source_context

{
  "surface": "signal_canvas",
  "mode": "paper",
  "query": "CLIP Architecture for Abdominal CT Image-Text Alignment and Zero-Shot Learning: Investigating Batch Composition and Data Scaling",
  "normalized_query": "2604.13561",
  "route": "/signal-canvas/clip-architecture-for-abdominal-ct-image-text-alignment-and-zero-shot-learning-investigating-batch-composition-and-data",
  "paper_ref": "clip-architecture-for-abdominal-ct-image-text-alignment-and-zero-shot-learning-investigating-batch-composition-and-data",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}