AI Paper Atlas

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

OmniDPO: A Preference Optimization Framework to Address Omni-Modal Hallucination

Junzhe Chen, Tianshu Zhang, Shiyu Huang, Yuwei Niu, Chao Sun, Rongzhou Zhang, Guanyu Zhou, Lijie Wen

AAAI 2026 · Vision-language and multimodal › Cross-modal learning · approach: Audio-visual speech separation · 7 citations (OpenAlex)

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 modality-aware visual and audio preference losses on top of text DPOQwen2.5-Omni-7B · AVHBench Audio-driven Video Hallucinationaccuracy: 71.74 -> 84.42Improved
add text-only DPOQwen2.5-Omni-7B · AVHBench Audio-driven Video Hallucinationaccuracy: 74.12 -> 71.74Hurt
add VCD contrastive decodingQwen2.5-Omni-7B · AVHBench Audio-driven Video Hallucinationaccuracy: 74.12 -> 77.40Improved
add text-only DPOMiniCPM-o-2.6 · CMM overallhallucination resistance (hr): 63.9 -> 66.2Improved
add OmniDPOMiniCPM-o-2.6 · CMM overallhallucination resistance (hr): 63.9 -> 68.8Improved
add audio preference loss only (no visual loss) (ablation)OmniDPO on Qwen2.5-Omni-7B · AVHBench Video-driven Audio Hallucinationaccuracy: 67.6 -> 76.7Improved
add video preference loss only (no audio loss) (ablation)OmniDPO on Qwen2.5-Omni-7B · CMM Audio-Language (AL)hallucination resistance (hr): 78.0 -> 63.0Hurt
add video preference loss only (no audio loss) (ablation)OmniDPO on Qwen2.5-Omni-7B · CMM Visual-Audio-Language (VAL)hallucination resistance (hr): 90.5 -> 66.0Hurt
combine audio and video preference losses (ablation)OmniDPO on Qwen2.5-Omni-7B · CMM Audio-Language (AL)hallucination resistance (hr): 82.5 (audio only) -> 83.5Improved
add video preference loss only (ablation)OmniDPO on Qwen2.5-Omni-7B · AVHBench Audio-driven Video Hallucinationaccuracy: 74.1 -> 82.4Improved
add OmniDPO trainingQwen2.5-Omni-7B · AVHBench Audio-driven Video HallucinationYes rate (%): 64.44 -> 44.28Improved
add OmniDPO trainingQwen2.5-Omni-7B and MiniCPM-o-2.6 · MMAU and MMMU (zero-shot)accuracy: average +2.4% and +3.6%Improved

Later papers that compared with this method

2 later accepted papers report OmniDPO as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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