activity
20182022
most citedOn the Predictive Power of Neural Language Models for Human Real-Time Comprehension Behavior

110 citations · 203 across the 10 of their papers we have counts for

collaborators
Showing cs.CLShow all

16 papers · 1 filter

cs.CL20222 cited

Probing for Incremental Parse States in Autoregressive Language Models

Tiwalayo Eisape, Vineet Gangireddy, Roger P. Levy +1

Next-word predictions from autoregressive neural language models show remarkable sensitivity to syntax. This work evaluates the extent to which this behavior arises as a result of…

cs.CL20222 cited

When Does Syntax Mediate Neural Language Model Performance? Evidence from Dropout Probes

Mycal Tucker, Tiwalayo Eisape, Peng Qian +2

Recent causal probing literature reveals when language models and syntactic probes use similar representations. Such techniques may yield "false negative" causality results: models…

cs.CL2021

Revisiting the Uniform Information Density Hypothesis

Clara Meister, Tiago Pimentel, Patrick Haller +3

The uniform information density (UID) hypothesis posits a preference among language users for utterances structured such that information is distributed uniformly across a signal.…

cs.CL2021

Controlled Evaluation of Grammatical Knowledge in Mandarin Chinese Language Models

Yiwen Wang, Jennifer Hu, Roger Levy +1

Prior work has shown that structural supervision helps English language models learn generalizations about syntactic phenomena such as subject-verb agreement. However, it remains u…

cs.CL2021

Structural Guidance for Transformer Language Models

Peng Qian, Tahira Naseem, Roger Levy +1

Transformer-based language models pre-trained on large amounts of text data have proven remarkably successful in learning generic transferable linguistic representations. Here we s…

cs.CL2021

What if This Modified That? Syntactic Interventions via Counterfactual Embeddings

Mycal Tucker, Peng Qian, Roger Levy

Neural language models exhibit impressive performance on a variety of tasks, but their internal reasoning may be difficult to understand. Prior art aims to uncover meaningful prope…