Research map / Vision-language and multimodal / Robot action and navigation
StableVLA: Towards Robust Vision-Language-Action Models without Extra Data
YIYANG FU, Chubin Zhang, Shukai Gong, Yufan Deng, Kaiwei Sun, Qiyang Min, Qibin Hou, Yansong Tang, Jianan Wang, Daquan Zhou
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 MLP pathway to IB-Adapter (Fused IB-Adapter) (ablation) | StableVLA · LIBERO (corrupted avg) | success rate (%): 76.0 -> 79.1 | Improved |
| add MLP pathway to IB-Adapter (Fused IB-Adapter) (ablation) | StableVLA · CALVIN (corrupted avg) | avg completed tasks: 1.44 -> 2.13 | Improved |
| use sigmoid instead of softmax gating (ablation) | Fused IB-Adapter · LIBERO (corrupted avg) | success rate (%): 62.8 -> 79.1 | Improved |
| use sigmoid instead of softmax gating (ablation) | Fused IB-Adapter · CALVIN (corrupted avg) | avg completed tasks: 0.46 -> 2.13 | Improved |
| use sigmoid instead of softmax gating (ablation) | Fused IB-Adapter · CALVIN (clean) | avg completed tasks: 0.46 -> 4.17 | Improved |
| replace MLP projector with Fused IB-Adapter | VLA-Adapter · LIBERO-Spatial severity 5 | success rate (%): 58.5 -> 82.0 | Improved |
| replace MLP projector with Fused IB-Adapter | VLA-Adapter · LIBERO-Object severity 5 | success rate (%): 29.3 -> 70.2 | Improved |
| replace MLP projector with Fused IB-Adapter | VLA-Adapter · LIBERO-Long severity 5 | success rate (%): 26.2 -> 45.3 | Improved |
| replace MLP projector with Fused IB-Adapter | VLA-Adapter · LIBERO-Long (clean) | success rate (%): 94.4 -> 93.6 | No clear effect |
| replace MLP projector with Fused IB-Adapter | VLA-Adapter · CALVIN severity 5 | avg completed tasks: 1.44 -> 1.51 | Improved |
| replace MLP projector with Fused IB-Adapter | VLA-Adapter · Real robot Pick and place (avg change) | avg delta (pp): -49.2 -> -17.5 | Improved |
| replace MLP projector with Fused IB-Adapter | VLA-Adapter · Real robot Pack doll (avg change) | avg delta (pp): -31.7 -> -14.2 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- VLA-Adapter (0.5B) — VLA-Adapter: An Effective Paradigm for Tiny-Scale Vision-Language-Action Model (AAAI 2026) · better on 7 of 7 shared results
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
- VLM-RobustBench: A Comprehensive Benchmark for Robustness of Vision-Language Models (ICML 2026 · similarity 0.90)
- MVI-Bench: A Comprehensive Benchmark for Evaluating Robustness to Misleading Visual Inputs in LVLMs (ICML 2026 · similarity 0.90)
- EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models (NeurIPS 2025 · similarity 0.89)
- CogVLA: Cognition-Aligned Vision-Language-Action Models via Instruction-Driven Routing & Sparsification (NeurIPS 2025 · similarity 0.88)
- Machine Vision Therapy: Multimodal Large Language Models Can Enhance Visual Robustness via Denoising In-Context Learning (ICML 2024 · similarity 0.88)
- The Illusion of Progress? A Critical Look at Test-Time Adaptation for Vision-Language Models (NeurIPS 2025 · similarity 0.88)
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.
