Research map / Vision-language and multimodal / Robot action and navigation
DiTEA: Mixture-of-Experts for Vision-Language-Action Model in Robotic Manipulation
Chengxuan Li, Xingwan Wang
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 shared expert to base action head (ablation) | DiTEA · SIMPLER WidowX | success rate (%): 40.2 -> 42.4 | Improved |
| add Task-Instruction Gate (ablation) | DiTEA · SIMPLER WidowX | success rate (%): 40.2 -> 43.8 | Improved |
| add load balancing loss (ablation) | DiTEA · SIMPLER WidowX | success rate (%): 40.2 -> 41.1 | Improved |
| add TIG on top of shared expert (ablation) | DiTEA · SIMPLER WidowX | success rate (%): 42.4 -> 48.7 | Improved |
| add all three components (ablation) | DiTEA · SIMPLER WidowX | success rate (%): 40.2 -> 51.5 | Improved |
| scale action head parameters instead of MoE (ablation) | DiTEA · SIMPLER WidowX | success rate (%): 40.2 -> 45.4 | Improved |
| replace baseline with DiTEA vs CogACT on real-world tasks | DiTEA · Real-world Franka | success rate (%): 34.0 -> 40.2 | Improved |
| DiTEA vs OpenVLA on instruction following | DiTEA · instruction-following tasks (Figure 5) | success rate | Mixed |
Closest papers by meaning
- VLA-REPLICA: A Low-Cost, Reproducible Benchmark for Real-World Evaluation of Vision-Language-Action Models (NeurIPS 2026 · similarity 0.90)
- PriorVLA: Prior-Preserving Adaptation for Vision-Language-Action Models (NeurIPS 2026 · similarity 0.89)
- NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models (NeurIPS 2026 · similarity 0.89)
- ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models (NeurIPS 2026 · similarity 0.89)
- Exploring the Limits of Vision-Language-Action Manipulation in Cross-task Generalization (NeurIPS 2025 · similarity 0.89)
- CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models (CVPR 2025 · similarity 0.89)
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.
