Channel Avatar

Allen Institute for AI @UCEqgmyWChwvt6MFGGlmUQCQ@youtube.com

None subscribers - no pronouns set

More from this channel (soon)


38:12
AI-enabled scientific discovery in natural world imagery
55:55
Concept Bottleneck Models for Text Classification
57:04
The Pre-trainer's toolkit: From dataset construction to model scaling
57:47
Cultivating Insights: AI-Infused Workflow Designs for Nurturing the Scientific Idea Garden
58:42
Towards a more contextualized view of the web
56:22
Optimization within Latent Spaces
59:11
Training Human-AI Teams
53:20
Making Health Knowledge Accessible Through Personalized Language Processing
57:24
Robot Learning by Understanding Egocentric Videos
57:21
Project Sidewalk: Crowd+AI Techniques to Map and Assess Every Sidewalk in the World
58:41
LMQL Programming Large Language Models
57:57
Does Generative AI Infringe Copyright?
59:26
Figuring out how the world works: causality in a world full of real people
59:47
Machine-Checked Proofs, and the Rise of Formal Methods in Mathematics
01:00:19
Beyond Test Accuracies for Studying Deep Neural Networks
58:19
Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hacking
59:47
Integrated Systems for Computational Scientific Discovery
58:41
Objective Mismatch in Reinforcement Learning from Human Feedback
49:49
Language AI for RNA Virus and RNA Vaccine
52:18
OpenWebMath: An Open Dataset of High-Quality Mathematical Web Text
46:21
The Worlds I See: Curiosity, Exploration and Discovery at the Dawn of AI
57:10
Optimal Transport Posterior Alignment for Cross-lingual Semantic Parsing
56:28
On Parameter Efficiency of Neural Language Models
53:23
Benchmarking Compositionality with Formal Languages
01:00:14
Studying Large Language Model Generalization with Influence Functions
58:20
Modular Language Models
56:51
Towards Cost-Efficient Use of Pre-trained Models
58:18
Reliability and interactive debugging for language models
55:10
The University of Washington eScience Institute: a Home for Data-Intensive Discovery
55:58
Reliable Evaluation and High-Quality Data: Building Blocks for Helpful Question Answering Systems
04:44
SATLAS: Open Geospatial Data Generated by AI
58:45
Vision Without Labels
56:25
Skill it! A Data-Driven Skills Framework for Understanding and Training Language Models
56:52
From Compression to Convection: A Latent Variable Perspective
01:03:51
Avenging Polanyi's Revenge
54:09
Generative AI & Copyright
58:43
Machine Learning in Climate Action
05:54
Do language models have coherent mental models of everyday things?
01:03:21
Imaginative Vision Language Models
57:51
Structure Modeling in Language Models
05:47
When Not to Trust Language Models: Investigating Effectiveness of Parametric&Non-Parametric Memories
01:00:27
Enhancing the Reliability and Continual Improvement of Neural Dialogue Systems
54:27
Toward Intelligent Writing Support Beyond Completing Sentences
57:57
Towards robust long-form text generation systems
58:04
Evaluating ethical and social risks from large models
01:06:09
"Open" AI: considering the ethical upsides and downsides of "Open" AI development
59:04
The BigScience Workshop
09:56
Moving Forward by Moving Backward: Embedding Action Impact over Action Semantics | AI2
58:10
Recycling finetuned models to pretrain: on loss spaces, fusing and evolving pretraining
58:04
Robot learning and perception for contact-rich manipulation
59:17
Environment-oriented modeling for grounded NLP and collective intelligence systems
58:10
Empowering Human-like Decision Making in AI Models through Explanations
57:58
Understanding and Improving Compositional Generalization | AI2
34:13
Artificial Intelligence and Ethics: Towards a Robust Normative Framework
58:50
No Language Left Behind Unlocking Text Data for Under Resourced | AI2
56:30
What Do NLP Researchers Believe? Results of the NLP Community Metasurvey
00:44
Global storm-resolving climate simulation in Python
04:31
DREAM: Improving Situational QA by First Elaborating the Situation
07:55
Just-DREAM-about-it: Figurative Language Understanding with DREAM-FLUTE
02:42
Towards Teachable Reasoning Systems: Using a Dynamic Memory of User Feedback for System Improvement