04:00
2026-07-14
arxiv.org
large-language-models
Silent Failures in Quantized LLM Reasoning: A Taxonomy-Based Analysis of Hollow Convergence and Failure Mode Shifts
Researchers found that quantizing large language models to lower precision can silently alter their reasoning processes even when overall task accuracy remains stable. A taxonomy-based analysis of 30,…