Research map / Image generation / Autoregressive and latent generation
Residual Decoder Adapter: ID-Preserving Tokenizer Adaption for Autoregressive Text Rendering
Dongxing Mao, Alex Jinpeng Wang, Jiahao Tang, Kevin Qinghong Lin, Linjie Li, Zhengyuan Yang, Lijuan Wang, Min Li, Jingru Tan
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 |
|---|---|---|---|
| add RDA | Janus Pro* 1B text-specific · StyledTextVisionBlend | OCR accuracy: 24.52 -> 58.26 | Improved |
| add RDA | Janus Pro* 1B text-specific · StyledTextSynth | OCR accuracy: 12.75 -> 36.81 | Improved |
| add RDA | Janus Pro 7B general · AnyText-Benchmark | OCR accuracy: 8.85 -> 10.07 | Improved |
| add RDA | Janus Pro 7B general · Mario-Eval | OCR accuracy: 6.75 -> 8.33 | Improved |
| add RDA | LlamaGen-VQ tokenizer · AnyText-Benchmark reconstruction (low resolution) | OCR accuracy: 21.26 -> 36.79 | Improved |
| add RDA | LlamaGen-VQ tokenizer · AnyText-Benchmark reconstruction (low resolution) | LPIPS: 0.1912 -> 0.1832 | Improved |
| add residual perceptual loss (ablation) | RDA · AnyText-Benchmark reconstruction (high resolution) | OCR accuracy: 58.30 -> 66.48 | Improved |
| add Sobel edge loss (ablation) | RDA · AnyText-Benchmark reconstruction (high resolution) | OCR accuracy: 64.31 -> 66.48 | Improved |
| add final-image perceptual loss (ablation) | RDA · AnyText-Benchmark reconstruction (high resolution) | OCR accuracy: 64.58 -> 66.48 | Improved |
| add frequency-domain loss (ablation) | RDA · AnyText-Benchmark reconstruction (high resolution) | OCR accuracy: 66.13 -> 66.48 | No clear effect |
| add residual decoder without hint codebook (ablation) | LlamaGen-VQ tokenizer · AnyText-Benchmark reconstruction (high resolution) | OCR accuracy: 58.04 -> 64.25 | Improved |
| add hint codebook (dim 16) (ablation) | RDA residual decoder · AnyText-Benchmark reconstruction (high resolution) | OCR accuracy: 64.25 -> 66.48 | Improved |
Closest papers by meaning
- Speculative Jacobi-Denoising Decoding for Accelerating Autoregressive Text-to-image Generation (NeurIPS 2025 · similarity 0.89)
- Vision Foundation Models as Effective Visual Tokenizers for Autoregressive Generation (NeurIPS 2025 · similarity 0.89)
- LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding (ICLR 2025 · similarity 0.88)
- ImageFolder: Autoregressive Image Generation with Folded Tokens (ICLR 2025 · similarity 0.88)
- Hawk: Leveraging Spatial Context for Faster Autoregressive Text-to-Image Generation (NeurIPS 2025 · similarity 0.88)
- When Worse is Better: Navigating the Compression Generation Trade-off In Visual Tokenization (NeurIPS 2025 · similarity 0.88)
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.
