Dataset and Qwen2.5-3B adapters (SFT → GRPO) for multi-turn mystery investigation with tools, unlock DAGs, and cited case closes.
Vaidik
VaidikML0508
AI & ML interests
exploring another way to use gradient decent
Recent Activity
liked a model 4 days ago
77ethers/alpa liked a model 4 months ago
VaidikML0508/qwen2.5-3b-investigation-grpo upvoted a collection 4 months ago
Mystery Investigation AgentOrganizations
None yet