| title | WinARDet: Compressed Window Auto-Regression for Long-Horizon Vision-Language Grounding |
|---|---|
| authors | Chenxu Wang, Hangyi Kuang, Bo-Wen Yin, Yuxuan Li, Quan-Sheng Zeng, Yang Zhang, Xu Guo, Xiang Li, Qibin Hou |
| date | 2026-10-14 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2610.19419 |
| affiliation | Nankai University |
| upvotes | 3 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | MLLM-based detectors formulate vision-language grounding as autoregressive coordinate generation, yet in dense scenes, they often collapse into a box repetition loop, repeatedly emitting identical boxes until the decoding budget is exhausted. We trace this failure to historical-answer attention takeover: Under full-history autoregression, repeated coordinate tokens in the growing answer history attract disproportionate attention, crowding out the attention sinks needed to attend to new visual evidence. % To break this loop, in this paper, we propose Compressed Window Auto-Regression (WinAR). This new mechanism partitions a long answer sequence into consecutive generation windows and restricts full token-level autoregression to the current window only while compressing each previous window into a fixed set of latent tokens that condition subsequent decoding. This design reduces the repeated exposure to raw historical coordinates that triggers the attention takeover, yet preserves cross-window dependencies essential for coherent multi-object generation. On RSDense, which contains 200-2,000 objects per image, our model WinARDet improves RexOmni F1-score (IoU=0.5) from 32.16 to 39.36 over the baseline, while reducing the redundancy rate from 35.48% to 13.00%. WinARDet also generalizes across decoding strategies and output formats, including parallel box decoding and polygon generation, providing a practical paradigm for long-horizon vision-language grounding across scene densities. |
cite
@misc{wang2026winardet,
title={WinARDet: Compressed Window Auto-Regression for Long-Horizon Vision-Language Grounding},
author={Chenxu Wang and Hangyi Kuang and Bo-Wen Yin and Yuxuan Li and Quan-Sheng Zeng and Yang Zhang and Xu Guo and Xiang Li and Qibin Hou},
year={2026},
eprint={2610.19419},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2610.19419}
}