LongAV-Compass: Towards Unified Evaluation of Minute-Scale Audio-Visual Generation Across T2AV, I2AV, and V2AV

Tengfei Liu, Yang Shi, Xuanyu Zhu, Jiafu Tang, Liu Yang, Qixun Wang, Zhuoran Zhang, Yuqi Tang, Fengxiang Wang, Yuhao Dong, Xinlong Chen, Bozhou Li, Bohan Zeng, Yue Ding, Xiaohan Zhang, Jialu Chen, Haotian Wang, Yuanxing Zhang, Pengfei Wan, Leye Wang

arXiv:2605.26244 · 2026-05-27 공개 · arXiv · PDF

clip dino-v2 multimodal-evaluation v2av audio-visual-generation mllm-assisted long-scale imagebind

Abstract

Audio-visual generation is rapidly advancing from short clips to minute-long content, while existing evaluation protocols remain largely confined to short-form settings. Existing benchmarks primarily focus on 5--10 second text-conditioned generation and rarely support unified evaluation across text, image, and video conditioning modalities. Moreover, they provide limited insight into how identity consistency, narrative coherence, and audio-visual alignment degrade over extended temporal horizons. To bridge this gap, we introduce LongAV-Compass, a systematic benchmark for minute-long audio-visual generation. LongAV-Compass contains 284 curated test cases spanning text-to-audio-video (T2AV), image-to-audio-video (I2AV), and video-to-audio-video (V2AV), organized by application scenario and generation complexity. The benchmark combines taxonomy-guided benchmark construction with a unified evaluation framework that integrates MLLM-assisted assessment with complementary perceptual and multimodal metrics, including DINO-v2, ArcFace, CLIP, and ImageBind. The framework evaluates more than 20 fine-grained dimensions covering within-segment quality, cross-segment consistency, global narrative coherence, semantic alignment, and audio-visual synchronization. Through experiments on 11 representative models together with human-alignment validation, LongAV-Compass provides a diagnostic testbed for analyzing the limitations of current systems in sustaining coherent, semantically aligned, and temporally consistent minute-scale audio-visual generation across diverse input modalities.

한국어 요약

📋 한 줄 요약

**[오디오-비주얼 생성 평가 / 분단위 벤치마크]** LongAV-Compass가 T2AV·I2AV·V2AV 통합 평가의 첫 분단위 벤치마크 — 284 테스트 케이스·20+ fine-grained 차원·MLLM 보조 평가로 11개 모델 대상 identity 일관성·narrative coherence·audio-visual 정렬 한계 진단.

🎯 핵심 기여도

💡 핵심 아이디어

분단위 오디오-비주얼 생성의 신뢰성 있는 평가는 short-form 메트릭의 단순 확장이 아닌 — taxonomy-guided 케이스 큐레이션·multi-modality 통합·MLLM 보조 + DINO-v2/ArcFace/CLIP/ImageBind perceptual 메트릭의 다단계 결합·20+ fine-grained 차원으로 within-segment 품질부터 cross-segment 일관성·global narrative coherence까지 측정해야 한다.

🔬 기술적 접근법

📊 주요 결과

💭 의의 및 한계

**의의**: Minute-long 오디오-비주얼 생성의 첫 통합 평가 표준, multi-modality conditioning 동시 지원으로 cross-modal 비교 가능, MLLM + perceptual 메트릭 결합으로 평가 다층화. **한계**: 284 케이스가 모든 응용 시나리오 커버 한계, MLLM 평가의 평가자 bias, 분단위 너머의 더 긴 horizon 일반화는 후속, human-alignment validation 규모 미명시.

🚀 실용적 활용