quanquangu.bsky.social
Professor @UCLA, Research Scientist @ByteDance | Recent work: SPIN, SPPO, DPLM 1/2, GPM, MARS | Opinions are my own
👤 Joined on 19 November 2024
Followers
Following
Total Posts
@quanquangu.bsky.social✈️ Excited to meet fellow enthusiasts at #NeurIPS2024! Here’s a preview of my presentation agenda:
Show PostPapers #2-3: arxiv.org/abs/2402.10210 and arxiv.org/abs/2405.00675 from the incredible @quanquangu.bsky.social. I really like how they explore new techniques for RLHF
Show Post▶️ Yue Wu presented "Self-Play Preference Optimization for Language Model Alignment" openreview.net/forum?id=Z1P... Yue Wu, Zhiqing Sun, @angelahzyuan.bsky.social, Kaixuan Ji, Yiming Yang, @quanquangu.bsky.social #NeurIPS2024 #AdaptiveFoundationModels
Show Posti don't buy the "pretraining ends" argument as well, probably the large marginal gain now we have will end, but still we generate more data and architecture will keep improve
Show Post