Research map / Label-efficient and robust learning / Recognition and long-tail
Feature Coding in the Era of Large Models: Dataset, Test Conditions, and Benchmark
Changsheng Gao, Yifan Ma, Qiaoxi Chen, Yenan Xu, Dong Liu, Weisi Lin
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| use VTM codec instead of Hyperprior as the core feature codec | feature coding baseline · Image classification | Accuracy near 2 BPFP: 93.20 -> 99.80 | Improved |
| use VTM instead of Hyperprior for text features | feature coding baseline · Common sense reasoning | Distortion-accuracy R^2: 0.4713 -> 0.7490 | Improved |
| reduce bitrate by raising QP in VTM baseline | VTM baseline · Image classification | Accuracy: 99.80 -> 7.20 | Hurt |
| use feature MSE as a proxy for task accuracy | both baselines · Common sense reasoning (text features) | Distortion-accuracy R^2: 0.4713 | Hurt |
| use image codecs on generated-image features | both baselines · Text-to-image synthesis | Distortion-accuracy R^2: 0.9983 and 0.9996 | Improved |
| apply a learned codec trained on another task's features | Hyperprior baseline · Cross-task generalization (Seg, CSR, TTI) | Accuracy / FID | Hurt |
| run VTM encoder on large-model features | VTM baseline · Complexity | Encoding time | Hurt |
| replace Bits Per Pixel with Bits Per Feature Point as bitrate measure | test conditions · metric design | bitrate fairness | Improved |
Closest papers by meaning
- Collaborative Decoding Makes Visual Auto-Regressive Modeling Efficient (CVPR 2025 · similarity 0.85)
- Language Modeling Is Compression (ICLR 2024 · similarity 0.85)
- Beyond Matryoshka: Revisiting Sparse Coding for Adaptive Representation (ICML 2025 · similarity 0.84)
- Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation (ICLR 2026 · similarity 0.84)
- Diversify, Contextualize, and Adapt: Efficient Entropy Modeling for Neural Image Codec (NeurIPS 2024 · similarity 0.84)
- Enhancing Dataset Distillation via Non-Critical Region Refinement (CVPR 2025 · similarity 0.84)
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.
