5 papers
Rigorous Interpretation Is a Form of Evaluation
Isabelle Lee, Emmy Liu, Cathy Jiao +4
Current machine learning models are evaluated through behavioral snapshots, with benchmark accuracies, win rates and outcome-based metrics. Model explanations and evaluations, howe…
What do Language Models Learn and When? The Implicit Curriculum Hypothesis
Emmy Liu, Kaiser Sun, Millicent Li +4
Large language models (LLMs) can perform remarkably complex tasks, yet the fine-grained details of how these capabilities emerge during pretraining remain poorly understood. Scalin…
To Memorize or to Retrieve: Scaling the Interaction Between Pretraining and Retrieval
Karan Singh, Michael Yu, Varun Gangal +4
Retrieval-augmented generation (RAG) improves language model (LM) performance by providing relevant context at test time for knowledge-intensive situations. In this work, we system…
Hybrid-Gym: Training Coding Agents to Generalize Across Tasks
Yiqing Xie, Emmy Liu, Gaokai Zhang +7
When assessing the quality of coding agents, predominant benchmarks focus on solving single issues on GitHub, such as SWE-Bench. In contrast, in real use, these agents solve more v…
Midtraining Bridges Pretraining and Posttraining Distributions
Emmy Liu, Graham Neubig, Chenyan Xiong
Midtraining, the practice of mixing specialized data with more general pretraining data in an intermediate training phase, has become widespread in language model development, yet…