StreamBench
Streaming Video Understanding and Multi-round Interaction with Memory-enhanced Knowledge
Haomiao Xiong, Zongxin Yang, Jiazuo Yu, Yunzhi Zhuge, Lu Zhang, Jiawen Zhu, et al. · 2025-01-23
- Modality
- Video & streaming
- Data
- Hybrid (real + synthetic/simulated)
- Task · ground truth
- QA/Gen. · Ans.+Time
- Scope
- Benchmark with memory tracks
What it tests
- CECross-modal evidenceDirectly evaluated
- LELong-range evidenceDirectly evaluated
- SCState changesPartially evaluated
- CFConflicting memoriesNot evaluated
- UQUnanswerable queriesNot evaluated
- MAMemory-grounded actionNot evaluated
- TLTemporal localizationDirectly evaluated
- TRTemporal relationsPartially evaluated
- MEMulti-event integrationDirectly evaluated
Survey notes
Explicit short-term and long-term memory-search tasks introduce query delays; multi-round conversation tasks require earlier interaction records.