AI Paper Atlas

Research map / Video / Video generation

T2AV-Compass: Towards Unified Evaluation for Text-to-Audio-Video Generation

Zhe Cao, Tao Wang, Jiaming Wang, Yanghai Wang, Yuanxing Zhang, Jiahao Wang, Jialu Chen, Miao Deng, Chenxi Liao, Yize Zhang, Yubin Guo, Zhaoxiang Zhang and 1 more

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
increase prompt event structure from single event to long-narrative (4+ events) (ablation)Wan-2.6 · T2AV-Compass event occurrence failure ratefailure rate (%): 22% -> 63%Hurt
increase prompt event structure to long-narrative (4+ events) (ablation)LTX-2 · T2AV-Compass event occurrence failure ratefailure rate (%): reaches 80% failureHurt
increase visual subject count from single to crowd (ablation)T2AV models · T2AV-Compass subject presence failure ratefailure rate (%): 8-10% -> 21-44%Hurt
use closed-source end-to-end models rather than open-source or cascadedT2AV systems · T2AV-Compass subjective AverageAverage score: Veo-3.1 70.29 vs LTX-2 63.72 (best open-source)Improved
use cascaded T2V then V2A pipeline instead of end-to-endWan-2.2 + Hunyuan-Foley · T2AV-Compass Video Realism vs alignmentVideo Realism: 89.63 Video Realism, but lags in audio-visual alignmentMixed
evaluate audio realism versus video realism across modelsT2AV systems · T2AV-Compass realismAudio Realism: best Audio Realism 53.84 vs best Video Realism 89.95Hurt
use Gemini-2.5-Pro as judge instead of Gemini-2.5-Flash or Qwen3-Omni-Flash (ablation)MLLM-as-a-Judge · Human agreement, 50 promptsOverall L1 distance (lower better): 1.087 vs 1.207 and 1.473Improved
use Gemini-2.5-Pro as judge for audio realismMLLM-as-a-Judge · Human agreement, Audio RealismL1 distance: 1.420 vs Inter-Human 1.042Hurt
repeat judging three times with Gemini 2.5 Pro (ablation)MLLM judge pipeline · 500 video samplescoefficient of variation: CV <= 1.02%No clear effect
swap judge to Gemini 2.5 Flash for five-system ranking (ablation)MLLM judge pipeline · five-system rankingSpearman rho: rho = 0.9248No clear effect

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