AdaFuse: Accelerating Dynamic Adapter Inference via Token-Level Pre-Gating and Fused Kernel Optimization

source_context

{
  "surface": "signal_canvas",
  "mode": "paper",
  "query": "AdaFuse: Accelerating Dynamic Adapter Inference via Token-Level Pre-Gating and Fused Kernel Optimization",
  "normalized_query": "2603.11873",
  "route": "/signal-canvas/adafuse-accelerating-dynamic-adapter-inference-via-token-level-pre-gating-and-fused-kernel-optimization",
  "paper_ref": "adafuse-accelerating-dynamic-adapter-inference-via-token-level-pre-gating-and-fused-kernel-optimization",
  "topic_slug": null,
  "benchmark_ref": null,
  "dataset_ref": null
}