REINFORCE

{
  "doc": "nguyen-2025-sfr-deepresearch.md",
  "context": "Classical RL algorithm adapted by the paper with modifications including clipped-surrogate losses and length-normalized advantages for agentic training.",
  "category": "ai",
  "norm_label": "REINFORCE",
  "graphify_id": "reinforce_algorithm",
  "source_file": "nguyen-2025-sfr-deepresearch.md",
  "imported_from": "/Users/antonborzov/Documents/Research.nosync/papers/extract_typed_out/nguyen-2025-sfr-deepresearch/graph.json",
  "extracted_type": "framework",
  "source_location": "§3.3",
  "graphify_file_type": "framework"
}