WFT approximates supervised fine-tuning for LLM personalization by transporting supervised residuals from author history to the current prompt through a dropout-estimated logit-space operator, without updating any weights.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Weightless Fine-Tuning: Personalizing LLMs via Logit-Space Transport
WFT approximates supervised fine-tuning for LLM personalization by transporting supervised residuals from author history to the current prompt through a dropout-estimated logit-space operator, without updating any weights.