Similares a Reinforce Adjoint Matching: Scaling Diffusion RL

20 vecinos (text_768) · 8.4K views · Microsoft Research · Estados Unidos


★	Reinforce Adjoint Matching: Scaling Diffusion RL	Microsoft Research @microsoftresearch	8.4K	Estados Unidos
16	Stanford CME296 Diffusion & Large Vision Models \| Spring 2026 \| Lecture 2 - Score matching	Stanford Online @stanfordonline	19.8K	Estados Unidos	87
2	Convergence Analysis for Fast High-Order ODE Solvers in Diffusion Probabilistic Models	Microsoft Research @microsoftresearch	1.4K	Estados Unidos	90
14	Normalized Discounted Cumulative Gain (NDCG)	DigitalSreeni @digitalsreeni	1.1K	Estados Unidos	87
13	Large-scale agentic quant research with Weights & Biases	Weights & Biases @weightsbiases	1.0K	Estados Unidos	87
6	A non-Markovian approach to diffusion-based sampling	Microsoft Research @microsoftresearch	428	Estados Unidos	89
15	New fine-tuning of language models: Match meaning, not tokens	Microsoft Research @microsoftresearch	379	Estados Unidos	87
10	Q-learning with Flow-Matching Policies	Microsoft Research @microsoftresearch	363	Estados Unidos	88
3	Meta Flow Maps	Microsoft Research @microsoftresearch	266	Estados Unidos	90
11	Physics and information theory of generative diffusion	Microsoft Research @microsoftresearch	237	Estados Unidos	88
20	Tractable Mapping Entropy and Generative Backmapping via Split-Flows	Microsoft Research @microsoftresearch	216	Estados Unidos	87
17	Where the Score Lives: What Wavelets Reveal About Diffusion Models	Microsoft Research @microsoftresearch	184	Estados Unidos	87
1	Extending measure dynamics beyond generative modeling	Microsoft Research @microsoftresearch	179	Estados Unidos	90
4	Blind denoising diffusion models and the blessings of dimensionality	Microsoft Research @microsoftresearch	120	Estados Unidos	89
8	Deep Image Retrieval: Learning global representations for image search	Xavi Giró-i-Nieto @xavigiro-i-nieto	104	—	89
5	Matching features, not tokens: Energy-based fine-tuning of language models	Microsoft Research @microsoftresearch	96	Estados Unidos	89
7	A Unified Approach to Analysis and Design of Denoising Markov Models	Microsoft Research @microsoftresearch	89	Estados Unidos	89
12	Coati Optimization Algorithm-Driven Contentadaptive Steganography with Transformer Attention	Computer Science & IT Conference Proceedings @computerscienceitconferenc7375	24	Australia	87
19	Demystifying Reward Design in Reinforcement Learning for Upper Extremity Interaction: Practical G...	ACM SIGCHI @sigchi	24	Estados Unidos	87
18	Adaptive Multi-Stage Vector Retrieval for Retrieval-Augmented Generation	Computer Science & IT Conference Proceedings @computerscienceitconferenc7375	8	Australia	87
9	Why Weight Decay IS L2 Regularization: Complete Guide Explained	THE FACT FACTORY @thefactfactoryf	5	—	88