research
∙
06/30/2023
Should you marginalize over possible tokenizations?
Autoregressive language models (LMs) map token sequences to probabilitie...
share
research
∙
03/08/2023
disco: a toolkit for Distributional Control of Generative Models
Pre-trained language models and other generative models have revolutioni...
research
∙
02/16/2023
Aligning Language Models with Preferences through f-divergence Minimization
Aligning language models with preferences can be posed as approximating ...
research
∙
06/01/2022
On Reinforcement Learning and Distribution Matching for Fine-Tuning Language Models with no Catastrophic Forgetting
The availability of large pre-trained models is changing the landscape o...
research
∙
12/10/2021
Sampling from Discrete Energy-Based Models with Quality/Efficiency Trade-offs
Energy-Based Models (EBMs) allow for extremely flexible specifications o...
research
∙
12/01/2021
Controlling Conditional Language Models with Distributional Policy Gradients
Machine learning is shifting towards general-purpose pretrained generati...
research
∙
06/09/2021
Energy-Based Models for Code Generation under Compilability Constraints
Neural language models can be successfully trained on source code, leadi...
research
∙
12/21/2020
A Distributional Approach to Controlled Text Generation
We propose a Distributional Approach to address Controlled Text Generati...
research
∙
12/18/2019
Distributional Reinforcement Learning for Energy-Based Sequential Models
Global Autoregressive Models (GAMs) are a recent proposal [Parshakova et...
research
∙
11/12/2019
Character-based NMT with Transformer
Character-based translation has several appealing advantages, but its pe...
research
∙
10/31/2019
Machine Translation of Restaurant Reviews: New Corpus for Domain Adaptation and Robustness
We share a French-English parallel corpus of Foursquare restaurant revie...
research
∙
09/16/2019
Global Autoregressive Models for Data-Efficient Sequence Learning
Standard autoregressive seq2seq models are easily trained by max-likelih...
research
∙
12/24/2018
Moment Matching Training for Neural Machine Translation: A Preliminary Study
In previous works, neural sequence models have been shown to improve sig...
research
∙
11/04/2018
Char2char Generation with Reranking for the E2E NLG Challenge
This paper describes our submission to the E2E NLG Challenge. Recently, ...
research
∙
09/20/2018
Symbolic Priors for RNN-based Semantic Parsing
Seq2seq models based on Recurrent Neural Networks (RNNs) have recently r...
research
∙
07/08/2016
Log-Linear RNNs: Towards Recurrent Neural Networks with Flexible Prior Knowledge
We introduce LL-RNNs (Log-Linear RNNs), an extension of Recurrent Neural...
research
∙
05/05/2016
LSTM-based Mixture-of-Experts for Knowledge-Aware Dialogues
We introduce an LSTM-based method for dynamically integrating several wo...
research
∙
10/07/2015
Assisting Composition of Email Responses: a Topic Prediction Approach
We propose an approach for helping agents compose email replies to custo...
research
∙
07/03/2012