Research map / Video / Video-language understanding
SciEducator: Scientific Video Understanding and Educating via Deming-Cycle Multi-Agent System
Zhiyu Xu, Weilong Yan, Yufei Shi, Xin Meng, Tao He, Huiping Zhuang, Ming Li, Hehe Fan
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 Knew (new knowledge) and F (failure analysis) in study stage (ablation) | SciEducator · SciVBench Physics | Accuracy: 45.94 -> 65.31 | Improved |
| add failure analysis F (given Knew) (ablation) | SciEducator · SciVBench Daily Life | Accuracy: 45.83 -> 62.24 | Improved |
| add new knowledge Knew (given F) (ablation) | SciEducator · SciVBench Daily Life | Accuracy: 34.64 -> 62.24 | Improved |
| add empirical prior E to evaluator (ablation) | Evaluator Agent · SciVBench (500 QA) | Accuracy: 57.50 -> 64.00 | Improved |
| add IDF keyword weighting to evaluator (ablation) | Evaluator Agent · SciVBench (500 QA) | Accuracy: 59.90 -> 64.00 | Improved |
| add perceptual evaluation to evaluator (ablation) | Evaluator Agent · SciVBench (500 QA) | Accuracy: 54.50 -> 64.00 | Improved |
| increase maximum PDSA rounds from 1 to 5 (ablation) | SciEducator · SciVBench Education Subset | Relevance win rate (%): 2.50 -> 90.00 | Improved |
| increase maximum PDSA rounds 1 to 5 (understanding) (ablation) | SciEducator · SciVBench Physics/Chemistry/Daily Life | Relevance and Accuracy | Improved |
| use SciEducator multi-agent system instead of GPT-4o | GPT-4o · SciVBench Physics | Accuracy: 34.69 -> 65.31 | Improved |
| use SciEducator instead of best baseline for education | Gemini 2.0 flash · SciVBench Education Subset | Attractiveness win rate (%): 0.00 -> 97.50 | Improved |
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.
- VideoAgent — VideoAgent: A Memory-augmented Multimodal Agent for Video Understanding (ECCV 2024) · better on 4 of 4 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
- SciVideoBench: Benchmarking Scientific Video Reasoning in Large Multimodal Models (ICML 2026 · similarity 0.88)
- VideoWebArena: Evaluating Long Context Multimodal Agents with Video Understanding Web Tasks (ICLR 2025 · similarity 0.86)
- Video-STaR: Self-Training Enables Video Instruction Tuning with Any Supervision (ICLR 2025 · similarity 0.85)
- MicroVQA: A Multimodal Reasoning Benchmark for Microscopy-Based Scientific Research (CVPR 2025 · similarity 0.84)
- ReAgent-V: A Reward-Driven Multi-Agent Framework for Video Understanding (NeurIPS 2025 · similarity 0.84)
- DoraemonGPT: Toward Understanding Dynamic Scenes with Large Language Models (Exemplified as A Video Agent) (ICML 2024 · similarity 0.84)
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.
