Research map / Vision-language and multimodal / Robot action and navigation
Neural Implicit Action Fields: From Discrete Waypoints to Continuous Functions for Vision-Language-Action Models
Haoyun Liu, Jianzhuang Zhao, Xinyuan Chang, Tianle Shi, Chuanzhang Meng, Jiayuan Tan, Feng Xiong, Tong Lin, Dongjie Huo, Mu Xu, SongLin Dong, Zhiheng Ma and 2 more
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 |
|---|---|---|---|
| represent actions as a continuous SIREN function instead of discrete B-spline/waypoint tokens | Florence-2 Large VLA · CALVIN ABC→D avg length | Avg. Len: 4.42 -> 4.47 | Improved |
| represent actions as a continuous SIREN function instead of B-spline control points | Florence-2 Large VLA · LIBERO average | success rate (%): 92.5 -> 97.9 | Improved |
| use NIAF head instead of OFT head on Qwen3-VL | Qwen3-VL · LIBERO multi-task | average success rate (%): 96.6 -> 97.7 | Improved |
| NIAF versus continuous-output baselines BEAST-CT and OFT (ablation) | Florence-2 Large · CALVIN ABC→D and LIBERO-Long | success rate | Improved |
| replace flow-matching action expert with NIAF head on pi0.5 | pi0.5 · Real-world Shape Insertion | success rate | Improved |
| replace pi0.5 action head with NIAF | pi0.5 · Real-world Towel Folding | success rate | Hurt |
| replace flow-matching target with B-spline control points | pi0.5 · Real-world Shape Insertion | success rate | Hurt |
| add explicit velocity supervision with analytic derivative enabling impedance control | NIAF · Real-world Item Placement | velocity profile smoothness (qualitative) | Improved |
Closest papers by meaning
- Beyond World-Frame Action Heads: Motion-Centric Action Frames for Vision-Language-Action Models (NeurIPS 2026 · similarity 0.90)
- ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models (NeurIPS 2026 · similarity 0.90)
- World–Value–Action Model: Implicit Planning for Vision–Language–Action Systems (NeurIPS 2026 · similarity 0.89)
- NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models (NeurIPS 2026 · similarity 0.89)
- ATI-VLA: Action-Centric Predictive Vision–Language–Action Models via Actionable Alignment Then Adaptive Injection (NeurIPS 2026 · similarity 0.88)
- FineVLA: Fine-Grained Instruction Alignment for Steerable Vision-Language-Action Policies (NeurIPS 2026 · 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.
