sayzardAkshay (@akshay_pachaar)
해당 연구의 논문(arXiv: 2510.08191v1)과 구현 코드가 공개되었습니다. 코드 리포지토리는 GitHub의 TencentCloudADP 계정 아래 youtu-agent 저장소의 training_free_GRPO 브랜치로 제공되어 연구 재현과 실험 확인이 가능합니다(논문·코드 링크 포함).
Akshay (@akshay_pachaar)
텐센트 연구진이 발표한 'Training-Free GRPO'는 파라미터 업데이트(미세조정) 없이도 강화학습(RL)과 동등한 성능을 달성할 수 있다고 주장하며, 기존 RL 비용(약 $10,000)을 약 $18 수준으로 대폭 절감할 수 있다고 소개합니다. 핵심은 모델 파라미터를 바꾸지 않고 보상 최적화 방식을 변경해 성능을 얻는다는 점입니다.
Hacker NewsFast-DLLM: Training-Free Acceleration of Diffusion LLM
https://arxiv.org/abs/2505.22618
#HackerNews #FastDLLM #TrainingFree #DiffusionLLM #AIresearch #MachineLearning
Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding
Diffusion-based large language models (Diffusion LLMs) have shown promise for non-autoregressive text generation with parallel decoding capabilities. However, the practical inference speed of open-sourced Diffusion LLMs often lags behind autoregressive models due to the lack of Key-Value (KV) Cache and quality degradation when decoding multiple tokens simultaneously. To bridge this gap, we introduce a novel block-wise approximate KV Cache mechanism tailored for bidirectional diffusion models, enabling cache reuse with negligible performance drop. Additionally, we identify the root cause of generation quality degradation in parallel decoding as the disruption of token dependencies under the conditional independence assumption. To address this, we propose a confidence-aware parallel decoding strategy that selectively decodes tokens exceeding a confidence threshold, mitigating dependency violations and maintaining generation quality. Experimental results on LLaDA and Dream models across multiple LLM benchmarks demonstrate up to \textbf{27.6$\times$ throughput} improvement with minimal accuracy loss, closing the performance gap with autoregressive models and paving the way for practical deployment of Diffusion LLMs.
optionpocketElite Closing Academy | Sales Training Birmingham - Helping you turn more of your leads into paying customers.
Website :https://eliteclosingacademy.com/
link : https://youtu.be/4uzN-4a02tg
#Sales #Training #Birmingham #Wigan #Nottingham #Programs #Birmingham #SalesTraining #Techniques #TrainingFree #TrainingMethods #Courses #B2BSalesTraining
Elite Closing Academy | Sales Training Birmingham - Helping you turn more of your leads into paying customers.
The Elite Closing Academy is for entrepreneurs, business owners and sales teams who want to dramatically increase their sales without coming across as hard sell, pushy or manipulative.
Website :https://eliteclosingacademy.com/
link :https://youtu.be/4uzN-4a02tg
#Sales #Training #Birmingham #Wigan #Nottingham #Programs #Birmingham #SalesTraining #Techniques #TrainingFree