---
base_model: Qwen/Qwen3-VL-4B-Thinking
library_name: transformers
model_name: PhysicalAI-reason-VLA
tags:
- generated_from_trainer
- sft
- trl
- vision-language
- autonomous-driving
- reasoning
license: mit
datasets:
- mjf-su/PhysicalAI-reason-US
---

# PhysicalAI-reason-VLA

A vision-language driving policy fine-tuned from [mjf-su/PhysicalAI-base-VLA](https://huggingface.co/mjf-su/PhysicalAI-base-VLA) (itself based on [Qwen/Qwen3-VL-4B-Thinking](https://huggingface.co/Qwen/Qwen3-VL-4B-Thinking)) using supervised fine-tuning with [TRL](https://github.com/huggingface/trl).

This model extends the base waypoint-prediction VLA with **structured chain-of-thought reasoning** and **discrete driving decisions**, trained on 10k Gemini-annotated driving scenes for 2 epochs.

---

## Input / Output

**Inputs**
- A forward-facing camera image
- Past ego-vehicle waypoints in the vehicle's relative frame

**Output**

```
<think>
{
  "scene": "2–3 sentence static scene description",
  "move_justification": "2–3 sentence causal explanation linking scene to decisions",
}
</think>
<action>
<longitudinal_token><lateral_token>
</action>
<wp>[x.xx,y.yy,t.tttt]</wp>
<wp>[x.xx,y.yy,t.tttt]</wp>
...
```

The model produces three outputs in sequence: a reasoning trace (`<think>`), discrete longitudinal and lateral driving decisions (`<action>`), and future trajectory waypoints (`<wp>`).

---

## Decision Tokens

Each `<action>` block contains exactly one longitudinal and one lateral token.

**Longitudinal** — `<stop>` · `<yield>` · `<follow>` · `<gap_search>` · `<pass>` · `<adapt>` · `<cruise>`

**Lateral** — `<turn_left>` · `<turn_right>` · `<lc_left>` · `<lc_right>` · `<merge>` · `<nudge_out_left>` · `<nudge_out_right>` · `<nudge_in_left>` · `<nudge_in_right>` · `<pull_over>` · `<abort>` · `<lane_keep>`

These are registered as genuine single tokens in the vocabulary (not subword decompositions), enabling efficient probability measurement over the full decision space with a single forward pass.

---

## Training

| | |
|---|---|
| **Base model** | [mjf-su/PhysicalAI-base-VLA](https://huggingface.co/mjf-su/PhysicalAI-base-VLA) |
| **Dataset** | [mjf-su/PhysicalAI-reason-US](https://huggingface.co/datasets/mjf-su/PhysicalAI-reason-US) |
| **Annotation** | Gemini batch API (chain-of-thought labels on real US driving data) |
| **Samples** | 10,000 |
| **Epochs** | 2 |
| **Method** | Completion-only SFT via TRL |

---

## Quick Start

```python
from transformers import AutoProcessor, AutoModelForImageTextToText
from PIL import Image

model_id = "mjf-su/PhysicalAI-reason-VLA"
processor = AutoProcessor.from_pretrained(model_id)
model = AutoModelForImageTextToText.from_pretrained(model_id, device_map="auto")

image = Image.open("forward_camera.jpg")

past_waypoints = "<wp>[0.00,0.00,0.0000]</wp>\n<wp>[0.51,0.00,0.0001]</wp>\n..."

messages = [
    {
        "role": "system",
        "content": [{"type": "text", "text": "You are a helpful AI assistant ..."}]
    },
    {
        "role": "user",
        "content": [
            {"type": "image"},
            {"type": "text", "text": f"[PAST-VEHICLE-MOTION]:\n{past_waypoints}"}
        ]
    }
]

prompt = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = processor(text=[prompt], images=[image], return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=512)
print(processor.batch_decode(outputs, skip_special_tokens=True)[0])
```

---

## Citation

```bibtex
@misc{vonwerra2022trl,
  title        = {{TRL: Transformer Reinforcement Learning}},
  author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching
                  and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul
                  and Quentin Gallou{\'e}dec},
  year         = 2022,
  journal      = {GitHub repository},
  publisher    = {GitHub},
  howpublished = {\url{https://github.com/huggingface/trl}}
}
```