AI Paper Atlas

Research map / Video / Video generation

Towards Streaming Synchronized Spatial Audio Generation via Autoregressive Diffusion Transformer

Ke Lei, Yu Zhang, Changhao Pan, Xueyi Pu, Wenxiang Guo, Ruiqi Li, Zhou Zhao

ICML 2026 · Video › Video generation · approach: Video-to-audio synthesis · 0 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 VideoMAE encoder with SVAC contrastive learning instead of frame-level CLIP features (ablation)SwanSphere video encoder · video-to-spatial audioFD / angular error: FD 140.28 -> 120.28; angular 1.34 -> 1.03Improved
add physics-based negatives (temporal offset, audio and video rotation) to semantic negatives in SVAC (ablation)SVAC · video-to-spatial audioangular error: 1.12 -> 1.03Improved
add multi-objective ODPO fine-tuning (ablation)SwanSphere-L · video-to-spatial audioFD / angular error: FD 133.91 -> 120.28; angular 1.22 -> 1.03Improved
add history conditioning via history encoder (ablation)SwanSphere · video-to-spatial audioFD: 128.15 -> 120.28Improved
scale model from 0.43B to 1.09B (ablation)SwanSphere · video-to-spatial audioFD / angular error: FD 139.81 -> 120.28; angular 1.33 -> 1.03Improved
use causal autoregressive diffusion (LocDiT) instead of a global DiT (ablation)SwanSphere · video-to-spatial audiofirst-chunk latency / FD: 6.47s -> 0.21s; FD 123.08 -> 120.28Improved
use global DiT instead of autoregressive diffusion (ablation)SwanSphere · video-to-spatial audioelevation error: 1.14 -> 0.91 (elevation error)Mixed
add caption conditioning to panoramic video (ablation)SwanSphere · caption test subsetFD / angular error: FD 120.28 -> 118.31; angular 1.03 -> 0.96Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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