SEP 10, 2026 · PREPRINT
Rebalancing Token Importance in Language Models with TF-IDF Weighted Cross-Entropy Loss
arXiv
Early-stage computational work demonstrating a proof-of-concept method across multiple model scales with surrogate metrics (memorization length, perplexity) but lacking peer review, clinical or real-world validation, and independent replication.
Reported
Memorization reduction (LoRA, acr…14%
Memorization reduction (full-weig…58%
Computational overhead<3%