LSG Attention: Extrapolation of Pretrained Transformers to Long Sequences
Charles Condevaux, Sébastien Harispe
Topics & Concepts
ExtrapolationComputer scienceTransformerArtificial intelligenceMachine learningElectrical engineeringMathematicsStatisticsVoltageEngineeringBlind Source Separation TechniquesMagnetic Properties and ApplicationsComputational Physics and Python Applications