mlsanityData to 7 Oct 2026

Teaching Large Language Models to Regress Accurate Image Quality Scores using Score Distribution

arXiv 2501.11561 ↗First posted 20 Jan 2025cs.CVcited by 139

On the map: Multimodal LLMs / vision-language models → Image editing and quality assessment

Abstract (arXiv)

With the rapid advancement of Multi-modal Large Language Models (MLLMs), MLLM-based Image Quality Assessment (IQA) methods have shown promising performance in linguistic quality description. However, current methods still fall short in accurately scoring image quality. In this work, we aim to leverage MLLMs to regress accurate quality scores. A key challenge is that the quality score is inherently continuous, typically modeled as a Gaussian distribution, whereas MLLMs generate discrete token outputs. This mismatch necessitates score discretization. Previous approaches discretize the mean score into a one-hot label, resulting in information loss and failing to capture inter-image relationships. We propose a distribution-based approach that discretizes the score distribution into a soft label. This method preserves the characteristics of the score distribution, achieving high accuracy and maintaining inter-image relationships. Moreover, to address dataset variation, where different IQA datasets exhibit various distributions, we introduce a fidelity loss based on Thurstone's model. This loss captures intra-dataset relationships, facilitating co-training across multiple IQA datasets. With these designs, we develop the distribution-based Depicted image Quality Assessment model for Score regression (DeQA-Score). Experiments across multiple benchmarks show that DeQA-Score stably outperforms baselines in score regression. Also, DeQA-Score can predict the score distribution that closely aligns with human annotations. Codes and model weights have been released in https://depictqa.github.io/deqa-score/.

Key facts. Cited by 139 papers in the map (data through 7 Oct 2026). Placed in Image editing and quality assessment, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 68 of its 95 references that are in the map. First posted on arXiv 20 Jan 2025.

Builds on

68 of 95 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024Adaptive Image Quality Assessment via Teaching Large Multimodal Model to Compare29 May
  2. 2024A Comprehensive Study of Multimodal Large Language Models for Image Quality Assessment16 Mar
  3. 20242AFC Prompting of Large Multimodal Models for Image Quality Assessment2 Feb
  4. 2023Depicting Beyond Scores: Advancing Image Quality Assessment through Multi-modal Language Models14 Dec

Other references · 64

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2023GPT-4 Technical Reportcited by 18,132
  3. 2022Training language models to follow instructions with human feedbackcited by 15,712
  4. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  5. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  6. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
  7. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  8. 2023Visual Instruction Tuningcited by 7,792
  9. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,337
  10. 2023InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuningcited by 2,855
  11. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,689
  12. 2016Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answeringcited by 2,498
38 more
  1. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,105
  2. 2018SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processingcited by 1,973
  3. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,881
  4. 2019Towards VQA Models That Can Readcited by 1,696
  5. 2022ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoningcited by 1,449
  6. 2015Microsoft COCO Captions: Data Collection and Evaluation Servercited by 1,447
  7. 2020DocVQA: A Dataset for VQA on Document Imagescited by 1,288
  8. 2021MUSIQ: Multi-scale Image Quality Transformercited by 1,058
  9. 2023mPLUG-Owl: Modularization Empowers Large Language Models with Multimodalitycited by 1,038
  10. 2022Exploring CLIP for Assessing the Look and Feel of Imagescited by 974
  11. 2020Image Quality Assessment: Unifying Structure and Texture Similaritycited by 792
  12. 2023mPLUG-Owl2: Revolutionizing Multi-modal Large Language Model with Modality Collaborationcited by 545
  13. 2022MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessmentcited by 514
  14. 2023Q-Align: Teaching LMMs for Visual Scoring via Discrete Text-Defined Levelscited by 489
  15. 2018nocaps: novel object captioning at scalecited by 444
  16. 2017NIMA: Neural Image Assessmentcited by 387
  17. 2019KonIQ-10k: An ecologically valid database for deep learning of blind image quality assessmentcited by 351
  18. 2019Blind Image Quality Assessment Using A Deep Bilinear Convolutional Neural Networkcited by 301
  19. 2023InternLM-XComposer: A Vision-Language Large Model for Advanced Text-image Comprehension and Compositioncited by 297
  20. 2023Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspectivecited by 276
  21. 2023Q-Bench: A Benchmark for General-Purpose Foundation Models on Low-level Visioncited by 247
  22. 2016Deep Neural Networks for No-Reference and Full-Reference Image Quality Assessmentcited by 224
  23. 2019From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Qualitycited by 221
  24. 2020PIPAL: a Large-Scale Image Quality Assessment Dataset for Perceptual Image Restorationcited by 177
  25. 2023LAMM: Language-Assisted Multi-Modal Instruction-Tuning Dataset, Framework, and Benchmarkcited by 169
  26. 2018PieAPP: Perceptual Image-Error Assessment through Pairwise Preferencecited by 158
  27. 2023Q-Instruct: Improving Low-level Visual Abilities for Multi-modality Foundation Modelscited by 152
  28. 2017RankIQA: Learning from Rankings for No-reference Image Quality Assessmentcited by 150
  29. 2020MetaIQA: Deep Meta-learning for No-Reference Image Quality Assessmentcited by 144
  30. 2016Learning a No-Reference Quality Metric for Single-Image Super-Resolutioncited by 143
  31. 2023AGIQA-3K: An Open Database for AI-Generated Image Quality Assessmentcited by 142
  32. 2020Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and Wildcited by 134
  33. 2024Towards Open-ended Visual Quality Comparisoncited by 85
  34. 2023Vary: Scaling up the Vision Vocabulary for Large Vision-Language Modelscited by 84
  35. 2023AIGCIQA2023: A Large-scale Image Quality Assessment Database for AI Generated Images: from the Perspectives of Quality, Authenticity and Correspondencecited by 61
  36. 2023A Perceptual Quality Assessment Exploration for AIGC Imagescited by 54
  37. 2022Shift-tolerant Perceptual Similarity Metriccited by 44
  38. 2021GraphIQA: Learning Distortion Graph Representations for Blind Image Quality Assessmentcited by 43

50 most cited of 64 All 64 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 71; 2026: 68.

139 papers.

Sort:

By area:

  1. 2025Flow-GRPO: Training Flow Matching Models via Online RLcited by 605
  2. 2025Emu3.5: Native Multimodal Models are World Learnerscited by 119
  3. 2025Q-Insight: Understanding Image Quality via Visual Reinforcement Learningcited by 93
  4. 2025VisualQuality-R1: Reasoning-Induced Image Quality Assessment via Reinforcement Learning to Rankcited by 77
  5. 2025VideoScore2: Think before You Score in Generative Video Evaluationcited by 37
  6. 2025ArtiMuse: Fine-Grained Image Aesthetics Assessment with Joint Scoring and Expert-Level Understandingcited by 36
  7. 2025Toward Generalized Image Quality Assessment: Relaxing the Perfect Reference Quality Assumptioncited by 34
  8. 2025Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preferencecited by 30
  9. 2026Flow-OPD: On-Policy Distillation for Flow Matching Modelscited by 29
  10. 2025LoRA-Edit: Controllable First-Frame-Guided Video Editing via Mask-Aware LoRA Fine-Tuningcited by 27
  11. 2025VQ-Insight: Teaching VLMs for AI-Generated Video Quality Understanding via Progressive Visual Reinforcement Learningcited by 26
  12. 2025AR-GRPO: Training Autoregressive Image Generation Models via Reinforcement Learningcited by 22
38 more
  1. 2026DenseGRPO: From Sparse to Dense Reward for Flow Matching Model Alignmentcited by 20
  2. 2025UniPercept: Towards Unified Perceptual-Level Image Understanding across Aesthetics, Quality, Structure, and Texturecited by 19
  3. 2026Reinforce Adjoint Matching: Scaling RL Post-Training of Diffusion and Flow-Matching Modelscited by 17
  4. 2025Group Critical-token Policy Optimization for Autoregressive Image Generationcited by 16
  5. 2025NTIRE 2025 Challenge on Short-form UGC Video Quality Assessment and Enhancement: Methods and Resultscited by 14
  6. 2025Q-Ponder: A Unified Training Pipeline for Reasoning-based Visual Quality Assessmentcited by 14
  7. 2025CulturalFrames: Assessing Cultural Expectation Alignment in Text-to-Image Models and Evaluation Metricscited by 14
  8. 2025Reinforcing Diffusion Models by Direct Group Preference Optimizationcited by 13
  9. 2026LLaDA2.0-Uni: Unifying Multimodal Understanding and Generation with Diffusion Large Language Modelcited by 12
  10. 2025NTIRE 2025 Challenge on Short-form UGC Video Quality Assessment and Enhancement: KwaiSR Dataset and Studycited by 11
  11. 2026ClearAIR: A Human-Visual-Perception-Inspired All-in-One Image Restorationcited by 11
  12. 2025Transition Matching: Scalable and Flexible Generative Modelingcited by 10
  13. 2025Self-Evolving Vision-Language Models for Image Quality Assessment via Voting and Rankingcited by 10
  14. 2025Consolidating Reinforcement Learning for Multimodal Discrete Diffusion Modelscited by 9
  15. 2026Understanding Reward Hacking in Text-to-Image Reinforcement Learningcited by 9
  16. 2026Alleviating Sparse Rewards by Modeling Step-Wise and Long-Term Sampling Effects in Flow-Based GRPOcited by 9
  17. 2025NTIRE 2025 challenge on Text to Image Generation Model Quality Assessmentcited by 8
  18. 2025RadarQA: Multi-modal Quality Analysis of Weather Radar Forecastscited by 8
  19. 2025PhotoFramer: Multi-modal Image Composition Instructioncited by 8
  20. 2026Generative World Renderercited by 8
  21. 2025Fine-grained Image Quality Assessment for Perceptual Image Restorationcited by 7
  22. 2025Revisiting MLLM Based Image Quality Assessment: Errors and Remedycited by 7
  23. 2026TDM-R1: Reinforcing Few-Step Diffusion Models with Non-Differentiable Rewardcited by 7
  24. 2026OP-GRPO: Efficient Off-Policy GRPO for Flow-Matching Modelscited by 7
  25. 2025LoCoT2V-Bench: Benchmarking Long-Form and Complex Text-to-Video Generationcited by 6
  26. 2026PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusioncited by 6
  27. 2025GenDR: Lighten Generative Detail Restorationcited by 5
  28. 2025Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy Optimizationcited by 5
  29. 2025UARE: A Unified Vision-Language Model for Image Quality Assessment, Restoration, and Enhancementcited by 5
  30. 2025Anchoring Values in Temporal and Group Dimensions for Flow Matching Model Alignmentcited by 5
  31. 2025Restore-R1: Efficient Image Restoration Agents via Reinforcement Learning with Multimodal LLM Perceptual Feedbackcited by 4
  32. 2025Bridging the Perception Gap in Image Super-Resolution Evaluationcited by 3
  33. 2025Position: Agentic Systems Constitute a Key Component of Next-Generation Intelligent Image Processingcited by 3
  34. 2025Q-CLIP: Unleashing the Power of Vision-Language Models for Video Quality Assessment through Unified Cross-Modal Adaptationcited by 3
  35. 2025HandEval: Taking the First Step Towards Hand Quality Evaluation in Generated Imagescited by 3
  36. 2025SoliReward: Mitigating Susceptibility to Reward Hacking and Annotation Noise in Video Generation Reward Modelscited by 3
  37. 2026Flow-Map GRPO: Reinforcement Learning for Few-Step Flow-Map Generators via Anchored Stochastic Compositioncited by 3
  38. 2025DeQA-Doc: Adapting DeQA-Score to Document Image Quality Assessmentcited by 2

50 most cited of 139 All 139 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025iDETEX: Empowering MLLMs for Intelligent DETailed EXplainable IQA
  2. 2026Decoupling Perception and Calibration: Label-Efficient Image Quality Assessment Framework
  3. 2024Enhancing Descriptive Image Quality Assessment with A Large-scale Multi-modal Dataset
  4. 2024Grounding-IQA: Grounding Multimodal Language Model for Image Quality Assessment
  5. 2025Investigate the Low-level Visual Perception in Vision-Language based Image Quality Assessment
  6. 2025Next Token Is Enough: Realistic Image Quality and Aesthetic Scoring with Multimodal Large Language Model
  7. 2026Zoom-IQA: Image Quality Assessment with Reliable Region-Aware Reasoning
  8. 2025Q-SiT: Teaching LMMs for Image Quality Scoring and Interpreting