Update README.md
Browse files
README.md
CHANGED
|
@@ -3,6 +3,9 @@ license: mit
|
|
| 3 |
base_model:
|
| 4 |
- Qwen/QwQ-32B
|
| 5 |
---
|
|
|
|
|
|
|
|
|
|
| 6 |
- Native agentic search reasoning model using ReAct framework towards autonomous information seeking agency and Deep Research-like model.
|
| 7 |
- We introduce a four-stage training paradigm comprising browsing data construction, trajectory sampling, supervised fine-tuning for effective cold start, and reinforcement learning for improved generalization, enabling the agent to autonomously acquire autonomous search and reasoning skills.
|
| 8 |
- Our data-centric approach integrates trajectory-level supervision fine-tuning and reinforcement learning (DAPO) to develop a scalable pipeline for training agentic systems via SFT or RL.
|
|
|
|
| 3 |
base_model:
|
| 4 |
- Qwen/QwQ-32B
|
| 5 |
---
|
| 6 |
+
|
| 7 |
+
You can download the model then run the inference scipts in https://github.com/Alibaba-NLP/WebAgent.
|
| 8 |
+
|
| 9 |
- Native agentic search reasoning model using ReAct framework towards autonomous information seeking agency and Deep Research-like model.
|
| 10 |
- We introduce a four-stage training paradigm comprising browsing data construction, trajectory sampling, supervised fine-tuning for effective cold start, and reinforcement learning for improved generalization, enabling the agent to autonomously acquire autonomous search and reasoning skills.
|
| 11 |
- Our data-centric approach integrates trajectory-level supervision fine-tuning and reinforcement learning (DAPO) to develop a scalable pipeline for training agentic systems via SFT or RL.
|