Published Feb 29, 2024

Sasha Rush: Building Better NLP Systems

Join Sasha Rush, Associate Professor at Cornell University, as he delves into groundbreaking efficiency techniques and controllability in NLP, exploring the potential of instruction tuning and state space models to enhance model performance, safety, and interpretability.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

Hugo Larochelle: Deep Learning as Science
Answers 383 questions
Sasha Luccioni: Connecting the Dots Between AI's Environmental and Social Impacts
Answers 383 questions
Kyunghyun Cho: Neural Machine Translation, Language, and Doing Good Science
Answers 383 questions
Soumith Chintala: PyTorch
Answers 383 questions
Sebastian Raschka: AI Education and Research
Answers 383 questions
Christopher Manning: Linguistics and the Development of NLP
Answers 383 questions
Rishi Bommasani on Foundation Models
Answers 383 questions
Nicholas Thompson: AI and Journalism
Answers 383 questions
Kate Park: Data Engines for Vision and Language
Answers 383 questions
Luis Voloch: AI and Biology
Answers 383 questions
Stuart Russell: The Foundations of Artificial Intelligence
Answers 383 questions
Joel Lehman: Open-Endedness and Evolution through Large Models
Answers 383 questions
Linus Lee: At the Boundary of Machine and Mind
Answers 383 questions
Ben Wellington: ML for Finance and Storytelling through Data
Answers 383 questions
Some Changes at The Gradient
Answers 383 questions

Dexa/Daniel Bashir

Sasha Rush: Building Better NLP Systems

Topics covered

Popular Clips

Sequence Level Distillation

Building Research Communities

Ideal Deep Learning Library

Pytorch Revolution

Pytorch Implementation Style

Innovative Efficiency Approaches

Evolution of NLP

Controllable NLP Systems

NLP Model Safety

Researcher's Evolution

State Space Models

Efficient Language Model Approximations

Unraveling Prompting Mysteries

Evolution of NLP

Episode Highlights

Efficient NLP Techniques

Controllability in NLP

NLP Model Architecture

Related Episodes

Hugo Larochelle: Deep Learning as Science

Sasha Luccioni: Connecting the Dots Between AI's Environmental and Social Impacts

Kyunghyun Cho: Neural Machine Translation, Language, and Doing Good Science

Soumith Chintala: PyTorch

Sebastian Raschka: AI Education and Research

Christopher Manning: Linguistics and the Development of NLP

Rishi Bommasani on Foundation Models

Nicholas Thompson: AI and Journalism

Kate Park: Data Engines for Vision and Language

Luis Voloch: AI and Biology

Stuart Russell: The Foundations of Artificial Intelligence

Joel Lehman: Open-Endedness and Evolution through Large Models

Linus Lee: At the Boundary of Machine and Mind

Ben Wellington: ML for Finance and Storytelling through Data

Some Changes at The Gradient

Sasha Rush: Building Better NLP Systems

Topics covered

Popular Clips

Episode Highlights

Efficient NLP TechniquesSasha Rush explores sequence-level knowledge distillation and innovative efficiency techniques in NLP. He emphasizes the need for creative approaches to enhance model performance and efficiency.

Efficient NLP Techniques

Controllability in NLPSasha Rush explores the transformative role of instruction tuning in NLP systems, emphasizing its impact on controllability and safety. He shares insights into the challenges and advancements in making models more reliable and interpretable.

Controllability in NLP

NLP Model Architecture

Related Episodes