Published Mar 23, 2023

Sewon Min: The Science of Natural Language

Sewon Min delves into advancements in natural language processing, examining Dense Passage Retrieval and in-context learning, and addressing the complexities of model hallucination and quality assurance benchmarks for effective question-answering systems.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

Kyunghyun Cho: Neural Machine Translation, Language, and Doing Good Science
Answers 383 questions
Linus Lee: At the Boundary of Machine and Mind
Answers 383 questions
Terry Winograd: AI, HCI, Language, and Cognition
Answers 383 questions
Eric Jang on Robots Learning at Google and Generalization via Language
Answers 383 questions
Christopher Manning: Linguistics and the Development of NLP
Answers 383 questions
Yejin Choi: Teaching Machines Common Sense and Morality
Answers 383 questions
Max Woolf: Data Science at BuzzFeed and AI Content Generation
Answers 383 questions
Chip Huyen: Machine Learning Tools and Systems
Answers 383 questions
Been Kim: Interpretable Machine Learning
Answers 383 questions
Ed Grefenstette: Language, Semantics, Cohere
Answers 383 questions
Stevan Harnad: AI's Symbol Grounding Problem
Answers 383 questions
Joon Park: Generative Agents and Human-Computer Interaction
Answers 383 questions
Luis Voloch: AI and Biology
Answers 383 questions
Sebastian Raschka: AI Education and Research
Answers 383 questions
Hugo Larochelle: Deep Learning as Science
Answers 383 questions

Dexa/Daniel Bashir

Sewon Min: The Science of Natural Language

Topics covered

Popular Clips

Understanding Chain of Thought Prompting

Multi-hop Question Answering

Rethinking Demonstrations

Evaluating External Knowledge

Non Parametric Modeling

Navigating PhD Stress

Improving Question Answering

Building General Purpose ML Systems

Evolving NLP Research

Training Challenges

Improving Retrieval Methods

AI Research Journey

Compositionality in Meta ICL

Phenomenal Research

Non Parametric Language Modeling

Episode Highlights

Text Retrieval and Integration

In-Context Learning

Challenges in QA Benchmarks

Related Episodes

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

Linus Lee: At the Boundary of Machine and Mind

Terry Winograd: AI, HCI, Language, and Cognition

Eric Jang on Robots Learning at Google and Generalization via Language

Christopher Manning: Linguistics and the Development of NLP

Yejin Choi: Teaching Machines Common Sense and Morality

Max Woolf: Data Science at BuzzFeed and AI Content Generation

Chip Huyen: Machine Learning Tools and Systems

Been Kim: Interpretable Machine Learning

Ed Grefenstette: Language, Semantics, Cohere

Stevan Harnad: AI's Symbol Grounding Problem

Joon Park: Generative Agents and Human-Computer Interaction

Luis Voloch: AI and Biology

Sebastian Raschka: AI Education and Research

Hugo Larochelle: Deep Learning as Science

Sewon Min: The Science of Natural Language

Topics covered

Popular Clips

Episode Highlights

Text Retrieval and IntegrationSewon Min explores Dense Passage Retrieval and non-parametric language models, highlighting their potential to enhance question-answering tasks and address challenges like model hallucination and verification.

Text Retrieval and Integration

In-Context LearningSewon Min explores the intricacies of in-context learning (ICL), highlighting its mechanisms, the role of demonstrations, and task transfer efficiency. Her insights reveal the potential and challenges of ICL in advancing natural language processing.

In-Context Learning

Challenges in QA BenchmarksSewon Min, a PhD student in NLP, shares insights on constructing effective QA benchmarks and the challenges of ambiguity in QA systems. She emphasizes the importance of external knowledge and validation in ensuring robust, real-world applicable models.

Challenges in QA Benchmarks

Related Episodes