טירונים שני-מסדריים של LLM בעלי זיכרון לינארי: דרך Nystr"om Sketching

תקציר מקורי באנגליתarXiv:2610.04204v2 Announce Type: replace-cross Abstract: Training large language models (LLMs) entails a fundamental trade-off: memory-efficient optimizers such as Adam discard cross-parameter curvature, whereas full-curvature methods such as SOAP can accelerate convergence at prohibitive memory costs. We introduce Clean, a memory-efficient and full-curvature optimizer designed to resolve this bottleneck. Clean leverages the randomized Nystrom method to accurately approximate the left and right preconditioners in SOAP, and to reduce the optimizer's memory complexity from quadratic to linear in terms of model dimensions. We subsequently reintegrate the off-subspace components to capture curvature information beyond the low-rank approximation, preserving rich curvature at minimal memory cos