Research map / Vision-language and multimodal / Visual question answering
Visual Dialogue State Tracking for Question Generation
Wei Pang, Xiaojie Wang
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 visual dialogue state tracking with UoOR and CMM (ablation) | VDST · GuessWhat?! New Object, 5Q | success rate (sampling): 62.73 -> 66.22 | Improved |
| add visual dialogue state tracking with UoOR and CMM (ablation) | VDST · GuessWhat?! New Object, 5Q | success rate (greedy): 64.30 -> 67.07 | Improved |
| add object-self difference attention (ablation) | VDST · GuessWhat?! New Object, 5Q | success rate (sampling): 64.06 -> 66.22 | Improved |
| add object-self difference attention (ablation) | VDST · GuessWhat?! New Object, 5Q | success rate (greedy): 65.21 -> 67.07 | Improved |
| add visual dialogue state tracking with UoOR and CMM (ablation) | VDST · GuessWhat?! New Game, 5Q | success rate (sampling): 60.07 -> 63.85 | Improved |
| add visual dialogue state tracking with UoOR and CMM (ablation) | VDST · GuessWhat?! New Game, 5Q | success rate (greedy): 60.93 -> 64.36 | Improved |
| add object-self difference attention (ablation) | VDST · GuessWhat?! New Game, 5Q | success rate (sampling): 60.89 -> 63.85 | Improved |
| add object-self difference attention (ablation) | VDST · GuessWhat?! New Game, 5Q | success rate (greedy): 61.71 -> 64.36 | Improved |
| track and update visual dialogue state | VDST · GuessWhat?! dialogues, 8Q | games with repeated questions (%): 52.19 -> 21.9 | Improved |
| track and update visual dialogue state | VDST · GuessWhat?! dialogues, 8Q | lexical diversity: 0.115 -> 0.150 | Improved |
Closest papers by meaning
- Guessing State Tracking for Visual Dialogue (ECCV 2020 · similarity 0.94)
- Learning Better Visual Dialog Agents With Pretrained Visual-Linguistic Representation (CVPR 2021 · similarity 0.90)
- GuessWhat?! Visual Object Discovery Through Multi-Modal Dialogue (CVPR 2017 · similarity 0.88)
- Divide-and-Conquer: Tree-structured Strategy with Answer Distribution Estimator for Goal-Oriented Visual Dialogue (AAAI 2025 · similarity 0.87)
- ConVQG: Contrastive Visual Question Generation with Multimodal Guidance (AAAI 2024 · similarity 0.85)
- Explicitly Guided Difficulty-Controllable Visual Question Generation (AAAI 2025 · similarity 0.85)
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.
