Capture the Flags: Family-Based Evaluation of Agentic LLMs via Semantics-Preserving Transformations

source_context

{
  "surface": "signal_canvas",
  "mode": "paper",
  "query": "Capture the Flags: Family-Based Evaluation of Agentic LLMs via Semantics-Preserving Transformations",
  "normalized_query": "2602.05523",
  "route": "/signal-canvas/capture-the-flags-family-based-evaluation-of-agentic-llms-via-semantics-preserving-transformations",
  "paper_ref": "capture-the-flags-family-based-evaluation-of-agentic-llms-via-semantics-preserving-transformations",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}