Upload README.md with huggingface_hub

README.md

@@ -21,7 +21,7 @@ The **SFT checkpoint** of **MechVL** — the domain-specialized multimodal model
 > **MechVQA: Benchmarking and Enhancing Multimodal LLMs on Comprehensive Mechanical Drawing Understanding** (ICML 2026)
 
 [![arXiv](https://img.shields.io/badge/arXiv-2605.30794-b31b1b.svg)](https://arxiv.org/abs/2605.30794)
-[![GitHub](https://img.shields.io/badge/Code-GitHub-181717.svg)](https://github.com/xiaofengShi/MechVQA)
+[![Code](https://img.shields.io/badge/Code-GitHub-181717.svg)](https://github.com/xiaofengShi/MechVQA)
 
 ## Model description
 
@@ -33,21 +33,21 @@ MechVL-4B-SFT is initialized from `Qwen3-VL-4B-Instruct` and trained with **full
 | Architecture | Qwen3VLForConditionalGeneration |
 | Stage | 1 / 2 — SFT (→ RL) |
 | MechVQA Total | **76.36** |
-| RL checkpoint | [xiaofengalg/MechVL-4B-RL](https://modelscope.cn/models/xiaofengalg/MechVL-4B-RL) |
+| RL checkpoint | [MonteXiaofeng/MechVL-4B-RL](https://huggingface.co/MonteXiaofeng/MechVL-4B-RL) |
 
-## Usage (ModelScope)
+## Usage (transformers)
 
 ```python
 import torch
-from modelscope import AutoModelForCausalLM, AutoProcessor
+from transformers import AutoProcessor, AutoModelForImageTextToText
 
-model = AutoModelForCausalLM.from_pretrained(
-    "xiaofengalg/MechVL-4B-SFT", torch_dtype=torch.bfloat16, device_map="auto"
+model = AutoModelForImageTextToText.from_pretrained(
+    "MonteXiaofeng/MechVL-4B-SFT", dtype=torch.bfloat16, device_map="auto"
 )
-processor = AutoProcessor.from_pretrained("xiaofengalg/MechVL-4B-SFT")
+processor = AutoProcessor.from_pretrained("MonteXiaofeng/MechVL-4B-SFT")
 
 messages = [{"role": "user", "content": [
-    {"type": "image", "image": "path/to/drawing.png"},
+    {"type": "image", "url": "path/to/drawing.png"},
     {"type": "text", "text": "图纸中标注的零件总长度是多少？"},
 ]}]
 inputs = processor.apply_chat_template(
@@ -58,11 +58,11 @@ out = model.generate(**inputs, max_new_tokens=1024)
 print(processor.decode(out[0], skip_special_tokens=True))
 ```
 
-Also available on [HuggingFace](https://huggingface.co/MonteXiaofeng/MechVL-4B-SFT). For batch vLLM inference, see [`scripts/batch_infer.py`](https://github.com/xiaofengShi/MechVQA/blob/main/scripts/batch_infer.py).
+For **batch vLLM inference** (SFT/RL dual-mode), see [`scripts/batch_infer.py`](https://github.com/xiaofengShi/MechVQA/blob/main/scripts/batch_infer.py).
 
 ## Training
 
-Full-parameter SFT on the LLM module (vision tower frozen) over the MechVQA training split. See [§4.1 of the paper](https://arxiv.org/abs/2605.30794).
+Full-parameter SFT on the LLM module (vision tower frozen) over the MechVQA training split, with a unified response schema (rationale + concise final answer). See [§4.1 of the paper](https://arxiv.org/abs/2605.30794).
 
 ## Citation