跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2607.21179 · v2

Out of Sight, Still in Mind: Token Compression for Omni-LLMs

首次提交 2026-07-23 19:09
版本更新 2026-11-17 19:17
雷达观测 2026-11-18 11:33
arXiv ↗PDF ↗

摘要

The goal of this paper is to reduce the input token cost of Omni-modal large language models (Omni-LLMs) at inference time. Omni-LLMs reason jointly over audio, video and text, but the cost of the three streams is highly unbalanced: visual tokens account for the vast majority of the input, and are highly redundant. In this paper, we propose ReMo, a training-free framework that compresses visual tokens by redistributing their information across modalities: a visual token is kept only if its information appears nowhere else. ReMo achieves this in two ways: (i) it aligns audio and video in a shared space via closed-form calibration, and removes visual tokens already explained by the audio or by other visual tokens; and (ii) it replaces object-level visual tokens, identified by a lightweight object detector, with compact text proxies, short descriptions of each object and its location, conveying the same content in far fewer tokens. On Qwen2.5-Omni at two model scales, ReMo removes 54% of the input tokens with comparable accuracy. It performs comparably to the full-token model, reaching 101.2% and 101.4% of its average accuracy over five audio-visual benchmarks.

作者备注:Preprint

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v2,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-17 19:17 · 最新观测版本

Out of Sight, Still in Mind: Token Compression for Omni-LLMs

The goal of this paper is to reduce the input token cost of Omni-modal large language models (Omni-LLMs) at inference time. Omni-LLMs reason jointly over audio, video and text, but the cost of the three streams is highly unbalanced: visual tokens account for the vast majority of the input, and are highly redundant. In this paper, we propose ReMo, a training-free framework that compresses visual tokens by redistributing their information across modalities: a visual token is kept only if its information appears nowhere else. ReMo achieves this in two ways: (i) it aligns audio and video in a shared space via closed-form calibration, and removes visual tokens already explained by the audio or by other visual tokens; and (ii) it replaces object-level visual tokens, identified by a lightweight object detector, with compact text proxies, short descriptions of each object and its location, conveying the same content in far fewer tokens. On Qwen2.5-Omni at two model scales, ReMo removes 54% of the input tokens with comparable accuracy. It performs comparably to the full-token model, reaching 101.2% and 101.4% of its average accuracy over five audio-visual benchmarks.

查看此版本解读 →查看此版本原文 ↗