AI Paper Atlas

Research map / Video / Video generation

OmniForcing: Unleashing Real-time Joint Audio-Visual Generation

Yaofeng Su, Yuming Li, Zeyue Xue, Jie Huang, Siming Fu, Haoran Li, Haoyang Huang, Nan Duan

ECCV 2026 · Video › Video generation · approach: Video-to-audio synthesis · 22 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
use Identity RoPE on sink tokens instead of incremental positions (ablation)OmniForcing Stage II · Stage II stabilization ablationloss (sigma=0.5): 0.402 vs. 0.081Improved
add audio sink tokens (S>=4) versus no stabilizer (ablation)OmniForcing Stage II · Stage II stabilization ablationconvergence: No Stabilizer NaN; S=4 loss 0.141Improved
use S=1 or S=2 sink tokens instead of S>=4 (ablation)OmniForcing Stage II · Stage II stabilization ablationconvergence: NaN gradients at S=2 and S=1Hurt
increase sink pool from 4 to 16 tokens (ablation)OmniForcing Stage II · Stage II stabilization ablationloss (sigma=0.5): 0.141 -> 0.081Improved
increase sink pool from 16 to 24 tokens (ablation)OmniForcing Stage II · Stage II stabilization ablationloss (sigma=0.5): 0.081 -> 0.110Hurt
replace QK-Norm with audio sink tokens (ablation)OmniForcing Stage II · Stage II stabilization ablationloss (sigma=0.5): 0.232 -> 0.081Improved
replace Tanh-Gated Attention with audio sink tokens (ablation)OmniForcing Stage II · Stage II stabilization ablationloss (sigma=0.5): 1.258 -> 0.081Improved
increase sink pool from 8 to 16 tokens (ablation)OmniForcing Stage II · Stage II stabilization ablationmax gradient norm: 21.95 -> 9.23Improved
distill bidirectional LTX-2 into streaming OmniForcingLTX-2 19B · JavisBenchruntime: 197s -> 5.7sImproved
distill bidirectional LTX-2 into streaming OmniForcingLTX-2 19B · JavisBenchJavisScore: 0.253 -> 0.208Hurt
distill bidirectional LTX-2 into streaming OmniForcingLTX-2 19B · VBench (same prompts)aesthetic quality: 0.569 -> 0.595Improved
distill bidirectional LTX-2 into streaming OmniForcingLTX-2 19B · JavisBenchFVD: 125.4 -> 137.2Hurt

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