Research map / Detection and segmentation / Open-set and few-shot detection
LlamaSeg: Image Segmentation via Autoregressive Mask Generation
Jiru Deng, Tengjin Weng, Tianyu Yang, Wenhan Luo, Zhiheng Li, Wenhao Jiang
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 |
|---|---|---|---|
| increase mask token resolution to 384 pix. | LlamaSeg-L · ADE20K | mIoU: 52.0 -> 55.9 | Improved |
| increase mask token resolution to 384 pix. | LlamaSeg-L · COCO-Stuff | mIoU: 55.7 -> 58.1 | Improved |
| increase mask token resolution to 384 pix. | LlamaSeg-L · PAS-20 | mIoU: 75.1 -> 77.1 | Improved |
| add SA-OVRS plus referring-data pre-training (ablation) | LlamaSeg-B · ADE20K | mIoU: 48.9 -> 50.5 | Improved |
| add SA-OVRS plus referring-data pre-training (ablation) | LlamaSeg-B · refCOCO | cIoU: 46.0 -> 50.9 | Improved |
| fine-tune VQGAN tokenizer on masks for 5000 steps (ablation) | VQGAN mask tokenizer · mask reconstruction | total IoU: 96.8 -> 93.0 | Hurt |
| fine-tune VQGAN tokenizer on masks for 10000 steps (ablation) | VQGAN mask tokenizer · mask reconstruction | mAHD: 6.1 -> 10.6 | Hurt |
| use greedy search instead of beam search (B=3) (ablation) | LlamaSeg-B · RefCOCO val | cIoU: 47.3 -> 50.9 | Improved |
| use greedy search instead of random sampling (ablation) | LlamaSeg-B · RefCOCO val | cIoU: 49.2 -> 50.9 | Improved |
| use greedy search instead of Top-P (P=0.9) (ablation) | LlamaSeg-B · RefCOCO val | cIoU: 50.2 -> 50.9 | Improved |
| scale LlamaSeg from B to L | LlamaSeg · refCOCO val | cIoU: 50.9 -> 56.5 | Improved |
| scale LlamaSeg from B to L | LlamaSeg · refCOCO+ val | cIoU: 38.9 -> 41.6 | Improved |
Closest papers by meaning
- Seg-VAR:Image Segmentation with Visual Autoregressive Modeling (NeurIPS 2025 · similarity 0.88)
- ARGenSeg: Image Segmentation with Autoregressive Image Generation Model (NeurIPS 2025 · similarity 0.87)
- Text4Seg: Reimagining Image Segmentation as Text Generation (ICLR 2025 · similarity 0.87)
- LLaFS: When Large Language Models Meet Few-Shot Segmentation (CVPR 2024 · similarity 0.87)
- ALTo: Adaptive-Length Tokenizer for Autoregressive Mask Generation (NeurIPS 2025 · similarity 0.86)
- Exploring Simple Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.86)
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.
