Oyvind Tafjord

Neologism Learning for Controllability and Self-Verbalization

SimpleToM: Exposing the Gap between Explicit ToM Inference and Implicit ToM Application in LLMs

OLMoE: Open Mixture-of-Experts Language Models

Tulu 3: Pushing Frontiers in Open Language Model Post-Training

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

2 OLMo 2 Furious (COLM’s Version)

From Models to Microtheories: Distilling a Model's Topical Knowledge for Grounded Question-Answering

Establishing Task Scaling Laws via Compute-Efficient Model Ladders

DataDecide: How to Predict Best Pretraining Data with Small Experiments

SimpleToM: Exposing the Gap between Explicit ToM Inference and Implicit ToM Application in LLMs