4 papers
RTTC: Reward-Guided Collaborative Test-Time Compute
J. Pablo Muñoz, Jinjie Yuan
Test-Time Compute (TTC) has emerged as a powerful paradigm for enhancing the performance of Large Language Models (LLMs) at inference, leveraging strategies such as Test-Time Train…
Is GPT-3 Text Indistinguishable from Human Text? Scarecrow: A Framework for Scrutinizing Machine Text
Yao Dou, Maxwell Forbes, Rik Koncel-Kedziorski +2
Modern neural language models can produce remarkably fluent and grammatical text. So much, in fact, that recent work by Clark et al. (2021) has reported that conventional crowdsour…
Go Forth and Prosper: Language Modeling with Ancient Textual History
Rik Koncel-Kedziorski, Noah A. Smith
We introduce a technique for improving document-level language models (LM) by leveraging "ancient history": text that is outside the LM's current context window. We learn an auxili…
Explaining Relationships Between Scientific Documents
Kelvin Luu, Xinyi Wu, Rik Koncel-Kedziorski +3
We address the task of explaining relationships between two scientific documents using natural language text. This task requires modeling the complex content of long technical docu…