TBPO posits a token-level Bradley-Terry model and derives a Bregman-divergence density-ratio matching loss that generalizes DPO while preserving token-level optimality.
Title resolution pending
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
representative citing papers
Medusa augments LLMs with multiple decoding heads and tree-based attention to predict and verify several tokens in parallel, yielding 2.2-3.6x inference speedup via two fine-tuning regimes.
GradShield removes data points likely to cause safety misalignment during LLM finetuning by computing a Finetuning Implicit Harmfulness Score and applying adaptive thresholding, keeping attack success rates below 6% while preserving utility.
Human-AI hybrids achieve only +0.4pp over AI alone on diverse tasks because confidence routing fails to identify the small set of cases where humans can correct AI errors.
citing papers explorer
-
TokenRatio: Principled Token-Level Preference Optimization via Ratio Matching
TBPO posits a token-level Bradley-Terry model and derives a Bregman-divergence density-ratio matching loss that generalizes DPO while preserving token-level optimality.
-
Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads
Medusa augments LLMs with multiple decoding heads and tree-based attention to predict and verify several tokens in parallel, yielding 2.2-3.6x inference speedup via two fine-tuning regimes.
-
GradShield: Alignment Preserving Finetuning
GradShield removes data points likely to cause safety misalignment during LLM finetuning by computing a Finetuning Implicit Harmfulness Score and applying adaptive thresholding, keeping attack success rates below 6% while preserving utility.
-
Toward Human-AI Complementarity Across Diverse Tasks
Human-AI hybrids achieve only +0.4pp over AI alone on diverse tasks because confidence routing fails to identify the small set of cases where humans can correct AI errors.