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

On Linear Representations and Pretraining Data Frequency in Language Models

Jack MerulloNoah A. SmithSarah WiegreffeYanai Elazar
2025
ICLR

Pretraining data has a direct impact on the behaviors and quality of language models (LMs), but we only understand the most basic principles of this relationship. While most work focuses on… 

WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild

Bill Yuchen LinYuntian DengK. ChanduYejin Choi
2025
ICLR

We introduce WildBench, an automated evaluation framework designed to benchmark large language models (LLMs) using challenging, real-world user queries. WildBench consists of 1,024 tasks carefully… 

AI Safety Should Prioritize the Future of Work

Sanchaita HazraBodhisattwa Prasad MajumderTuhin Chakrabarty
2025
International Conference on Machine Learning (ICML)

Current efforts in AI safety prioritize filtering harmful content, preventing manipulation of human behavior, and eliminating existential risks in cybersecurity or biosecurity. While pressing, this… 

CodeScientist: End-to-End Semi-Automated Scientific Discovery with Code-based Experimentation

Peter JansenOyvind TafjordMarissa RadenskyPeter Clark
2025
ACL (Findings)

Despite the surge of interest in autonomous scientific discovery (ASD) of software artifacts (e.g., improved ML algorithms), current ASD systems face two key limitations: (1) they largely explore… 

OLMoE: Open Mixture-of-Experts Language Models

Niklas MuennighoffLuca SoldainiDirk GroeneveldHanna Hajishirzi
2025
arXiv.org

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… 

ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning

Bill Yuchen LinRonan Le BrasKyle RichardsonYejin Choi
2025
ICML

We investigate the logical reasoning capabilities of large language models (LLMs) and their scalability in complex non-monotonic reasoning. To this end, we introduce ZebraLogic, a comprehensive… 

2 OLMo 2 Furious

Pete WalshLuca SoldainiDirk GroeneveldHanna Hajishirzi
2025
arXiv.org

We present OLMo 2, the next generation of our fully open language models. OLMo 2 includes dense autoregressive models with improved architecture and training recipe, pretraining data mixtures, and… 

Transformers as Transducers

Lena StroblDana AngluinDavid ChiangAshish Sabharwal
2025
TACL

We study the sequence-to-sequence mapping capacity of transformers by relating them to finite transducers, and find that they can express surprisingly large classes of (total functional)… 

DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents

Peter JansenMarc-Alexandre CoteTushar KhotPeter Clark
2024
NeurIPS Datasets and Benchmarks

Automated scientific discovery promises to accelerate progress across scientific domains. However, developing and evaluating an AI agent's capacity for end-to-end scientific reasoning is challenging… 

Paloma: A Benchmark for Evaluating Language Model Fit

Ian MagnussonAkshita BhagiaValentin HofmannJesse Dodge
2024
NeurIPS

Language models (LMs) commonly report perplexity on monolithic data held out from training. Implicitly or explicitly, this data is composed of domains$\unicode{x2013}$varying distributions of…