SOURCE-LINKED INTELLIGENCE

Dense Structural Compression of Transformers via Gauge-Correct Channel Removal

arXiv · AI, language, vision and robotics · article · Sep 7, 2026 · UTC

Inference energy per token drives the cost and carbon footprint of deployed transformers. It is dominated by dense matrix products that incur fused multiply-accumulate (FMA) operations and memory traffic. To reduce these computations while retaining dense tensors for high GPU throughput, we develop a methodology from first principles to adapt structural complexity during training to maximize inference utility per unit compute. Channel penalties drive entire tensor slices to zero to enable physical removal while preserving density and the network function. The natural approach, penalizing the n

Read original source ↗ Open in workspace

recordType

paper

region

Global