Update README.md

README.md

@@ -1,3 +1,117 @@
 ---
 license: mit
+task_categories:
+- visual-question-answering
+- image-to-text
+language:
+- en
+tags:
+- physical-tool-use
+- mllm-evaluation
+- tool-planning
+- multimodal
+pretty_name: PhysTool-Bench
+size_categories:
+- 1K<n<10K
 ---
+
+# Dataset Card for PhysTool-Bench
+
+## 📊 Dataset Summary
+
+**PhysTool-Bench** is a multimodal benchmark designed to evaluate how well Multimodal Large Language Models (MLLMs) perceive, select, and sequence physical tools in real-world scenes. Unlike traditional tool-use benchmarks that focus on digital APIs, this dataset probes an MLLM's ability to ground functional reasoning in cluttered, physical environments. 
+
+The benchmark features 2,510 high-quality queries covering 2,678 unique physical tools across 57 distinct categories (e.g., manufacturing, healthcare, farming).
+
+### Key Features
+- **Two‑Task Evaluation:** Decouples pure visual recognition from functional planning and sequencing.
+- **Real‑World Clutter:** Each scene contains an average of 8.6 tools (3.1 required targets and 5.5 visually/functionally similar distractors).
+- **Sequential Logic:** 86.9% of the tasks require a strict execution order, rigorously testing the model's physical commonsense.
+
+---
+
+## 🎯 Supported Tasks
+
+The dataset separates evaluation into two distinct tracks to pinpoint whether model failures stem from visual bottlenecks or poor physical reasoning.
+
+### Task I: Tool Recognition
+* **Input:** A real-world scenario image.
+* **Objective:** Enumerate all visible tools in the cluttered scene.
+* **Purpose:** Measures pure visual enumeration and recognition capabilities.
+
+### Task II: Tool Selection & Planning
+* **Input:** A real-world scenario image paired with a brief task instruction.
+* **Objective:** Output the exact, ordered sequence of tools required to complete the specified task.
+* **Purpose:** Measures functional mapping, physical commonsense, and multi-step planning capabilities.
+
+---
+
+## 📁 Dataset Structure
+
+Unlike standard text-to-text datasets, **PhysTool-Bench** relies on a decoupled structure to support complex visual reasoning evaluations. The repository contains the raw images and two primary metadata files:
+
+* `images/`: Directory containing all high-resolution physical scenario images.
+* `generation_checkpoint.json`: The input file used for model inference. It contains the image paths and `task_instruct` prompts for Task II.
+* `corrected_tools.json`: The ground truth file used for evaluation. It contains the refined taxonomy, required tools (`target_tools`), `target_steps` for ordered tasks, and `negative_tools` (distractors).
+
+### Example: Loading the Raw Data
+You can easily download and explore the raw dataset using the `huggingface_hub` or standard Python tools:
+
+```python
+import json
+import os
+from huggingface_hub import snapshot_download
+from PIL import Image
+
+# 1. Download the dataset folder
+dataset_path = snapshot_download(repo_id="ModalityDance/PhysTool-Bench", repo_type="dataset")
+
+# 2. Load the input metadata
+with open(os.path.join(dataset_path, "generation_checkpoint.json"), "r") as f:
+    inputs = json.load(f)
+
+# 3. Explore a sample
+sample = inputs[0]
+print(f"Task Instruction: {sample['task_instruct']}")
+
+# Load corresponding image
+img_path = os.path.join(dataset_path, sample['image_path'])
+Image.open(img_path).show()
+```
+
+---
+
+## ⚠️ Inference & Evaluation (Important)
+
+Due to the complex nature of physical tool planning, **standard HuggingFace pipelines (`pipeline("visual-question-answering")`) are not sufficient for evaluating this benchmark.** To properly run PhysTool-Bench, please use our **[Official GitHub Repository](https://github.com/ModalityDance/PhysTool-Bench)**.
+
+### Why use the official codebase?
+
+- **Environment Isolation:** Different MLLMs require conflicting dependency versions (e.g., PyTorch, Transformers, Accelerate). Our repo provides standalone inference scripts for major models.
+- **LLM-as-a-Judge Evaluation:** We do not use simple string matching. Because models often output synonyms or functionally identical tools, our evaluation pipeline (`eval_gemini.py`) utilizes the Gemini API (`gemini-3.1-pro-preview`) to perform semantic one-to-one mapping against the ground truth before calculating metrics like SR@k and Pairwise Order Accuracy (POA).
+
+**Head over to [ModalityDance/PhysTool-Bench](https://github.com/ModalityDance/PhysTool-Bench) for the complete quickstart guide, environment setups, and automated evaluation scripts.**
+
+
+---
+
+## 📚 Citation
+
+If you use **PhysTool-Bench** in your research, please cite our paper:
+
+```bibtex
+@article{PhysTool-Bench2026,
+  title        = {Beyond APIs: Probing the Limits of MLLMs in Physical Tool Use},
+  author       = {Zhixin Ma and Yutong Zhou and Yongqi Li},
+  journal      = {arXiv preprint arXiv:{xxxx.xxxxx}},
+  year         = {2026}
+}
+
+```
+
+---
+
+
+## 📜 License
+
+The dataset (images and annotations) is released under the **MIT** license for non-commercial research use.