Rethinking Calibration for High-Sparsity LLM Pruning: From Single Sources to Capability Coverage.

arXiv:2606.03328v4 Announce Type: replace-cross Abstract: Post-training LLM pruning relies on a small calibration set, usually chosen heuristically and judged by WikiText perplexity or by mean accuracy on general-domain commonsense benchmarks. This single-axis