← Search

Sidharth Mudgal

2 accepted papers

2024

Controlled Decoding from Language Models

ICML 2024poster

KL-regularized reinforcement learning (RL) is a popular alignment framework to control the language model responses towards high reward outcomes. We pose a tokenwise RL objective and propose a modular solver for it, called *controlled decoding (CD)*. CD exerts control through a separate *prefix scor…

Cited by 86SourcePDFScholar
2021

A Scalable Framework for Learning From Implicit User Feedback to Improve Natural Language Understanding in Large-Scale Conversational AI Systems

EMNLP 2021main

Natural Language Understanding (NLU) is an established component within a conversational AI or digital assistant system, and it is responsible for producing semantic understanding of a user request. We propose a scalable and automatic approach for improving NLU in a large-scale conversational AI sys…

Cited by 21SourcePDFScholar