Generating texts under constraint through discriminator-guided MCTS

by   Antoine Chaffin, et al.

Large pre-trained language models (LM) based on Transformers allow to generate very plausible long texts. In this paper, we explore how this generation can be further controlled to satisfy certain constraints (eg. being non-toxic, positive or negative, convey certain emotions, etc.) without fine-tuning the LM. Precisely, we formalize constrained generation as a tree exploration process guided by a discriminator according to how well the associated sequence respects the constraint. Using a discriminator to guide this generation, rather than fine-tuning the LM, in addition to be easier and cheaper to train, allows to apply the constraint more finely and dynamically. We propose several original methods to search this generation tree, notably the Monte Carlo Tree Search (MCTS) which provides theoretical guarantees on the search efficiency, but also simpler methods based on re-ranking a pool of diverse sequences using the discriminator scores. We evaluate these methods on two types of constraints and languages: review polarity and emotion control in French and English. We show that MCTS achieves state-of-the-art results in constrained generation, without having to tune the language model, in both tasks and languages. We also demonstrate that our other proposed methods based on re-ranking can be really effective when diversity among the generated propositions is encouraged.


page 1

page 2

page 3

page 4


Exploring Versatile Generative Language Model Via Parameter-Efficient Transfer Learning

Fine-tuning pre-trained generative language models to down-stream langua...

Language Generation via Combinatorial Constraint Satisfaction: A Tree Search Enhanced Monte-Carlo Approach

Generating natural language under complex constraints is a principled fo...

Making Large Language Models Better Reasoners with Alignment

Reasoning is a cognitive process of using evidence to reach a sound conc...

Which Discriminator for Cooperative Text Generation?

Language models generate texts by successively predicting probability di...

Sequential Monte Carlo Steering of Large Language Models using Probabilistic Programs

Even after fine-tuning and reinforcement learning, large language models...

Controlling Perceived Emotion in Symbolic Music Generation with Monte Carlo Tree Search

This paper presents a new approach for controlling emotion in symbolic m...

GPoeT-2: A GPT-2 Based Poem Generator

This project aims to produce the next volume of machine-generated poetry...

Please sign up or login with your details

Forgot password? Click here to reset