Lili Mou

Stochastic Wasserstein Autoencoder for Probabilistic Sentence Generation

2018 · arXiv (Cornell University)

The variational autoencoder (VAE) imposes a probabilistic distribution (typically Gaussian) on the latent space and penalizes the Kullback--Leibler (KL) divergence between the posterior and prior. In NLP, VAEs are extremely difficult to train due to …

Towards Neural Speaker Modeling in Multi-Party Conversation: The Task, Dataset, and Models

2018

Neural network-based dialog systems are attracting increasing attention in both academia and industry.Recently, researchers have begun to realize the importance of speaker modeling in neural dialog systems, but there lacks established tasks and datasets.In this …

Knowledge Distillation for Language Models

2025

Yuqiao Wen, Freda Shi, Lili Mou. Proceedings of the 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 5: Tutorial Abstracts). 2025.

Classifying Relations via Long Short Term Memory Networks along Shortest Dependency Path

2015 · arXiv (Cornell University)

Relation classification is an important research arena in the field of natural language processing (NLP). In this paper, we present SDP-LSTM, a novel neural network to classify the relation of two entities in a sentence. …

RUBER: An Unsupervised Method for Automatic Evaluation of Open-Domain Dialog Systems

2018 · Proceedings of the AAAI Conference on Artificial Intelligence

Open-domain human-computer conversation has been attracting increasing attention over the past few years. However, there does not exist a standard automatic evaluation metric for open-domain dialog systems; researchers usually resort to human annotation for model …

How to Make Context More Useful? An Empirical Study on Context-Aware Neural Conversational Models

2017

Generative conversational systems are attracting increasing attention in natural language processing (NLP). Recently, researchers have noticed the importance of context information in dialog processing, and built various models to utilize context. However, there is no …

Disentangled Representation Learning for Non-Parallel Text Style Transfer

2019

This paper tackles the problem of disentangling the latent representations of style and content in language models. We propose a simple yet effective approach, which incorporates auxiliary multi-task and adversarial objectives, for style prediction and …

Distilling Task-Specific Knowledge from BERT into Simple Neural Networks

2019 · arXiv (Cornell University)

In the natural language processing literature, neural networks are becoming increasingly deeper and complex. The recent poster child of this trend is the deep language representation model, which includes BERT, ELMo, and GPT. These developments …

Natural Language Inference by Tree-Based Convolution and Heuristic Matching

2016

In this paper, we propose the TBCNNpair model to recognize entailment and contradiction between two sentences. In our model, a tree-based convolutional neural network (TBCNN) captures sentencelevel semantics; then heuristic matching layers like concatenation, element-wise …

Generating Sentences from Disentangled Syntactic and Semantic Spaces

2019

Variational auto-encoders (VAEs) are widely used in natural language generation due to the regularization of the latent space. However, generating sentences from the continuous latent space does not explicitly model the syntactic information. In this …

Sequence to Backward and Forward Sequences: A Content-Introducing\n Approach to Generative Short-Text Conversation

2016 · arXiv (Cornell University)

Using neural networks to generate replies in human-computer dialogue systems\nis attracting increasing attention over the past few years. However, the\nperformance is not satisfactory: the neural network tends to generate safe,\nuniversally relevant replies which carry little …

CGMH: Constrained Sentence Generation by Metropolis-Hastings Sampling

2019 · Proceedings of the AAAI Conference on Artificial Intelligence

In real-world applications of natural language generation, there are often constraints on the target sentences in addition to fluency and naturalness requirements. Existing language generation techniques are usually based on recurrent neural networks (RNNs). However, …

Unsupervised Paraphrasing by Simulated Annealing

2020

We propose UPSA, a novel approach that accomplishes Unsupervised Paraphrasing by Simulated Annealing. We model paraphrase generation as an optimization problem and propose a sophisticated objective function, involving semantic similarity, expression diversity, and language fluency …