Research map / Video / Video-language understanding
LiViBench: An Omnimodal Benchmark for Interactive Livestream Video Understanding
Xiaodong Wang, Langling Huang, Zhirong Wu, Xu Zhao, Teng Xu, Xuhong Xia, Peixi Peng
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add audio (ablation) | LiVi-LLM-7B · LiViBench Overall | accuracy: 61.4 -> 63.9 | Improved |
| add ASR speech (ablation) | LiVi-LLM-7B · LiViBench Overall | accuracy: 61.4 -> 63.4 | Improved |
| add audio (ablation) | MiniCPM-o-26 · LiViBench Overall | accuracy: 56.0 -> 54.7 | Hurt |
| add audio (ablation) | Qwen2.5-Omni-7B · LiViBench Overall | accuracy: 57.8 -> 60.3 | Improved |
| add ASR speech (ablation) | MiniCPM-o-26 · LiViBench Overall | accuracy: 56.0 -> 57.9 | Improved |
| add ASR speech (ablation) | LiVi-LLM-7B · LiViBench Fine tasks | accuracy: 69.1 -> 68.7 | Hurt |
| add raw comments (ablation) | LiVi-LLM-7B · LiViBench Overall | accuracy: 63.9 -> 63.0 | Hurt |
| add raw comments (ablation) | Qwen2.5-VL-7B · LiViBench Overall | accuracy: 58.3 -> 56.6 | Hurt |
| add VCR comment retrieval instead of raw comments (ablation) | LiVi-LLM-7B · LiViBench Overall | accuracy: 63.0 -> 64.4 | Improved |
| add VCR comment retrieval (ablation) | LLaVA-Video-7B · LiViBench Overall | accuracy: 52.6 -> 55.0 | Improved |
| add VCR comment retrieval (ablation) | Qwen2.5-VL-7B · LiViBench Overall | accuracy: 58.3 -> 57.7 | Hurt |
| add Stage I synthetic livestream instruction tuning (ablation) | Qwen2.5-Omni-7B · LiViBench Overall | accuracy: 60.3 -> 62.9 | Improved |
Closest papers by meaning
- MMOU: A Massive Multi-Task Omni Understanding and Reasoning Benchmark for Long and Complex Real-World Videos (NeurIPS 2026 · similarity 0.90)
- Needle In A Video Haystack: A Scalable Synthetic Evaluator for Video MLLMs (ICLR 2025 · similarity 0.90)
- LiveStar: Live Streaming Assistant for Real-World Online Video Understanding (NeurIPS 2025 · similarity 0.90)
- VideoAutoArena: An Automated Arena for Evaluating Large Multimodal Models in Video Analysis through User Simulation (CVPR 2025 · similarity 0.89)
- LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding (NeurIPS 2024 · similarity 0.89)
- OmniVideoBench: Towards Audio-Visual Understanding Evaluation for Omni MLLMs (ICLR 2026 · similarity 0.89)
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.
