Ai2Latest and greatest: Ai2’s release notesAlong with our rebrand, we’re excited to debut a new release note process. Because we’re making regular updates and new asset roll-outs in…22h ago22h ago
Harsh TrivediAppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding AgentsJul 28Jul 28
Faeze BrahmanBroadening the Scope of Noncompliance: When and How AI Models Should Not Comply with User Requestsby Faeze Brahman and Sachin KumarJul 3Jul 3
Nouha DziriThe AI2 Safety Toolkit: Datasets and Models for Safe and Responsible LLMs DevelopmentLarge Language models (LLMs) have revolutionized how humans perform their daily tasks and are rapidly becoming more integral to humans’…Jun 28Jun 28
Ai2PolygloToxicityPrompts: Multilingual Evaluation of Neural Toxic Degeneration in Large Language…The presence of low-quality data on the internet leads to undesirable, unsafe, or toxic knowledge being instilled in large language models…Jun 24Jun 24
Ai2Data-driven Discovery with Large Generative ModelsHow do you boil the ocean? That impossible task is what researchers in every field try to accomplish when they sort through the existing…May 161May 161
Piper WoltersSatlasPretrain Models: Foundation Models for Satellite and Aerial ImageryWe’re excited to announce SatlasPretrain Models, a suite of open geospatial foundation models. Accompanied by their source code…Apr 23Apr 23
Jordan StewardRestoring Bahamas’ Seas: CBS Mornings Spotlights Skylight’s SupportThrough a mix of innovative partnerships and Skylight’s AI, learn how the country is protecting its ocean life.Apr 22Apr 22
Ai2OLMo 1.7–7B: A 24 point improvement on MMLUToday, we’ve released an updated version of our 7 billion parameter Open Language Model, OLMo 1.7–7B. This model scores 52 on MMLU, sitting…Apr 171Apr 171
Ai2Making a switch — Dolma moves to ODC-BYWe’re moving the Dolma dataset to the ODC-BY license. Here’s why.Apr 15Apr 15