Extremely quantized LLMs degrade in smoothness, sparsifying the decoding tree and hurting generation quality; a smoothness-preserving principle delivers gains beyond numerical fitting.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
EnergyFlow shows that denoising score matching on diffusion policies recovers the gradient of the expert's soft Q-function under maximum-entropy optimality, enabling non-adversarial reward extraction and improved policy generalization.
citing papers explorer
-
Fitting Is Not Enough: Smoothness in Extremely Quantized LLMs
Extremely quantized LLMs degrade in smoothness, sparsifying the decoding tree and hurting generation quality; a smoothness-preserving principle delivers gains beyond numerical fitting.
-
Recovering Hidden Reward in Diffusion-Based Policies
EnergyFlow shows that denoising score matching on diffusion policies recovers the gradient of the expert's soft Q-function under maximum-entropy optimality, enabling non-adversarial reward extraction and improved policy generalization.