Research map / Image generation / Autoregressive and latent generation
DSMix: Distortion-Induced Saliency Map Based Pre-training for No-Reference Image Quality Assessment
Jinsong Shi, Pan Gao, Xiaojiang Peng, Jie Qin
This paper tried a common change: Add a reconstruction / auxiliary loss
add feature reconstruction loss to distillation · CSIQ · SRCC: 0.933 -> 0.941 Improved
What 987 accepted papers report for this change
- 882 improved
- 41 no clear effect
- 20 mixed
- 44 hurt
- OPTION: Optimal Transport–Guided Flow Matching for Incomplete and Unaligned Multi-View Clustering ICML 2026 · add reconstruction loss L_rec · Scene-15 (aligned, 70% missing) · ACC: 32.4 -> 37.5Improved
- Detached Skip-Links and $R$-Probe: Decoupling Feature Aggregation from Gradient Propagation for MLLM OCR ICML 2026 · add R-probe auxiliary loss · OCRBench · score: 714.0 -> 721.0Improved
- Rethinking Time-Series Imputation as Conditional Inference along Temporal Evolution ICML 2026 · add reconstruction loss on observed entries (joint vs missing-only supervision) · ETTh1 10% missing · Masked MAE: 0.2002 -> 0.1658Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add DSMix augmentation to pretraining (ablation) | Quality encoder (ResNet-50) · CLIVE | SRCC: 0.773 -> 0.824 | Improved |
| add semantic knowledge distillation from ImageNet ResNet-50 (ablation) | DSMix pretraining · CLIVE | SRCC: 0.824 -> 0.873 | Improved |
| add cosine distance loss to distillation (ablation) | KD · CSIQ | SRCC: 0.941 -> 0.957 (with L_FR); 0.933 -> 0.950 alone | Improved |
| add feature reconstruction loss to distillation (ablation) | KD · CSIQ | SRCC: 0.933 -> 0.941 | Improved |
| use more KADIS pretraining data (ablation) | DSMix pretraining · CSIQ | SRCC: 0.913 -> 0.948 -> 0.957 (25%, 50%, 100%) | Improved |
| use larger encoder (ablation) | Quality encoder · CLIVE | SRCC: 0.831 -> 0.857 -> 0.873 (ResNet-18/34/50) | Improved |
| replace HyperIQA pretrained weights with DSMix | HyperIQA · FLIVE | SRCC: 0.544 -> 0.631 | Improved |
| replace TReS pretrained weights with DSMix | TReS · KADID | SRCC: 0.859 -> 0.904 | Improved |
| DSMix linear probing vs prior methods on authentic datasets | DSMix · KonIQ | SRCC: 0.915 vs QPT 0.927 | No clear effect |
Closest papers by meaning
- Transformer-Based No-Reference Image Quality Assessment via Supervised Contrastive Learning (AAAI 2024 · similarity 0.86)
- Distilling Spatially-Heterogeneous Distortion Perception for Blind Image Quality Assessment (CVPR 2025 · similarity 0.86)
- Towards Syn-to-Real IQA: A Novel Perspective on Reshaping Synthetic Data Distributions (NeurIPS 2025 · similarity 0.86)
- Causal-IQA: Towards the Generalization of Image Quality Assessment Based on Causal Inference (ICML 2024 · similarity 0.85)
- Toward Generalized Image Quality Assessment: Relaxing the Perfect Reference Quality Assumption (CVPR 2025 · similarity 0.85)
- Adaptive Image Quality Assessment via Teaching Large Multimodal Model to Compare (NeurIPS 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.
