---
license: apache-2.0
library_name: pytorch
tags:
- robotics
- manipulation
- baselines
pipeline_tag: robotics
---

# HapticWAM baselines

> **HapticWAM: Distilling Imagined Touch into a World-Action Model without Inference-Time Tactile Sensing**
> — paper [arXiv:2609.23888](https://arxiv.org/abs/2609.23888), submitted to **ICRA 2027**.
> Code: [github.com/Advanced-Robotic-Manipulation/HapticWAM](https://github.com/Advanced-Robotic-Manipulation/HapticWAM) · all repos: [HapticWAM — ICRA 2027](https://huggingface.co/collections/armteam/hapticwam-icra-2027-6ab234bdfe24c383b5f9fb57).

The three baseline policies the HapticWAM student is compared against, at the exact steps the
paper deploys, plus the evaluation sweeps that chose those steps.

| Folder | Files | Size | Baseline | Deployed checkpoint |
|---|---|---|---|---|
| `diffusion_100k/` | 7 | 1.05 GB | Diffusion Policy | `pretrained_model/` at 100k steps |
| `xvla_20k/` | 7 | 1.76 GB | X-VLA 0.9B | `pretrained_model/` at 20k steps |
| `pi05_phantom_expert_v1/` | 18 | 9.69 GB | pi0.5 expert fine-tune | **`020000/pretrained_model/`** |
| `eval_baselines/` | 10 | 1.2 MB | evaluation JSONs, see below |
| `eval_pi05/` | 11 | 1.2 MB | pi0.5 step-selection sweep |

`index.jsonl` lists every file here with its size, LFS sha256 and the path it was copied from.

## What the eval files cover

`eval_baselines/` holds the diffusion-policy sweep (`eval_dp_025000` … `eval_dp_100000`) and the
X-VLA sweep (`eval_xvla_005000` … `eval_xvla_020000`), which is how 100k and 20k were chosen.
It also holds `A_visiononly_nfe1.json` and `B_nodistill_nfe1.json`. Those two score the
vision-only and no-distillation Cosmos arms, whose checkpoints live in
`armteam/hapticwam-ablations` rather than here; they are kept together because they are one
baseline table in the paper.

`eval_pi05/` holds the step sweep for the pi0.5 expert. The **deployed** pi0.5 arm is step
**20000** (`pi05_020000_terminal.json`). The `eval_resume_*` files score a run that was resumed
to 60k steps; that checkpoint is not deployed and lives in `armteam/hapticwam-ablations` as
`pi05_phantom_expert_v1_resume60k/`.

`pi05_phantom_expert_v1/` also carries `020000/training_state/` (optimizer and scheduler state,
2.2 GB) and `paligemma_tokenizer/`, both needed to resume or to run the policy.

## Data

All three baselines were trained on
[`armteam/hapticwam-teleop-dataset`](https://huggingface.co/datasets/armteam/hapticwam-teleop-dataset).
Models: [`armteam/hapticwam-teacher`](https://huggingface.co/armteam/hapticwam-teacher),
[`armteam/hapticwam-student`](https://huggingface.co/armteam/hapticwam-student).

## Part of the HapticWAM release

Ten repos on the hub, gathered in the [HapticWAM — ICRA 2027 collection](https://huggingface.co/collections/armteam/hapticwam-icra-2027-6ab234bdfe24c383b5f9fb57).

| Repo | Kind | Holds |
|---|---|---|
| [`armteam/hapticwam-teacher`](https://huggingface.co/armteam/hapticwam-teacher) | model | the tactile-input **teacher**. Deployed checkpoint `teacher_v6_simft/teacher_002000.pt`; also holds the Cosmos prompt cache `text_embeddings.pt` |
| [`armteam/hapticwam-student`](https://huggingface.co/armteam/hapticwam-student) | model | the distilled **pad-free student**, the model that runs on the rig. Deployed checkpoint `hid_simft/student_001000.pt` |
| `armteam/hapticwam-baselines` **← you are here** | model | the pi0.5, Diffusion Policy and X-VLA baselines at the deployed steps |
| [`armteam/hapticwam-ablations`](https://huggingface.co/armteam/hapticwam-ablations) | model | every training arm that is not deployed, and the complete evaluation sweeps |
| [`armteam/hapticwam-teleop-dataset`](https://huggingface.co/datasets/armteam/hapticwam-teleop-dataset) | dataset | the training corpus — 1,115 teleoperated episodes, packed per task |
| [`armteam/hapticwam-teleop-raw`](https://huggingface.co/datasets/armteam/hapticwam-teleop-raw) | dataset | the same teleoperation as loose, as-recorded sessions (provenance) |
| [`armteam/hapticwam-sim-episodes`](https://huggingface.co/datasets/armteam/hapticwam-sim-episodes) | dataset | Isaac Sim expert episodes, used for the sim fine-tune |
| [`armteam/hapticwam-rig-episodes`](https://huggingface.co/datasets/armteam/hapticwam-rig-episodes) | dataset | the closed-loop rig takes the reported numbers are computed from |
| [`armteam/hapticwam-rollouts`](https://huggingface.co/datasets/armteam/hapticwam-rollouts) | dataset | policy-driven rollouts — the DAgger rounds and the deploy days |
| [`armteam/hapticwam-evidence`](https://huggingface.co/datasets/armteam/hapticwam-evidence) | dataset | per-take evidence behind the paper's tables — scored CSVs, probe JSONs, figures |

Code, training and deployment scripts: [github.com/Advanced-Robotic-Manipulation/HapticWAM](https://github.com/Advanced-Robotic-Manipulation/HapticWAM).

## Licence

Model weights and the HapticWAM code: **Apache-2.0**. The datasets listed above: **CC-BY-4.0**.