AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

OmniFit: Bridging Modalities via Layer-Adaptive Token Compression for Omnimodal Large Language Models

Zining Wang, Zhihang Yuan, Yingjie Zhai, Wenshuo Li, Han Shu, Ruihao Gong, Jinyang Guo, Xianglong Liu

ICML 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal fusion with missing modalities

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add LAHP to RandomDrop (ablation)OmniFit on Qwen2.5-Omni-3B · VideoMME, retain 20%accuracy: 50.1 -> 55.3Improved
add LAHP to ARTS (ablation)OmniFit on Qwen2.5-Omni-3B · VideoMME, retain 20%accuracy: 55.4 -> 62.0Improved
add modality preference profiling on top of ARTS w/ TRP (ablation)OmniFit on Qwen2.5-Omni-3B · WorldSense, retain 20%accuracy: 36.5 -> 45.1Improved
add token redundancy profiling to ARTS (ablation)OmniFit on Qwen2.5-Omni-3B · WorldSense, retain 20%accuracy: 28.6 -> 36.5Improved
replace RandomDrop with ARTS selection (ablation)OmniFit on Qwen2.5-Omni-3B · VideoMME, retain 20%accuracy: 50.1 -> 55.4Improved
use cumulative product schedule instead of linear decay (ablation)OmniFit on Qwen2.5-Omni-3B · VideoMME, retain 20%accuracy: 59.8 -> 62.0Improved
use cumulative product schedule instead of cosine decay (ablation)OmniFit on Qwen2.5-Omni-3B · MLVU, retain 20%accuracy: 65.1 -> 67.2Improved
increase calibration samples from 16 to 256 (ablation)OmniFit on Qwen2.5-Omni-3B · VideoMME, retain 20%accuracy: 58.2 -> 61.9Improved
increase calibration samples from 256 to 2048 (ablation)OmniFit on Qwen2.5-Omni-3B · VideoMME, retain 20%accuracy: 61.9 -> 62.0No clear effect
change calibration dataset from AVQA+Ola to other sources (ablation)OmniFit on Qwen2.5-Omni-3B · Daily-Omniaccuracy: 59.8 to 60.3No clear effect
vary DPC-KNN K and M (ablation)OmniFit on Qwen2.5-Omni-3B · VideoMME, retain 20%accuracy: 61.4 to 62.0No clear effect
use pruning instead of merging for dropped tokens (ablation)OmniFit on Qwen2.5-Omni-3B · VideoMME, retain 20%accuracy: 62.0 -> 60.5Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy