate multiple LLM-based and tool-augmented visual verifiers within iterative refinement pipelines, and perform extensive manual annotation of refinement trajectories to assess verifier behavior and feedback quality. Our findings show that even imperfect verifiers can determine with moderate accuracy whether visual instructions are applied to code, achieving F1-scores up to 0.815. Feedback improves iterative refinement, especially for weaker models, adding 11--20 perfect customizations for Qwen3-vl-30b-a3b-Instruct, while stronger models like Gemini-3 gain fewer improvements (+5) but benefit more from accurate verification that prevents premature acceptance. Feedback is effective only when it precisely identifies image issues, provides actionable guidance, addresses all relevant problems, and remains grounded in the original instruction.

Charly Reux, Mathieu Acher, Djamel Eddine Khelladi +2

UR, INSA Rennes, DiverSe · Univ Rennes, Inria, IRISA, INSA, Rennes, France · CNRS, IUF, IRISA, UR, DiverSe +7

LLM Self-RefinementCode GenerationVision-Language Model-as-a-Judge
AskPDF
Apr 7, 2026

Vision-Guided Iterative Refinement for Frontend Code Generation

Code generation with large language models often relies on multi-stage human-in-the-loop refinement, which is effective but very costly - particularly in domains such as frontend web development where the solution quality depends on rendered visual output. We present a fully automated critic-in-the-loop framework in which a vision-language model serves as a visual critic that provides structured feedback on rendered webpages to guide iterative refinement of generated code. Across real-world user requests from the WebDev Arena dataset, this approach yields consistent improvements in solution quality, achieving up to 17.8% increase in performance over three refinement cycles. Next, we investigate parameter-efficient fine-tuning using LoRA to understand whether the improvements provided by the critic can be internalized by the code-generating LLM. Fine-tuning achieves 25% of the gains from the best critic-in-the-loop solution without a significant increase in token counts. Our findings indicate that automated, VLM-based critique of frontend code generation leads to significantly higher quality solutions than can be achieved through a single LLM inference pass, and highlight the importance of iterative refinement for the complex visual outputs associated with web development.

Hannah Sansford, Derek H. C. Law, Wei Liu +3

School of Mathematics, University of Bristol, UK · Amazon AGI

Code GenerationVision-Language Model-as-a-Judge
AskPDF
Apr 7, 2026

LLM-as-Judge for Semantic Judging of Powerline Segmentation in UAV Inspection

The deployment of lightweight segmentation models on drones for autonomous power line inspection presents a critical challenge: maintaining reliable performance under real-world conditions that differ from training data. Although compact architectures such as U-Net enable real-time onboard inference, their segmentation outputs can degrade unpredictably in adverse environments, raising safety concerns. In this work, we study the feasibility of using a large language model (LLM) as a semantic judge to assess the reliability of power line segmentation results produced by drone-mounted models. Rather than introducing a new inspection system, we formalize a watchdog scenario in which an offboard LLM evaluates segmentation overlays and examine whether such a judge can be trusted to behave consistently and perceptually coherently. To this end, we design two evaluation protocols that analyze the judge's repeatability and sensitivity. First, we assess repeatability by repeatedly querying the LLM with identical inputs and fixed prompts, measuring the stability of its quality scores and confidence estimates. Second, we evaluate perceptual sensitivity by introducing controlled visual corruptions (fog, rain, snow, shadow, and sunflare) and analyzing how the judge's outputs respond to progressive degradation in segmentation quality. Our results show that the LLM produces highly consistent categorical judgments under identical conditions while exhibiting appropriate declines in confidence as visual reliability deteriorates. Moreover, the judge remains responsive to perceptual cues such as missing or misidentified power lines, even under challenging conditions. These findings suggest that, when carefully constrained, an LLM can serve as a reliable semantic judge for monitoring segmentation quality in safety-critical aerial inspection tasks.

Akram Hossain, Rabab Abdelfattah, Xiaofeng Wang +1

School of Computing Sciences and Computer Engineering, The University of Southern Mississippi, Hattiesburg, MS 39406 USA. · Electrical Engineering Department, University of South Carolina, Columbia, SC, USA. · Computer Science Department, Faculty of Computers & Artificial Intelligence, Fayoum University, Fayoum, Egypt.

LLM ReliabilityVision-Language Model-as-a-JudgeIndustrial InspectionSemantic Segmentation
AskPDF
Mar 30, 2026

GEditBench v2: A Human-Aligned Benchmark for General Image Editing

Recent advances in image editing have enabled models to handle complex instructions with impressive realism. However, existing evaluation frameworks lag behind: current benchmarks suffer from narrow task coverage, while standard metrics fail to adequately capture visual consistency, i.e., the preservation of identity, structure and semantic coherence between edited and original images. To address these limitations, we introduce GEditBench v2, a comprehensive benchmark with 1,200 real-world user queries spanning 23 tasks, including a dedicated open-set category for unconstrained, out-of-distribution editing instructions beyond predefined tasks. Furthermore, we propose PVC-Judge, an open-source pairwise assessment model for visual consistency, trained via two novel region-decoupled preference data synthesis pipelines. Besides, we construct VCReward-Bench using expert-annotated preference pairs to assess the alignment of PVC-Judge with human judgments on visual consistency evaluation. Experiments show that our PVC-Judge achieves state-of-the-art evaluation performance among open-source models and even surpasses GPT-5.1 on average. Finally, by benchmarking 16 frontier editing models, we show that GEditBench v2 enables more human-aligned evaluation, revealing critical limitations of current models, and providing a reliable foundation for advancing precise image editing.

Zhangqi Jiang, Zheng Sun, Xianfang Zeng +7

Nanyang Technological University · StepFun · Southeast University

Image Editing EvaluationText-Guided Image EditingVision-Language Model-as-a-Judge
AskPDF
Mar 30, 2026

AIBench: Evaluating Visual-Logical Consistency in Academic Illustration Generation

Although image generation has boosted various applications via its rapid evolution, whether the state-of-the-art models are able to produce ready-to-use academic illustrations for papers is still largely unexplored. Directly comparing or evaluating the illustration with VLM is native but requires oracle multi-modal understanding ability, which is unreliable for long and complex texts and illustrations. To address this, we propose AIBench, the first benchmark using VQA for evaluating logic correctness of the academic illustrations and VLMs for assessing aesthetics. In detail, we designed four levels of questions proposed from a logic diagram summarized from the method part of the paper, which query whether the generated illustration aligns with the paper on different scales. Our VQA-based approach raises more accurate and detailed evaluations on visual-logical consistency while relying less on the ability of the judger VLM. With our high-quality AIBench, we conduct extensive experiments and conclude that the performance gap between models on this task is significantly larger than general ones, reflecting their various complex reasoning and high-density generation ability. Further, the logic and aesthetics are hard to optimize simultaneously as in handcrafted illustrations. Additional experiments further state that test-time scaling on both abilities significantly boosts the performance on this task.

Zhaohe Liao, Kaixun Jiang, Zhihang Liu +11

Tongyi Lab, Alibaba Group · SJTU · FDU +3

Image Generation EvaluationImage GenerationVision-Language Model-as-a-Judge
AskPDF
Mar 20, 2026

Evaluating Image Editing with LLMs: A Comprehensive Benchmark and Intermediate-Layer Probing Approach

Evaluating text-guided image editing (TIE) methods remains a challenging problem, as reliable assessment should simultaneously consider perceptual quality, alignment with textual instructions, and preservation of original image content. Despite rapid progress in TIE models, existing evaluation benchmarks remain limited in scale and often show weak correlation with human perceptual judgments. In this work, we introduce TIEdit, a benchmark for systematic evaluation of text-guided image editing methods. TIEdit consists of 512 source images paired with editing prompts across eight representative editing tasks, producing 5,120 edited images generated by ten state-of-the-art TIE models. To obtain reliable subjective ratings, 20 experts are recruited to produce 307,200 raw subjective ratings, which accumulates into 15,360 mean opinion scores (MOSs) across three evaluation dimensions: perceptual quality, editing alignment, and content preservation. Beyond the benchmark itself, we further propose EditProbe, an LLM-based evaluator that estimates editing quality via intermediate-layer probing of hidden representations. Instead of relying solely on final model outputs, EditProbe extracts informative representations from intermediate layers of multimodal large language models to better capture semantic and perceptual relationships between source images, editing instructions, and edited results. Experimental results demonstrate that widely used automatic evaluation metrics show limited correlation with human judgments on editing tasks, while EditProbe achieves substantially stronger alignment with human perception. Together, TIEdit and EditProbe provide a foundation for more reliable and perceptually aligned evaluation of text-guided image editing methods.

Shiqi Gao, Zitong Xu, Kang Fu +3

Institute of Image Communication and Network Engineering, Shanghai Jiao Tong University, Shanghai 200240, China

Image Editing EvaluationText-Guided Image EditingVision-Language Model-as-a-Judge
AskPDF
Mar 16, 2026

EditHF-1M: A Million-Scale Rich Human Preference Feedback for Image Editing

Recent text-guided image editing (TIE) models have achieved remarkable progress, while many edited images still suffer from issues such as artifacts, unexpected editings, unaesthetic contents. Although some benchmarks and methods have been proposed for evaluating edited images, scalable evaluation models are still lacking, which limits the development of human feedback reward models for image editing. To address the challenges, we first introduce EditHF-1M, a million-scale image editing dataset with over 29M human preference pairs and 148K human mean opinion ratings, both evaluated from three dimensions, i.e., visual quality, instruction alignment, and attribute preservation. Based on EditHF-1M, we propose EditHF, a multimodal large language model (MLLM) based evaluation model, to provide human-aligned feedback from image editing. Finally, we introduce EditHF-Reward, which utilizes EditHF as the reward signal to optimize the text-guided image editing models through reinforcement learning. Extensive experiments show that EditHF achieves superior alignment with human preferences and demonstrates strong generalization on other datasets. Furthermore, we fine-tune the Qwen-Image-Edit using EditHF-Reward, achieving significant performance improvements, which demonstrates the ability of EditHF to serve as a reward model to scale-up the image editing. Both the dataset and code will be released in our GitHub repository: https://github.com/IntMeGroup/EditHF.

Zitong Xu, Huiyu Duan, Zhongpeng Ji +9

Shanghai Jiao Tong University · Vivo Mobile Communication Co., Ltd · University of Electronic and Science Technology of China +2

Image Editing EvaluationText-Guided Image EditingHuman Preference ModelingVision-Language Model-as-a-JudgeRL for Image Editing
AskPDF
Mar 12, 2026

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

Multimodal Large Language Models (MLLMs) have been widely adopted as MLLM-as-a-Judges due to their strong alignment with human judgment across various visual tasks. However, most existing judge models are optimized for single-task scenarios and struggle to generalize to diverse contexts, which is a critical requirement for reliable evaluation. To address this limitation, we propose Multi-Task Reinforcement Learning for MLLM-as-a-Judge (MT-RL-Judge), a framework that jointly optimizes the judge model across multiple tasks, leveraging the generalization capabilities of RL. Experimental results against several strong baselines demonstrate that MT-RL-Judge outperforms strong baselines in both judgment consistency and correlation with human preferences. Furthermore, our approach exhibits robust generalization on out-of-distribution tasks, further validating its effectiveness.

Junjie Wu, Xuan Kan, Zihao He +3

Meta AI · Hong Kong University of Science and Technology · Emory University

Vision-Language Model-as-a-JudgeMultimodal Model EvaluationRL for VLMsMulti-Task RL
AskPDF
Mar 11, 2026

HanMoVLM: Large Vision-Language Models for Professional Artistic Painting Evaluation

While Large Vision-Language Models (VLMs) demonstrate impressive general visual capabilities, they remain artistically blind and unable to offer professional evaluation of artworks within specific artistic domains like human experts. To bridge this gap, we transform VLMs into experts capable of professional-grade painting evaluation in the Chinese Artistic Domain, which is more abstract and demands extensive artistic training for evaluation. We introduce HanMo-Bench, a new dataset that features authentic auction-grade masterpieces and AI-generated works, grounded in real-world market valuations. To realize the rigorous judgment, we propose the HanMoVLM and construct a Chain-of-Thought (CoT) validated by experts. This CoT guides the model to perform expert-level reasoning: from content identification and Region of Interest (RoI) localization to professional evaluation, guided by both theme-specific evaluation and typical three-tier evaluation in Chinese paintings. Furthermore, we design a reward function to refine the reasoning process of the HanMoVLM to improve the accuracy. We demonstrate that HanMoVLM can serve as a critical backbone for Test-time Scaling in image generation. By acting as a high-quality verifier, HanMoVLM enables generative models to select the most artistically superior outputs from multiple candidates. Experimental results and human studies confirm that the proposed HanMoVLM effectively bridges the gap, achieving a high consistency with professional experts and significantly improving the quality of Chinese Painting generation.

Hongji Yang, Yucheng Zhou, Wencheng Han +3

SKL-IOTSC, CIS, University of Macau · CSE, Shandong University · Online-Video BU, Tencent

Vision-Language Model-as-a-JudgeVLM EvaluationImage Generation Evaluation
AskPDF
Mar 11, 2026

CUAAudit: Meta-Evaluation of Vision-Language Models as Auditors of Autonomous Computer-Use Agents

Computer-Use Agents (CUAs) are emerging as a new paradigm in human-computer interaction, enabling autonomous execution of tasks in desktop environment by perceiving high-level natural-language instructions. As such agents become increasingly capable and are deployed across diverse desktop environments, evaluating their behavior in a scalable and reliable manner becomes a critical challenge. Existing evaluation pipelines rely on static benchmarks, rule-based success checks, or manual inspection, which are brittle, costly, and poorly aligned with real-world usage. In this work, we study Vision-Language Models (VLMs) as autonomous auditors for assessing CUA task completion directly from observable interactions and conduct a large-scale meta-evaluation of five VLMs that judge task success given a natural-language instruction and the final environment state. Our evaluation spans three widely used CUA benchmarks across macOS, Windows, and Linux environments and analyzes auditor behavior along three complementary dimensions: accuracy, calibration of confidence estimates, and inter-model agreement. We find that while state-of-the-art VLMs achieve strong accuracy and calibration, all auditors exhibit notable performance degradation in more complex or heterogeneous environments, and even high-performing models show significant disagreement in their judgments. These results expose fundamental limitations of current model-based auditing approaches and highlight the need to explicitly account for evaluator reliability, uncertainty, and variance when deploying autonomous CUAs in real-world settings.

Marta Sumyk, Oleksandr Kosovan

Ukrainian Catholic University Lviv, Ukraine

Computer-Use Agent EvaluationAI Agent AuditingVision-Language Model-as-a-JudgeComputer-Use AgentsProbability Calibration
AskPDF
Mar 9, 2026

MJ1: Multimodal Judgment via Grounded Verification

Multimodal judges struggle to ground decisions in visual evidence. We present MJ1, a multimodal judge trained with reinforcement learning that enforces visual grounding through a structured grounded verification chain (observations →\rightarrow claims →\rightarrow verification →\rightarrow evaluation →\rightarrow scoring) and a counterfactual consistency reward that penalizes position bias. Even without training, our mechanism improves base-model accuracy on MMRB2 by +3.8 points on Image Editing and +1.7 on Multimodal Reasoning. After training, MJ1, with only 3B active parameters, achieves 77.0% accuracy on MMRB2 and surpasses orders-of-magnitude larger models like Gemini-3-Pro. These results show that grounded verification and consistency-based training substantially improve multimodal judgment without increasing model scale.

Bhavesh Kumar, Dylan Feng, Leonard Tang

Haize Labs

Position Bias in VLMsMLLM EvaluationVision-Language Model-as-a-JudgeMultimodal Model EvaluationMultimodal GroundingVision-Language Models
AskPDF
Feb 26, 2026

DesignSense: A Human Preference Dataset and Reward Modeling Framework for Graphic Layout Generation

Graphic layouts serve as an important and engaging medium for visual communication across different channels. While recent layout generation models have demonstrated impressive capabilities, they frequently fail to align with nuanced human aesthetic judgment. Existing preference datasets and reward models trained on text-to-image generation do not generalize to layout evaluation, where the spatial arrangement of identical elements determines quality. To address this critical gap, we introduce DesignSense-10k, a large-scale dataset of 10,235 human-annotated preference pairs for graphic layout evaluation. We propose a five-stage curation pipeline that generates visually coherent layout transformations across diverse aspect ratios, using semantic grouping, layout prediction, filtering, clustering, and VLM-based refinement to produce high-quality comparison pairs. Human preferences are annotated using a 4-class scheme (left, right, both good, both bad) to capture subjective ambiguity. Leveraging this dataset, we train DesignSense, a vision-language model-based classifier that substantially outperforms existing open-source and proprietary models across comprehensive evaluation metrics (54.6% improvement in Macro F1 over the strongest proprietary baseline). Our analysis shows that frontier VLMs remain unreliable overall and fail catastrophically on the full four-class task, underscoring the need for specialized, preference-aware models. Beyond the dataset, our reward model DesignSense yields tangible downstream gains in layout generation. Using our judge during RL based training improves generator win rate by about 3%, while inference-time scaling, which involves generating multiple candidates and selecting the best one, provides a 3.6% improvement. These results highlight the practical impact of specialized, layout-aware preference modeling on real-world layout generation quality.

Varun Gopal, Rishabh Jain, Aradhya Mathur +6

MDSR, Adobe

Human Preference ModelingReward ModelingVision-Language Model-as-a-Judge
AskPDF
Feb 25, 2026

E-comIQ-ZH: A Human-Aligned Dataset and Benchmark for Fine-Grained Evaluation of E-commerce Posters with Chain-of-Thought

Generative AI is widely used to create commercial posters. However, rapid advances in generation have outpaced automated quality assessment. Existing models emphasize generic esthetics or low level distortions and lack the functional criteria required for e-commerce design. It is especially challenging for Chinese content, where complex characters often produce subtle but critical textual artifacts that are overlooked by existing methods. To address this, we introduce E-comIQ-ZH, a framework for evaluating Chinese e-commerce posters. We build the first dataset E-comIQ-18k to feature multi dimensional scores and expert calibrated Chain of Thought (CoT) rationales. Using this dataset, we train E-comIQ-M, a specialized evaluation model that aligns with human expert judgment. Our framework enables E-comIQ-Bench, the first automated and scalable benchmark for the generation of Chinese e-commerce posters. Extensive experiments show our E-comIQ-M aligns more closely with expert standards and enables scalable automated assessment of e-commerce posters. All datasets, models, and evaluation tools will be released to support future research in this area.Code will be available at https://github.com/4mm7/E-comIQ-ZH.

Meiqi Sun, Mingyu Li, Junxiong Zhu

Taobao & Tmall Group, Alibaba Group

Text Rendering in T2I GenerationAutomated EvaluationT2I Generation EvaluationT2I GenerationVision-Language Model-as-a-Judge
AskPDF
Feb 23, 2026

Pixels Don't Lie (But Your Detector Might): Bootstrapping MLLM-as-a-Judge for Trustworthy Deepfake Detection and Reasoning Supervision

Deepfake detection models often generate natural-language explanations, yet their reasoning is frequently ungrounded in visual evidence, limiting reliability. Existing evaluations measure classification accuracy but overlook reasoning fidelity. We propose DeepfakeJudge, a framework for scalable reasoning supervision and evaluation, that integrates an out-of-distribution benchmark containing recent generative and editing forgeries, a human-annotated subset with visual reasoning labels, and a suite of evaluation models, that specialize in evaluating reasoning rationales without the need for explicit ground truth reasoning rationales. The Judge is optimized through a bootstrapped generator-evaluator process that scales human feedback into structured reasoning supervision and supports both pointwise and pairwise evaluation. On the proposed meta-evaluation benchmark, our reasoning-bootstrapped model achieves an accuracy of 96.2%, outperforming 30x larger baselines. The reasoning judge attains very high correlation with human ratings and 98.9% percent pairwise agreement on the human-annotated meta-evaluation subset. These results establish reasoning fidelity as a quantifiable dimension of deepfake detection and demonstrate scalable supervision for interpretable deepfake reasoning. Our user study shows that participants preferred the reasonings generated by our framework 70% of the time, in terms of faithfulness, groundedness, and usefulness, compared to those produced by other models and datasets. All of our datasets, models, and codebase are open-sourced.

Kartik Kuckreja, Parul Gupta, Muhammad Haris Khan +1

MBZUAI · Monash University *

Deepfake DetectionExplainable Deepfake DetectionMultimodal Deepfake DetectionMultimodal Deception DetectionFaithfulness of Language Model ExplanationsVision-Language Model-as-a-Judge+1
AskPDF
Jun 2, 2025

TIIF-Bench: How Does Your T2I Model Follow Your Instructions?

The rapid advancements of Text-to-Image (T2I) models have ushered in a new phase of AI-generated content, marked by their growing ability to interpret and follow user instructions. However, existing T2I model evaluation benchmarks fall short in limited prompt diversity and complexity, as well as coarse evaluation metrics, making it difficult to evaluate the fine-grained alignment performance between textual instructions and generated images. In this paper, we present TIIF-Bench Text-to-Image Instruction Following Benchmark), aiming to systematically assess T2I models' ability in interpreting and following intricate textual instructions. TIIF-Bench comprises 5,000 prompts organized along multiple dimensions and categorized into three levels of difficulty and complexity. To rigorously evaluate robustness to prompt length, each prompt is provided in both short and long versions with identical core semantics. We further propose a novel Global Normalized Edit Distance (GNED) metric for text rendering and provide aspect-ratio-diverse reference images for each prompt to assess style control. In addition, we collect 100 high-quality designer-level prompts covering diverse scenarios for comprehensive evaluation. To enable scalable and fine-grained evaluation, we explore the best paradigm for leveraging the world knowledge encoded in large Vision-Language Models (VLMs) as automated binary evaluators. Through extensive ablations, we develop a fully reproducible evaluator that provides interpretable reasoning and reliable verification, enabling our benchmark to discern subtle variations in T2I model outputs. Through comprehensive benchmarking of mainstream T2I models on TIIF-Bench, we analyze the strengths and weaknesses of current T2I systems and reveal the limitations of existing evaluation benchmarks. Project Page: https://a113n-w3i.github.io/TIIF_Bench/.

Xinyu Wei, Jinrui Zhang, Zeqing Wang +4

The Hong Kong Polytechnic University · OPPO Research Institute · Tsinghua University

Vision-Language Model-as-a-JudgeT2I Generation Evaluation
AskPDF
May 1, 2025

Multimodal Language Models as Text-to-Image Model Evaluators

The steady improvements of text-to-image (T2I) generative models lead to slow deprecation of automatic evaluation benchmarks that rely on static datasets, motivating researchers to seek alternative ways to evaluate T2I progress. We present Multimodal Text-to-Image Eval (MT2IE), an evaluation framework in which a single multimodal large language model (MLLM) acts as an evaluator agent, iteratively generating the evaluation prompts and scoring the resulting images. We show that MT2IE's image-text consistency scores have higher correlation with human judgment than metrics previously introduced in the literature. MT2IE generates prompts that are efficient at probing T2I model performance: closely recovering the official T2I model rankings of three structurally distinct benchmarks from just 20 generated evaluation prompts, 28-105x fewer than the benchmarks' own prompt sets. When compared to existing evaluation metrics such as CLIPScore, VIEScore, and VQAScore, MT2IE's T2I model rankings are more faithful and far more consistent across multiple evaluation seeds when using the same number of prompts. MT2IE can also adapt evaluation to the model being tested: rewriting each prompt based on the model's own measured performance to produce a bespoke per-model benchmark that still recovers the official rankings and keeps the evaluated model in an informative scoring range. We hope that these results will encourage the development of dynamic and interactive evaluation frameworks, and mitigate the deprecation of automatic evaluation benchmarks.

Jiahui Chen, Candace Ross, Reyhane Askari-Hemmat +5

1FAIR at Meta · University of Texas at Austin · ∗Work done at Meta +3

LLM-as-a-JudgeBenchmark DesignAutomated EvaluationVision-Language Model-as-a-JudgeT2I Generation EvaluationMLLM Evaluation
AskPDF