Metraipolis › archive › dc_ac42b2b60fcb

GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed — arXiv:2609.39600

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.39600
License (as archived): CC0-1.0
Archived: 2026-10-07T12:45:34.965Z
SHA-256 of the archived text: e093db9e470e63cfd8ea048e0ee5e140ded4312752a1f9b8b213ea02dff68309
Stable URL of this page: https://metraipolis.com/documento/dc_ac42b2b60fcb

How to cite

GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed — arXiv:2609.39600 — arXiv (en), https://arxiv.org/abs/2609.39600 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ac42b2b60fcb (Metraipolis archive dc_ac42b2b60fcb, sha256 e093db9e470e63cfd8ea048e0ee5e140ded4312752a1f9b8b213ea02dff68309)

BibTeX

@misc{dcac42b2b60fcb,
  title = {GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed — arXiv:2609.39600},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.39600},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ac42b2b60fcb; sha256 e093db9e470e63cfd8ea048e0ee5e140ded4312752a1f9b8b213ea02dff68309},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.39600. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Autoregressive (AR) grounding models serialize spatial predictions, introducing sequential latency and imposing a causal order on output tokens. We view grounding as visual evidence extraction: objects, locations, and spatial relations are jointly constrained by the image and query, yet their dependencies do not imply an intrinsic left-to-right generation order. This distinction makes bidirectional diffusion a natural fit, allowing spatial hypotheses to emerge in parallel and be jointly refined through iterative denoising. We introduce GroundAnything, a 4B-parameter grounding foundation model that reconciles fast parallel decoding with precise localization through blockwise denoising. Training combines grounding pretraining from public datasets and dedicated data engines, direct AR-to-diffusion conversion with joint AR and diffusion objectives, supervised fine-tuning, and GRPO-based reinforcement post-training. Across 30 grounding benchmarks, our autoregressive variant, GroundAnything-VLM, establishes a new overall state of the art among similarly sized models at 72.42%, remaining competitive with GPT-6 Astra (71.35%). With entropy-guided decoding, GroundAnything also surpasses the prior state of the art at this scale, averaging 61.75% versus 53.32% for the fast MTP-based LocateAnything model. We further explore decoding strategies, showing that an optional self-speculative mode achieves a $4.51\times$ speedup over the AR counterpart with a 0.74 percentage-point drop in COCO F1mIoU. Infrastructure experiments show that progressive inference optimizations translate parallel decoding into practical speedups. These support efficient visual grounding in latency-sensitive real-world systems.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.39600 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T12:14:14.115Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).