ServiceNow IA recherche

PipelineRL: Faster On-policy Reinforcement Learning for Long Sequence Generation

Rafael Pardinas, Ehsan Kamalloo, Alexandre Piche, Dzmitry Bahdanau
janvier 2026
Article Citation Code
Type
Article de revue
Publication
Transactions on Machine Learning Research (TMLR)
Reinforcement Learning
Rafael Pardinas
Rafael Pardinas
Applied Research Scientist

Applied Research Scientist at Frontier AI Research located at [‘London (remote), UK’].

Ehsan Kamalloo
Ehsan Kamalloo
Research Scientist

Research Scientist at Frontier AI Research located at [‘Toronto (remote), Canada’].