DeepAI
Log In Sign Up

PARENTing via Model-Agnostic Reinforcement Learning to Correct Pathological Behaviors in Data-to-Text Generation

10/21/2020
by   Clément Rebuffel, et al.
0

In language generation models conditioned by structured data, the classical training via maximum likelihood almost always leads models to pick up on dataset divergence (i.e., hallucinations or omissions), and to incorporate them erroneously in their own generations at inference. In this work, we build ontop of previous Reinforcement Learning based approaches and show that a model-agnostic framework relying on the recently introduced PARENT metric is efficient at reducing both hallucinations and omissions. Evaluations on the widely used WikiBIO and WebNLG benchmarks demonstrate the effectiveness of this framework compared to state-of-the-art models.

READ FULL TEXT

page 1

page 2

page 3

page 4

03/15/2018

Neural Text Generation: Past, Present and Beyond

This paper presents a systematic survey on recent development of neural ...
06/08/2020

ColdGANs: Taming Language GANs with Cautious Sampling Strategies

Training regimes based on Maximum Likelihood Estimation (MLE) suffer fro...
08/27/2021

ReGen: Reinforcement Learning for Text and Knowledge Base Generation using Pretrained Language Models

Automatic construction of relevant Knowledge Bases (KBs) from text, and ...
01/12/2020

Revisiting Challenges in Data-to-Text Generation with Fact Grounding

Data-to-text generation models face challenges in ensuring data fidelity...
10/12/2020

Improving Text Generation with Student-Forcing Optimal Transport

Neural language models are often trained with maximum likelihood estimat...
03/12/2021

Constrained Text Generation with Global Guidance – Case Study on CommonGen

This paper studies constrained text generation, which is to generate sen...
04/04/2022

Value Gradient weighted Model-Based Reinforcement Learning

Model-based reinforcement learning (MBRL) is a sample efficient techniqu...