Research map / Image generation / Autoregressive and latent generation
No-Reference Image Quality Assessment with Reinforcement Recursive List-Wise Ranking
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train with reinforcement list-wise ranking instead of regression on VSI scores (ablation) | ResNet-50 · TID2013 (opinion-unaware) | SROCC: 0.754 -> 0.885 | Helped |
| train with reinforcement list-wise ranking instead of regression on VSI scores (ablation) | ResNet-50 · LIVE (opinion-unaware) | SROCC: 0.838 -> 0.949 | Helped |
| train with reinforcement list-wise ranking instead of regression on VSI scores (ablation) | ResNet-50 · CSIQ (opinion-unaware) | SROCC: 0.764 -> 0.915 | Helped |
| train with ranking instead of MOS regression (ablation) | ResNet-50 · TID2013 (opinion-aware) | SROCC: 0.712 -> 0.806 | Helped |
| train with ranking instead of MOS regression (ablation) | ResNet-50 · CSIQ (opinion-aware) | SROCC: 0.876 -> 0.907 | Helped |
| train with ranking instead of MOS regression (ablation) | ResNet-50 · LIVE (opinion-aware) | SROCC: 0.950 -> 0.956 | Helped |
| increase discount factor gamma from 0 toward 1 (ablation) | RRLRIQA · LIVE and TID2013 | SROCC | Hurt |
| increase training list size N from 5 to 10 or 15 (ablation) | RRLRIQA · LIVE and TID2013 | SROCC: SROCC std 0.0119/0.0080 (N=10/15) on LIVE | Helped |
| proposed method vs best opinion-unaware competitor dipIQ | RRLRIQA · LIVE | SROCC: 0.946 -> 0.949 | No clear effect |
| proposed method vs opinion-unaware competitors | RRLRIQA · TID2013 | SROCC: 0.872 -> 0.885 | Helped |
Closest papers by meaning
- Q-Insight: Understanding Image Quality via Visual Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- VisualQuality-R1: Reasoning-Induced Image Quality Assessment via Reinforcement Learning to Rank (NeurIPS 2025 · similarity 0.88)
- Refine-IQA: Multi-Stage Reinforcement Finetuning for Perceptual Image Quality Assessment (AAAI 2026 · similarity 0.87)
- Reasoning as Representation: Rethinking Visual Reinforcement Learning in Image Quality Assessment (ICLR 2026 · similarity 0.86)
- Adaptive Image Quality Assessment via Teaching Large Multimodal Model to Compare (NeurIPS 2024 · similarity 0.86)
- Adaptive Feature Selection for No-Reference Image Quality Assessment by Mitigating Semantic Noise Sensitivity (ICML 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.