Zhijian Ou

Learning Sparse Structured Ensembles with SG-MCMC and Network Pruning

2018 · arXiv (Cornell University)

An ensemble of neural networks is known to be more robust and accurate than an individual network, however usually with linearly-increased cost in both training and testing. In this work, we propose a two-stage method …

Upgrading CRFS to JRFS and its Benefits to Sequence Modeling and Labeling

2020

Two important sequence tasks are sequence modeling and labeling. Sequence modeling involves determining the probabilities of sequences, e.g. language modeling. It is still difficult to improve language modeling with additional relevant tags, e.g. part-of-speech (POS) …

Advancing Semi-Supervised Task Oriented Dialog Systems by JSA Learning of Discrete Latent Variable Models

2022 · arXiv (Cornell University)

Developing semi-supervised task-oriented dialog (TOD) systems by leveraging unlabeled dialog data has attracted increasing interests. For semi-supervised learning of latent state TOD models, variational learning is often used, but suffers from the annoying high-variance of …

Jointly Reinforced User Simulator and Task-oriented Dialog System with Simplified Generative Architecture

2022 · arXiv (Cornell University)

Recently, there has been progress in supervised funetuning pretrained GPT-2 to build end-to-end task-oriented dialog (TOD) systems. However, online reinforcement learning of a GPT-2 based dialog system (DS), together with a end-to-end user simulator (US), …

A Generative User Simulator with GPT-based Architecture and Goal State Tracking for Reinforced Multi-Domain Dialog Systems

2022 · arXiv (Cornell University)

Building user simulators (USs) for reinforcement learning (RL) of task-oriented dialog systems (DSs) has gained more and more attention, which, however, still faces several fundamental challenges. First, it is unclear whether we can leverage pretrained …

Energy-Based Models with Applications to Speech and Language Processing

2024 · Foundations and Trends® in Signal Processing

Energy-Based Models (EBMs) are an important class of probabilistic models, also known as random fields and undirected graphical models. EBMs are un-normalized and thus radically different from other popular self-normalized probabilistic models such as hidden …

Entriever: Energy-based Retriever for Knowledge-Grounded Dialog Systems

2025

A retriever, which retrieves relevant knowledge pieces from a knowledge base given a context, is an important component in many natural language processing (NLP) tasks.Retrievers have been introduced in knowledge-grounded dialog systems to improve knowledge …

Pronunciation-Lexicon Free Training for Phoneme-based Crosslingual ASR via Joint Stochastic Approximation

2025 · arXiv (Cornell University)

Recently, pre-trained models with phonetic supervision have demonstrated their advantages for crosslingual speech recognition in data efficiency and information sharing across languages. However, a limitation is that a pronunciation lexicon is needed for such phoneme-based …