Skip to main content ->
Ai2

Research - Papers

Explore a selection of our published work on a variety of key research challenges in AI.

Filter papers

Application of the AI2 Climate Emulator to E3SMv2's global atmosphere model, with a focus on precipitation fidelity

James P. C. DuncanElynn WuJean-Christoph Golazand Christopher S. Bretherton
2024
Journal of Geophysical Research - Machine Learning

Can the current successes of global machine learning-based weather simulators be generalized beyond 2-week forecasts to stable and accurate multiyear runs? The recently developed AI2 Climate… 

OLMoE: Open Mixture-of-Experts Language Models

Niklas MuennighoffLuca SoldainiDirk GroeneveldHannaneh Hajishirzi
2024
arXiv

We introduce OLMoE, a fully open, state-of-the-art language model leveraging sparse Mixture-of-Experts (MoE). OLMoE-1B-7B has 7 billion (B) parameters but uses only 1B per input token. We pretrain… 

Pushing the frontiers in climate modelling and analysis with machine learning

V. EyringWilliam D. CollinsPierre GentineLaure Zanna
2024
Nature Climate Change

Climate modelling and analysis are facing new demands to enhance projections and climate information. Here we argue that now is the time to push the frontiers of machine learning beyond… 

Weather and climate predicted accurately — without using a supercomputer

Oliver Watt-Meyer
2024
Nature

A cutting-edge global model of the atmosphere combines machine learning with a numerical model based on the laws of physics. This ‘hybrid’ system accurately predicts the weather — and even shows… 

AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agents

Harsh TrivediTushar KhotMareike HartmannNiranjan Balasubramanian
2024
ACL

Autonomous agents that address day-to-day digital tasks (e.g., ordering groceries for a household), must not only operate multiple apps (e.g., notes, messaging, shopping app) via APIs, but also… 

Can Language Models Serve as Text-Based World Simulators?

Ruoyao WangGraham ToddZiang XiaoP. Jansen
2024
ACL

Virtual environments play a key role in benchmarking advances in complex planning and decision-making tasks but are expensive and complicated to build by hand. Can current language models themselves… 

Few-shot Dialogue Strategy Learning for Motivational Interviewing via Inductive Reasoning

Zhouhang XieBodhisattwa Prasad MajumderMengjie ZhaoJulian McAuley
2024
ACL Findings

We consider the task of building a dialogue system that can motivate users to adopt positive lifestyle changes: Motivational Interviewing. Addressing such a task requires a system that can infer… 

The Unreasonable Effectiveness of Easy Training Data for Hard Tasks

Peter HaseMohit BansalPeter ClarkSarah Wiegreffe
2024
ACL

How can we train models to perform well on hard test data when hard training data is by definition difficult to label correctly? This question has been termed the scalable oversight problem and has… 

Data Contamination Report from the 2024 CONDA Shared Task

Oscar SainzIker Garc'ia-FerreroAlon JacoviJinglin Yang
2024
arXiv

The 1st Workshop on Data Contamination (CONDA 2024) focuses on all relevant aspects of data contamination in natural language processing, where data contamination is understood as situations where… 

Answer, Assemble, Ace: Understanding How Transformers Answer Multiple Choice Questions

Sarah WiegreffeOyvind TafjordYonatan BelinkovAshish Sabharwal
2024
arXiv

Multiple-choice question answering (MCQA) is a key competence of performant transformer language models that is tested by mainstream benchmarks. However, recent evidence shows that models can have…