AI Paper Atlas

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

OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM

Hanrong Ye, Chao-Han Huck Yang, Arushi Goel, Wei Huang, Zhen Wan, Jinchuan Tian, An-Chieh Cheng, Ligeng Zhu, Yuanhang Su, Yuming Lou, Yong-Xiang Lin, Dong Yang and 20 more

ICLR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multi-modality image fusion

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 Temporal Embedding Grouping to token concatenation (ablation)OmniVinci (10B-token ablation) · Omni averageaverage accuracy: 45.51 -> 47.72Improved
use learned time embedding for absolute timestamps (ablation)TEG baseline · Omni averageaverage accuracy: 47.72 -> 47.30No clear effect
use RoTE for absolute timestamps (ablation)TEG baseline · Omni averageaverage accuracy: 47.72 -> 47.80No clear effect
add Constrained Rotary Time Embedding (ablation)TEG baseline · Omni averageaverage accuracy: 47.72 -> 50.25Improved
add OmniAlignNet on top of TEG and CRTE (ablation)OmniVinci (10B-token ablation) · Omni averageaverage accuracy: 50.25 -> 52.59Improved
add implicit learning from video audio (ablation)visual-only model · VideoMME w/o subtitlesaccuracy: 61.67 -> 63.76Improved
add implicit learning from video audio (ablation)visual-only model · VideoMME Shortaccuracy: 74.22 -> 71.31Hurt
add explicit learning with omni-modal data engine (ablation)Visual + Audio (IL) · VideoMME w/o subtitlesaccuracy: 63.76 -> 67.37Improved
add GRPO post-training with omni-modal MCQ data (ablation)OmniVinci · Omni averageaverage accuracy: 53.73 -> 54.52Improved
include audio input in GRPO training (ablation)OmniVinci RL · GRPO accuracy rewardaccuracy reward: +0.1 higher convergenceImproved
apply AWQ 4-bit quantizationOmniVinci · Tennis MCQinference time (s): 3.29 -> 1.85Improved

Later papers that compared with this method

1 later accepted paper reports OmniVinci as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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