134 citations · 236 across the 9 of their papers we have counts for
12 papers
Match the Distribution, Not the Compute: Post-Training Multi-Token Prediction Heads
Prachi Badarayani, Aidan Jay, Chenghui Zhou +8
Multi-token prediction (MTP) improves the throughput of autoregressive generation by enabling the language model to draft multiple next tokens per forward pass, while a verificatio…
Creating generalizable downstream graph models with random projections
Anton Amirov, Chris Quirk, Jennifer Neville
We investigate graph representation learning approaches that enable models to generalize across graphs: given a model trained using the representations from one graph, our goal is…
Probing Factually Grounded Content Transfer with Factual Ablation
Peter West, Chris Quirk, Michel Galley +1
Despite recent success, large neural models often generate factually incorrect text. Compounding this is the lack of a standard automatic evaluation for factuality--it cannot be me…
Text Editing by Command
Felix Faltings, Michel Galley, Gerold Hintz +4
A prevailing paradigm in neural text generation is one-shot generation, where text is produced in a single step. The one-shot setting is inadequate, however, when the constraints t…
Examination and Extension of Strategies for Improving Personalized Language Modeling via Interpolation
Liqun Shao, Sahitya Mantravadi, Tom Manzini +4
In this paper, we detail novel strategies for interpolating personalized language models and methods to handle out-of-vocabulary (OOV) tokens to improve personalized language model…
A Controllable Model of Grounded Response Generation
Zeqiu Wu, Michel Galley, Chris Brockett +8
Current end-to-end neural conversation models inherently lack the flexibility to impose semantic control in the response generation process, often resulting in uninteresting respon…