Shows
Daily Paper CastOpen Problems in Mechanistic Interpretability
🤗 Upvotes: 10 | cs.LG
Authors:
Lee Sharkey, Bilal Chughtai, Joshua Batson, Jack Lindsey, Jeff Wu, Lucius Bushnaq, Nicholas Goldowsky-Dill, Stefan Heimersheim, Alejandro Ortega, Joseph Bloom, Stella Biderman, Adria Garriga-Alonso, Arthur Conmy, Neel Nanda, Jessica Rumbelow, Martin Wattenberg, Nandi Schoots, Joseph Miller, Eric J. Michaud, Stephen Casper, Max Tegmark, William Saunders, David Bau, Eric Todd, Atticus Geiger, Mor Geva, Jesse Hoogland, Daniel Murfet, Tom McGrath
Title:
Open Problems in Mechanistic Interpretability
Arxiv:
http://arxiv.org/abs/2501.16496v1
Abstract:
Mechanistic interpretability aims to understand the computational mechanisms underlying neural networks' cap...2025-01-3025 minDaily Paper CastBridging the Data Provenance Gap Across Text, Speech and Video
🤗 Upvotes: 3 | cs.AI, cs.CL, cs.CY, cs.LG, cs.MM
Authors:
Shayne Longpre, Nikhil Singh, Manuel Cherep, Kushagra Tiwary, Joanna Materzynska, William Brannon, Robert Mahari, Manan Dey, Mohammed Hamdy, Nayan Saxena, Ahmad Mustafa Anis, Emad A. Alghamdi, Vu Minh Chien, Naana Obeng-Marnu, Da Yin, Kun Qian, Yizhi Li, Minnie Liang, An Dinh, Shrestha Mohanty, Deividas Mataciunas, Tobin South, Jianguo Zhang, Ariel N. Lee, Campbell S. Lund, Christopher Klamm, Damien Sileo, Diganta Misra, Enrico Shippole, Kevin Klyman, Lester JV Miranda, Niklas Muennighoff, Seonghyeon Ye, Seungone Kim, Vipul Gupta, Vivek Sharma, Xuhui Zhou, Caiming Xiong, Luis Villa, Ste...2024-12-2625 min
Exploring the unknown, together.Roads to Research: A Panel Conversation on Research Question IdeationLet's demystify ML research. This panel hosted by Marzieh Fadaee focuses on how to ask research questions, featuring insight and advice from experienced researchers.Discussion includes:-some practical strategies or techniques you use for generating and refining research ideas in the field of ML-common pitfalls or mistakes that ML researchers should avoid when formulating research questions-focus on a specific niche or sub-field of ML versus exploring a variety of topicsLook forward to a lively discussion, followed by audience questions.About the speakers- Stella Biderman: I am...2024-05-221h 02
IA Sob Controle - Inteligência Artificial16: Chatbot brasileiro, GPT em provas universitárias, LLaMA para C++Neste episódio, recebemos Hugo Abonizio, da Maritaca AI para falar sobre os principais desafios para desenvolver um chatbot especialmente treinado para entender bem o português, sobre como o treinamento monolingual de um LLM pode influenciar no resultado, e sobre as pesquisas da Maritaca investigando o desempenho das diferentes versões do GPT em provas universitárias.
Vem ver quem participou desse papo:
Marcus Mendes, host sob controle
Paulo Silveira, CEO da Alura
SĂ©rgio Lopes, CTO da Alura
Mário Souto, o “Dev Soutinho“, desenvolvedor no Nubank
Hugo Abonizio, Engenheiro de Machine Learning na Maritaca A...2023-07-2845 min
Science VsAI: Is It Out Of Control?Artificial Intelligence seems more human-like and capable than ever before — but how did it get so good so quickly? Today, we’re pulling back the curtain to find out exactly how AI works. And we'll dig into one of the biggest problems that scientists are worried about here: The ability of AI to trick us. We talk to Dr. Sasha Luccioni and Professor Seth Lazar about the science.This episode contains explicit language.There’s also a brief mention of suicide, so please take care when listening. Here are some c...2023-06-0842 min
The Changelog: Software Development, Open SourceANTHOLOGY — Open source AI (Interview)This week on The Changelog we’re taking you to the hallway track of The Linux Foundation’s Open Source Summit North America 2023 in Vancouver, Canada. Today’s anthology episode features: Beyang Liu (Co-founder and CTO at Sourcegraph), Denny Lee (Developer Advocate at Databricks), and Stella Biderman (Executive Director and Head of Research at EleutherAI).
Special thanks to our friends at GitHub for sponsoring us to attend this conference as part of Maintainer Month.
Join the discussionChangelog++ members get a bonus 3 minutes at the end of this episode and zero ads. Join today...2023-05-241h 38Changelog Master FeedANTHOLOGY — Open source AI (Changelog Interviews #541)This week on The Changelog we’re taking you to the hallway track of The Linux Foundation’s Open Source Summit North America 2023 in Vancouver, Canada. Today’s anthology episode features: Beyang Liu (Co-founder and CTO at Sourcegrpah), Denny Lee (Developer Advocate at Databricks), and Stella Biderman (Executive Director and Head of Research at EleutherAI).
Special thanks to our friends at GitHub for sponsoring us to attend this conference as part of Maintainer Month.
Leave us a comment
Changelog++ members get a bonus 3 minutes at the end of this episode and zero ads. Join t...2023-05-241h 38Changelog InterviewsANTHOLOGY — Open source AIThis week on The Changelog we’re taking you to the hallway track of The Linux Foundation’s Open Source Summit North America 2023 in Vancouver, Canada. Today’s anthology episode features: Beyang Liu (Co-founder and CTO at Sourcegraph), Denny Lee (Developer Advocate at Databricks), and Stella Biderman (Executive Director and Head of Research at EleutherAI).
Special thanks to our friends at GitHub for sponsoring us to attend this conference as part of Maintainer Month.
Join the discussionChangelog++ members get a bonus 3 minutes at the end of this episode and zero ads. Join today...2023-05-241h 38
Gradient Dissent: Conversations on AIHow EleutherAI Trains and Releases LLMs: Interview with Stella BidermanOn this episode, we’re joined by Stella Biderman, Executive Director at EleutherAI and Lead Scientist - Mathematician at Booz Allen Hamilton.EleutherAI is a grassroots collective that enables open-source AI research and focuses on the development and interpretability of large language models (LLMs).We discuss:- How EleutherAI got its start and where it's headed.- The similarities and differences between various LLMs.- How to decide which model to use for your desired outcome.- The benefits and challenges of reinforcement learning from human feedback....2023-05-0457 min
Machine Learning Street Talk (MLST)#84 LAURA RUIS - Large language models are not zero-shot communicators [NEURIPS UNPLUGGED]In this NeurIPSs interview, we speak with Laura Ruis about her research on the ability of language models to interpret language in context. She has designed a simple task to evaluate the performance of widely used state-of-the-art language models and has found that they struggle to make pragmatic inferences (implicatures). Tune in to learn more about her findings and what they mean for the future of conversational AI.
Laura Ruis
https://www.lauraruis.com/
https://twitter.com/LauraRuis
BLOOM
https://bigscience.huggingface.co...2022-12-0627 min