Hugging Face's logo

Collections

Discover the best community collections!

Collections trending this week
FactGuard — Evidence-Grounded Verification
Judges a claim only against given evidence (Qwen3.5/3.8 bases). Merged BF16 weights. Dev/counterfactual evals; real-answer accuracy unproven.
REA Environments on AgentEnv
50 RL environments REA built from recorded browser sessions of 43 self-hosted apps and 7 public APIs, replayed as MCP tools on AgentEnv.
FactGuard — Evidence-Grounded Verification
Judges a claim only against given evidence (Qwen3.5/3.8 bases). Merged BF16 weights. Dev/counterfactual evals; real-answer accuracy unproven.
REA Environments on AgentEnv
50 RL environments REA built from recorded browser sessions of 43 self-hosted apps and 7 public APIs, replayed as MCP tools on AgentEnv.