GAST: Gradient-aligned Sparse Tuning of Large Language Models with Data-layer Selection

source_context

{
  "surface": "signal_canvas",
  "mode": "paper",
  "query": "GAST: Gradient-aligned Sparse Tuning of Large Language Models with Data-layer Selection",
  "normalized_query": "2603.09865",
  "route": "/signal-canvas/gast-gradient-aligned-sparse-tuning-of-large-language-models-with-data-layer-selection",
  "paper_ref": "gast-gradient-aligned-sparse-tuning-of-large-language-models-with-data-layer-selection",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}