Research map / Video / Video-language understanding
Describing Unseen Videos via Multi-Modal Cooperative Dialog Agents
Ye Zhu, Yu Wu, Yi Yang, Yan Yan
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 full-update of attention outputs instead of partial update (ablation) | QA-Cooperative Q-BOT · AVSD v0.1 | CIDEr: 21.6 -> 23.1 | Improved |
| add initial states from visual and audio-visual LSTM (ablation) | QA-Cooperative Q-BOT (full update) · AVSD v0.1 | CIDEr: 20.3 -> 23.1 | Improved |
| add initial states from LSTMs (ablation) | QA-Cooperative Q-BOT (partial update) · AVSD v0.1 | CIDEr: 20.3 -> 21.6 | No clear effect |
| add caption input (ablation) | A-BOT / Q-BOT QA-C · AVSD v0.1 | CIDEr: 19.5 -> 23.1 | Improved |
| add audio input (ablation) | QA-C · AVSD v0.1 | CIDEr: 22.3 -> 23.1 | Improved |
| add history embedding to answer decoder (ablation) | QA-C · AVSD v0.1 | CIDEr: 23.0 -> 23.1 | No clear effect |
| keep correct QA pair order instead of shuffled (ablation) | QA-C · AVSD v0.1 | CIDEr: 18.9 -> 23.1 | Improved |
| add cooperative learning (QA-C MM) over pre-trained Q-BOT | Q-BOT · AVSD v0.1 | CIDEr: 21.5 -> 23.1 | Improved |
| use MM instead of IM history attention | strong baseline Q-BOT · AVSD v0.1 | CIDEr: 21.3 -> 26.3 | Improved |
| add history attention to A-BOT | S. et al. A-BOT · AVSD v0.1 | CIDEr: 21.6 -> 25.6 | Improved |
| add cooperative learning versus basic baseline (MM) | Q-BOT · AVSD v0.1 | CIDEr: 20.2 -> 23.1 | Improved |
Closest papers by meaning
- Saying the Unseen: Video Descriptions via Dialog Agents (TPAMI 2021 · similarity 0.94)
- Learning Cooperative Visual Dialog Agents With Deep Reinforcement Learning (ICCV 2017 · similarity 0.88)
- V^2Dial: Unification of Video and Visual Dialog via Multimodal Experts (CVPR 2025 · similarity 0.85)
- Seeing, Listening, Remembering, and Reasoning: A Multimodal Agent with Long-Term Memory (ICLR 2026 · similarity 0.85)
- Structure-Aware Multimodal Sequential Learning for Visual Dialog (AAAI 2024 · similarity 0.84)
- Artemis: Towards Referential Understanding in Complex Videos (NeurIPS 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.
