Add ToolWeave Stage 3 model card and branding

README.md

@@ -1,126 +1,39 @@
----
-base_model:
-  - Qwen/Qwen3-4B
-language:
-  - en
-license: apache-2.0
-tags:
-  - agent
-  - Agentic Learning
-  - tool use
-  - function-calling
-  - multi-turn
-  - reinforcement-learning
-  - GRPO
-  - BFCL
-task_categories:
-  - question-answering
-  - text-generation
-pipeline_tag: text-generation
-library_name: transformers
-datasets:
-  - gorilla-llm/Berkeley-Function-Calling-Leaderboard
-model-index:
-  - name: Qwen3-4B-RODS
-    results:
-      - task:
-          type: function-calling
-          name: Multi-Turn Tool Use
-        dataset:
-          name: BFCL V3 Multi-Turn
-          type: gorilla-llm/Berkeley-Function-Calling-Leaderboard
-        metrics:
-          - type: accuracy
-            value: 56.00
-            name: Overall Accuracy
----
+<div align="center">
 
-# RODS: Reward-Driven Online Data Synthesis for Multi-Turn Tool-Use Agents
+<img src="assets/toolweave-mark.svg" alt="ToolWeave mark" width="110">
 
-[![arXiv](https://img.shields.io/badge/arXiv-2606.19047-b31b1b.svg?logo=arXiv)](https://arxiv.org/abs/2606.19047) [![Paper](https://img.shields.io/badge/Hugging%20Face-Paper-yellow?logo=huggingface)](https://huggingface.co/papers/2606.19047) [![Model](https://img.shields.io/badge/Hugging%20Face-Model-yellow?logo=huggingface)](https://huggingface.co/RuishanFang/Qwen3-4B-RODS) [![GitHub](https://img.shields.io/badge/GitHub-Code-181717?logo=github)](https://github.com/inclusionAI/AWorld-RL/tree/main/RODS) [![Project Page](https://img.shields.io/badge/Project-AWorld-green)](https://github.com/inclusionAI/AWorld)
+# ToolWeave Stage 3
 
-## Model Overview
+**Reference checkpoint for the boundary-guided online-learning stage.**
 
-The **Qwen3-4B-RODS** model is a high-performance **Large Language Model (LLM)** fine-tuned for complex, multi-turn **Function Calling (FC)** and agentic tool-use tasks. Built upon the **Qwen3-4B-Instruct** base model, it has been trained using the novel **RODS (Reward-driven Online Data Synthesis)** framework combined with GRPO reinforcement learning.
+</div>
 
-RODS closes the loop between RL training and data generation: it repurposes the progress reward variance as a zero-cost capability boundary detector, continuously synthesizes structurally isomorphic training data at the agent's learning frontier, and manages a dynamic replay buffer that co-evolves with the policy. Starting from only **400 human-annotated seeds**, RODS achieves strong multi-turn tool-use performance with extreme data efficiency.
+## Model summary
 
-- **Base Model:** [Qwen3-4B-Instruct](https://huggingface.co/Qwen/Qwen3-4B)
-- **Size:** 4 Billion parameters
-- **Key Capability:** Advanced Multi-Turn Function Calling and Agentic Tool-Use
+This repository stores the Qwen3-4B-RODS reference checkpoint used in the ToolWeave Stage 3 workspace. It is retained here as a Stage 3 reference model while the formal ToolWeave Stage 3 training remains pending.
 
-## Evaluation Results
+- Reference model: Qwen3-4B-RODS
+- Intended role: Stage 3 reference / starting model
+- ToolWeave Stage 3 status: formal training pending
+- Release status: reference checkpoint only
 
-The model was evaluated on the Berkeley Function-Calling Leaderboard (BFCL).
+This repository must not be interpreted as a completed ToolWeave Stage 3 release or as the official RODS implementation. ToolWeave is a separate project that adapts public ideas and infrastructure with project-specific extensions.
 
-### BFCLv3 Multi-Turn Performance
-
-| Model | Size | Multi-Turn (Overall) | Base | Miss Func | Miss Param | Long Context |
-| :--- | :---: | :---: | :---: | :---: | :---: | :---: |
-| Qwen3-4B-Instruct (Base) | 4B | 22.13 | 26.50 | 21.00 | 15.50 | 25.50 |
-| **Qwen3-4B + RODS (ours)** | **4B** | **56.00** | **68.00** | **59.00** | **44.00** | **53.00** |
-| Claude-Sonnet-4-5-20250929 | - | 61.38 | 69.00 | 65.00 | 52.50 | 59.00 |
-| Grok-4-1-fast-reasoning | - | 58.88 | 70.50 | 59.50 | 43.00 | 62.50 |
-| Kimi-K2-Instruct | 1043B | 50.63 | 62.00 | 41.00 | 44.50 | 55.00 |
-| Qwen3-32B | 32B | 47.88 | 56.00 | 52.50 | 40.00 | 43.00 |
-| DeepSeek-V3.2-Exp | 671B | 44.88 | 55.00 | 49.00 | 27.00 | 48.50 |
-| GPT-4o-2024-11-20 | - | 42.50 | 55.50 | 34.50 | 29.00 | 51.00 |
-
------
-
-## Training Data and Framework
-
-### RODS Framework
-
-RODS is a closed-loop RL-data synthesis framework with three co-evolving modules:
-
-1. **Reward-Based Boundary Detection:** Uses GRPO rollout reward variance as a zero-cost probe to identify tasks at the agent's capability boundary, where gradient signal is richest.
-2. **Skill-Aligned Synthesis Pipeline:** A multi-agent pipeline (Planner → Executor → Rewriter → Critic) generates structurally isomorphic variants that preserve API topology and dependency depth while introducing novel narratives and environment states.
-3. **Dynamic Replay Buffer Management:** A dual-control lifecycle with staged injection and multi-layer retirement keeps the training pool anchored at the shifting capability boundary.
-
-### Training Details
-
-- **Method:** GRPO (Group Relative Policy Optimization)
-- **Rollouts:** K=16 per prompt
-- **Training stages:**
-  1. Format training (100 Base samples, format reward)
-  2. Base reasoning (100 Base samples, progress reward)
-  3. Full expansion (400 samples + dynamic synthesis, progress reward)
-- **Synthesis backbone:** Qwen3-32B via vLLM
-- **Hardware:** 8x A100 (training) + 8x A100 (synthesis)
-- **Active training pool:** ~800 samples (400 seeds + up to 400 generated)
-
-### Usage
+## Usage
 
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-model_name = "RuishanFang/Qwen3-4B-RODS"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto")
+model_id = "muradil211/ToolWeave_stage3"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
 ```
 
-For tool-use inference, follow the [Qwen3 function calling format](https://qwen.readthedocs.io/en/latest/framework/function_call.html). The model expects tools to be provided in the system prompt and generates structured `<tool_call>` responses.
-
------
-
-## Related Projects and Citation
-
-This work is part of the open-source project **[AWorld, InclusionAI](https://github.com/inclusionAI/AWorld/)**.
-
-If you use RODS in your research, please cite:
-
-```bibtex
-@article{fang2026rods,
-  title={RODS: Reward-Driven Online Data Synthesis for Multi-Turn Tool-Use Agents},
-  author={Fang, Ruishan and Lu, Siyuan and Zhuang, Chenyi and Lin, Tao},
-  journal={arXiv preprint arXiv:2606.19047},
-  year={2026}
-}
-```
+Tool-use inference requires the model's function schemas and the Qwen3-compatible tool-call format.
 
-### Contact
+## References
 
-For inquiries, please contact:
+- [ToolWeave project](https://github.com/Muradil-mamat-211/ToolWeave)
+- [Upstream RODS model](https://huggingface.co/RuishanFang/Qwen3-4B-RODS)
+- [RODS paper](https://arxiv.org/abs/2606.19047)
 
-- `fangruishan@westlake.edu.cn`