Update README.md

#2
by kgubernatorov - opened
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -10,13 +10,13 @@ tags:
10
  base_model: lerobot/smolvla_base
11
  ---
12
 
13
- # HapticsVLA β€” Contact-Rich Manipulation without Inference-Time Tactile Sensing
14
 
15
  Distilled from a tactile-conditioned [SA-RWFM teacher](https://huggingface.co/armteam/crab-smolvla-rwfm) via offline action-level knowledge distillation. At inference, this model is an **unmodified SmolVLA** β€” no tactile sensors, no extra modules, zero overhead.
16
 
17
  ## Key Result
18
 
19
- HapticsVLA achieves **86.7% mean success rate** (vs. 61.7% baseline, 75.0% tactile teacher) with the **lowest force error rate of any model (5.0%)**, including the tactile-equipped teacher β€” all without requiring tactile sensors at inference.
20
 
21
  ## Model Details
22
 
@@ -42,7 +42,7 @@ HapticsVLA achieves **86.7% mean success rate** (vs. 61.7% baseline, 75.0% tacti
42
  |-------|-------------|-----------------|-----------------|
43
  | SmolVLA (Baseline) | 61.7% | 26.7% | No |
44
  | SA-RWFM (Teacher) | 75.0% | 11.7% | **Yes** |
45
- | **HapticsVLA (Ours)** | **86.7%** | **5.0%** | **No** |
46
 
47
  ## Distillation Approach
48
 
@@ -67,13 +67,13 @@ See [Advanced-Robotic-Manipulation/crab](https://github.com/Advanced-Robotic-Man
67
  - [crab-smolvla-rwfm](https://huggingface.co/armteam/crab-smolvla-rwfm) β€” SA-RWFM tactile teacher
68
 
69
 
70
- ## BibTex
71
 
72
  If you use this model, please cite our paper:
73
 
74
  ```bibtex
75
- @article{gubernatorov2026hapticsvla,
76
- title={HapticsVLA: Contact-Rich Manipulation via Vision-Language-Action Model without Inference-Time Tactile Sensing},
77
  author={Gubernatorov, Konstantin and Sannikov, Mikhail and Mikhalchuk, Ilya and Kuznetsov, Egor and Artemov, Makar and Ouwatobi, Ogunwoye Faith and Fernando, Marcelino and Asanov, Artem and Guo, Ziang and Tsetserukou, Dzmitry},
78
  journal={arXiv preprint arXiv:2603.15257},
79
  year={2026}
 
10
  base_model: lerobot/smolvla_base
11
  ---
12
 
13
+ # HapticVLA β€” Contact-Rich Manipulation without Inference-Time Tactile Sensing
14
 
15
  Distilled from a tactile-conditioned [SA-RWFM teacher](https://huggingface.co/armteam/crab-smolvla-rwfm) via offline action-level knowledge distillation. At inference, this model is an **unmodified SmolVLA** β€” no tactile sensors, no extra modules, zero overhead.
16
 
17
  ## Key Result
18
 
19
+ HapticVLA achieves **86.7% mean success rate** (vs. 61.7% baseline, 75.0% tactile teacher) with the **lowest force error rate of any model (5.0%)**, including the tactile-equipped teacher β€” all without requiring tactile sensors at inference.
20
 
21
  ## Model Details
22
 
 
42
  |-------|-------------|-----------------|-----------------|
43
  | SmolVLA (Baseline) | 61.7% | 26.7% | No |
44
  | SA-RWFM (Teacher) | 75.0% | 11.7% | **Yes** |
45
+ | **HapticVLA (Ours)** | **86.7%** | **5.0%** | **No** |
46
 
47
  ## Distillation Approach
48
 
 
67
  - [crab-smolvla-rwfm](https://huggingface.co/armteam/crab-smolvla-rwfm) β€” SA-RWFM tactile teacher
68
 
69
 
70
+ ## Citation
71
 
72
  If you use this model, please cite our paper:
73
 
74
  ```bibtex
75
+ @article{gubernatorov2026hapticvla,
76
+ title={HapticVLA: Contact-Rich Manipulation via Vision-Language-Action Model without Inference-Time Tactile Sensing},
77
  author={Gubernatorov, Konstantin and Sannikov, Mikhail and Mikhalchuk, Ilya and Kuznetsov, Egor and Artemov, Makar and Ouwatobi, Ogunwoye Faith and Fernando, Marcelino and Asanov, Artem and Guo, Ziang and Tsetserukou, Dzmitry},
78
  journal={arXiv preprint arXiv:2603.15257},
79
  year={2026}