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 box
1 comments
WinARDet: Compressed Window Auto-Regression for Long-Horizon Vision-Language Grounding was first seen on Hugging Face · Daily papers on Oct 21, 2026. Since then it appeared in 2 hourly checks across 2 days, peaking at #30 on Oct 21, 23:18 UTC. It was last on the list on Oct 22, 01:18 UTC.
Position on Hugging Face · Daily papers at every check (lower is better).
Best rank per day (UTC).
Platform metric as reported by Hugging Face.
Show that this trended, linking back to this page.
<a href="https://trendoday.com/hugging-face/papers/2610.19419-winardet-compressed-window-auto-regression-for-long-horizon"><img src="https://trendoday.com/badge/hugging-face/papers/2610.19419-winardet-compressed-window-auto-regression-for-long-horizon.svg" alt="WinARDet: Compressed Window Auto-Regression for Long-Horizon Vision-Language Grounding on TrendBoard" height="44"></a>
[](https://trendoday.com/hugging-face/papers/2610.19419-winardet-compressed-window-auto-regression-for-long-horizon)