Metraipolis › archive › dc_9e00a865b313

Position Is All You Need: A Free Lunch Token Compression Strategy for MLLM-based Referring Expression Segmentation — arXiv:2608.26142

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.26142
License (as archived): CC0-1.0
Archived: 2026-10-17T14:24:03.787Z
SHA-256 of the archived text: c6e27dc11bad56cc684434485dda1c26d379ec21a2903f7067115b769ae4e4d2
Stable URL of this page: https://metraipolis.com/documento/dc_9e00a865b313

How to cite

Position Is All You Need: A Free Lunch Token Compression Strategy for MLLM-based Referring Expression Segmentation — arXiv:2608.26142 — arXiv (en), https://arxiv.org/abs/2608.26142 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9e00a865b313 (Metraipolis archive dc_9e00a865b313, sha256 c6e27dc11bad56cc684434485dda1c26d379ec21a2903f7067115b769ae4e4d2)

BibTeX

@misc{dc9e00a865b313,
  title = {Position Is All You Need: A Free Lunch Token Compression Strategy for MLLM-based Referring Expression Segmentation — arXiv:2608.26142},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.26142},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9e00a865b313; sha256 c6e27dc11bad56cc684434485dda1c26d379ec21a2903f7067115b769ae4e4d2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26142. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Referring Expression Segmentation (RES) aims to generate pixel-wise segmentation masks from complex and implicit textual queries. While recent advances in Multimodal Large Language Models (MLLMs) have substantially boosted RES performance, their prohibitive computational overhead remains a critical bottleneck, which, however, is rarely explored. To fill this gap, we first evaluate typical token compression methods on this task and observe a surprising performance degradation. In this paper, we aim to understand this phenomenon for a solution. By extensive experiments, we find that token compression for RES requires preserving the original position embeddings and local neighboring spatial structures, indicating that visual token position information is far more critical than in other tasks. Building on this insight, we ask: Can we design the token compression method purely based on the position information? Therefore, we propose PAYN, a plug-and-play, training-free token compression method that relies solely on position information. PAYN retains tokens that are adequately distributed in every local neighboring region while strictly preserving original positional indices, thereby maintaining spatial relational consistency. Experiments on multiple RES benchmarks demonstrate that our method outperforms existing token compression methods, verifying that position is indeed all you need for token compression in the MLLM-based RES task. Codes are avaliable at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.26142 — 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 2026-10-17T13:16:53.589Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).