Research map / Vision-language and multimodal / Robot action and navigation
Unified Vision-Language-Action Model
Yuqi Wang, Xinghang Li, Wenxuan Wang, Junbo Zhang, Yingyan Li, Yuntao Chen, Xinlong Wang, Zhaoxiang Zhang
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 world model post-training on videos (ablation) | UniVLA · LIBERO | success rate: 48.5 -> 94.2 | Improved |
| add world model post-training on videos (ablation) | UniVLA · CALVIN | average length: 1.46 -> 4.61 | Improved |
| add world model post-training on videos (ablation) | UniVLA · SimplerEnv-WidowX | success rate: 0.0 -> 64.6 | Improved |
| add world model post-training on videos (ablation) | UniVLA · LIBERO-Long | success rate: 17.4 -> 89.2 | Improved |
| post-train with action prediction only (ablation) | UniVLA · LIBERO | success rate: 48.5 -> 43.9 | Hurt |
| post-train with video prediction instead of world model (no text) (ablation) | UniVLA · LIBERO | success rate: 94.2 -> 78.9 | Hurt |
| post-train with text-to-image instead of world model (ablation) | UniVLA · CALVIN | average length: 4.61 -> 3.79 | Hurt |
| add post-training with 10% fine-tuning data (ablation) | UniVLA · CALVIN | average length: 0.15 -> 3.19 | Improved |
| add post-training for faster convergence at 2k iterations (ablation) | UniVLA · CALVIN | average length: 0.37 -> 4.21 | Improved |
| add visual prediction loss without post-training (ablation) | UniVLA · LIBERO | success rate: 48.5 -> 88.7 | Improved |
| add history window of 10 with 1+1 observations (ablation) | UniVLA · CALVIN | average length: 4.26 -> 4.61 | Improved |
| extend history from 1+1 to 1+2 observations (ablation) | UniVLA · CALVIN | average length: 4.61 -> 4.43 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- CoT-VLA — CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models (CVPR 2025) · better on 2 of 2 shared results
- UP-VLA — UP-VLA: A Unified Understanding and Prediction Model for Embodied Agent (ICML 2025) · better on 2 of 2 shared results
- Seer-Large — SEER: Backdoor Detection for Vision-Language Models through Searching Target Text and Image Trigger Jointly (AAAI 2024) · better on 1 of 1 shared result
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
- DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge (NeurIPS 2025 · similarity 0.92)
- NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models (NeurIPS 2026 · similarity 0.92)
- CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models (CVPR 2025 · similarity 0.92)
- ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models (NeurIPS 2026 · similarity 0.92)
- VLA-REPLICA: A Low-Cost, Reproducible Benchmark for Real-World Evaluation of Vision-Language-Action Models (NeurIPS 2026 · similarity 0.92)
- Exploring the Limits of Vision-Language-Action Manipulation in Cross-task Generalization (NeurIPS 2025 · similarity 0.91)
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.
