Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Learning Cooperative Visual Dialog Agents With Deep Reinforcement Learning
Abhishek Das, Satwik Kottur, Jose M. F. Moura, Stefan Lee, Dhruv Batra
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 |
|---|---|---|---|
| RL fine-tuning of both agents on image guessing reward | Q-bot and A-bot (SL-pretrained) · VisDial v0.5 test image guessing, round 10 | percentile rank: improving percentile rank by over 3% | Improved |
| RL fine-tuning, human judges guess the image | SL-pretrained dialog · Human study | mean rank of ground truth: 3.70 -> 2.73 | Improved |
| RL fine-tuning, human judges guess the image | SL-pretrained dialog · Human study | MRR: 0.518 -> 0.622 | Improved |
| add multi-task SL answer loss to RL image-guessing reward (Frozen-Q-multi) | Frozen-Q A-bot · VisDial v0.5 test | mean rank: improving the best reported result by 0.7 mean rank (3%) | Improved |
| RL fine-tuning only (no SL term) | A-bot · VisDial v0.5 test | MRR, recall@k, mean rank | No clear effect |
| freeze one agent during RL (Frozen-Q or Frozen-A) (ablation) | RL agents · Image guessing | percentile rank | Hurt |
Closest papers by meaning
- Visual Dialog (CVPR 2017 · similarity 0.91)
- Learning Better Visual Dialog Agents With Pretrained Visual-Linguistic Representation (CVPR 2021 · similarity 0.90)
- VisPlay: Self-Evolving Vision-Language Models (CVPR 2026 · similarity 0.88)
- MMDU: A Multi-Turn Multi-Image Dialog Understanding Benchmark and Instruction-Tuning Dataset for LVLMs (NeurIPS 2024 · similarity 0.87)
- DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning (ICLR 2026 · similarity 0.87)
- Divide-and-Conquer: Tree-structured Strategy with Answer Distribution Estimator for Goal-Oriented Visual Dialogue (AAAI 2025 · similarity 0.87)
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.
