Unveiling Downstream Performance Scaling of LLMs: A Clustering-Based Perspective Paper • 2502.17262 • Published Feb 24 • 21
MAGA: MAssive Genre-Audience Reformulation to Pretraining Corpus Expansion Paper • 2502.04235 • Published Feb 6 • 22
AttentionInfluence: Adopting Attention Head Influence for Weak-to-Strong Pretraining Data Selection Paper • 2505.07293 • Published 25 days ago • 26
