paper-with-me

홈 › Papers

Faster WIND: Accelerating Iterative Best-of-$N$ Distillation for LLM Alignment

2024-10-28 · Tong Yang, Jincheng Mei, Hanjun Dai, Zixin Wen, Shicong Cen, Dale Schuurmans, Yuejie Chi, Bo Dai

Recent advances in aligning large language models with human preferences have corroborated the growing importance of best-of-N distillation (BOND). However, the iterative BOND algorithm is prohibitively expensive in practice due to the sample and computation inefficiency. This paper addresses the problem by revealing a unified game-theoretic connection between iterative BOND and self-play alignment, which unifies seemingly disparate algorithmic paradigms. Based on the connection, we establish a novel framework, WIN rate Dominance (WIND), with a series of efficient algorithms for regularized win rate dominance optimization that approximates iterative BOND in the parameter space. We provides provable sample efficiency guarantee for one of the WIND variant with the square loss objective. The experimental results confirm that our algorithm not only accelerates the computation, but also achieves superior sample efficiency compared to existing methods.

📄 PDF Abstract BibTeX arXiv:2410.20727

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Accelerating Diffusion-based Combinatorial Optimization Solvers by Progressive Distillation

2023-08-12 · Junwei Huang, Zhiqing Sun, Yiming Yang

Graph-based diffusion models have shown promising results in terms of generating high-quality solutions to NP-complete (NPC) combinatorial optimization (CO) problems. However, those models are often inefficient in infere…

Combinatorial OptimizationDenoising

Presto! Distilling Steps and Layers for Accelerating Music Generation

2024-10-07 · Zachary Novack, Ge Zhu, Jonah Casebeer, Julian McAuley 외

Despite advances in diffusion-based text-to-music (TTM) methods, efficient, high-quality generation remains a challenge. We introduce Presto!, an approach to inference acceleration for score-based diffusion transformers …

DiversityMusic Generation

Algorithmic Optimisations for Iterative Deconvolution Methods

2013-04-26 · Martin Welk, Martin Erler

We investigate possibilities to speed up iterative algorithms for non-blind image deconvolution. We focus on algorithms in which convolution with the point-spread function to be deconvolved is used in each iteration, and…

Image Deconvolution

FasterVoiceGrad: Faster One-step Diffusion-Based Voice Conversion with Adversarial Diffusion Conversion Distillation

2025-08-25 · Takuhiro Kaneko, Hirokazu Kameoka, Kou Tanaka, Yuto Kondo arxiv

A diffusion-based voice conversion (VC) model (e.g., VoiceGrad) can achieve high speech quality and speaker similarity; however, its conversion process is slow owing to iterative sampling. FastVoiceGrad overcomes this li…

Voice Conversion

AddSR: Accelerating Diffusion-based Blind Super-Resolution with Adversarial Diffusion Distillation

2024-04-02 · Rui Xie, Chen Zhao, Kai Zhang, Zhenyu Zhang 외

Blind super-resolution methods based on stable diffusion showcase formidable generative capabilities in reconstructing clear high-resolution images with intricate details from low-resolution inputs. However, their practi…

Blind Super-ResolutionSuper-Resolution