AI Paper Atlas

Research map / Video / Video-language understanding

ViSpeak: Visual Instruction Feedback in Streaming Videos

Shenghao Fu, Qize Yang, Yuan-Ming Li, Yi-Xing Peng, Kun-Yu Lin, Xihan Wei, Jian-Fang Hu, Xiaohua Xie, Wei-Shi Zheng

ICCV 2025 · Video › Video-language understanding · approach: Event and anomaly reasoning in video · 22 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 a trained informative head instead of the language-model head for proactive output (ablation)ViSpeak · StreamingBench POPO score: 30.00 -> 34.80Improved
predict from the last visual token instead of the <seg> token (ablation)ViSpeak informative head · StreamingBench POPO score: 34.80 -> 36.00Improved
jointly train the informative head with the LLM (ablation)ViSpeak · StreamingBench POPO score: 36.00 -> 38.80Improved
jointly train the informative head with the LLM (ablation)ViSpeak · StreamingBench overallOverall: 60.91 -> 62.00Improved
combine the two input streams with different methods (Adaptive Sum, Linear, Add) (ablation)ViSpeak template alignment · MME, MVBench, Video-MMEMVBench: 54.12 / 52.95 / 54.27No clear effect
template alignment to two-stream format versus original VITA 1.5 (ablation)VITA 1.5 · MMEMME total: 2353.5 -> 2237.0Hurt
add offline video data (SMILE, IntentQA, Social-IQ) to ViSpeak-Instruct (ablation)ViSpeak · ViSpeak-BenchOverall: 2.70 -> 2.76Improved
stage 3 finetuning on ViSpeak-Instruct after stage 2ViSpeak · StreamingBench POPO score: 38.80 -> 50.80Improved
stage 3 finetuning on ViSpeak-InstructViSpeak · StreamingBench real-time visual understandingAll score: 74.36 -> 70.44Hurt
give models explicit prompts about response type on ViSpeak-Benchbaseline MLLMs · ViSpeak-BenchstabilityImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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