Lei Li

Career Trajectory Analysis of Information Technology Alumni

2016

Understanding alumni's career path is critical for an Information Technology (IT) program to engage existing students, get connected to alumni and adapt the curriculum to the ever-changing technology field. LinkedIn, the largest professional networking site, …

Research on risk assessment method of subway operation based on cloud model

2016

In view of the deficiency of the existing risk assessment method for subway operation, the new risk assessment method based on cloud model is studied. Firstly, risk assessment indicator system of the subway operation is …

Analysis of the Survey and Solve Path to "Telecom and Network Fraud" in Colleges and Universities

2017 · DEStech Transactions on Social Science Education and Human Science

Through the questionnaire survey on the "students' understanding of "telecom network fraud" to the students in colleges and universities, we More accurately grasp the college students in case of "telecommunications network fraud", college students' attitude …

Quantum anomaly detection with density estimation and multivariate Gaussian distribution

2019 · Physical Review A

We study quantum anomaly detection with density estimation and multivariate Gaussian distribution. Both algorithms are constructed using the standard gate-based model of quantum computing. Compared with the corresponding classical algorithms, the resource complexities of our …

CFO: Conditional Focused Neural Question Answering with Large-scale Knowledge Bases

2016

How can we enable computers to automatically answer questions like "Who created the character Harry Potter"? Carefully built knowledge bases provide rich sources of facts. However, it remains a challenge to answer factoid questions raised …

On the convergence of gradient descent for two layer neural networks

2019 · arXiv (Cornell University)

It has been shown that gradient descent can yield the zero training loss in the over-parametrized regime (the width of the neural networks is much larger than the number of data points). In this work, …

Adaptive Gradient Methods Can Be Provably Faster than SGD after Finite Epochs

2020 · arXiv (Cornell University)

Adaptive gradient methods have attracted much attention of machine learning communities due to the high efficiency. However their acceleration effect in practice, especially in neural network training, is hard to analyze, theoretically. The huge gap …

On the Sentence Embeddings from Pre-trained Language Models

2020 · arXiv (Cornell University)

Pre-trained contextual representations like BERT have achieved great success in natural language processing. However, the sentence embeddings from the pre-trained language models without fine-tuning have been found to poorly capture semantic meaning of sentences. In …

Taxonomy Completion via Triplet Matching Network

2021 · arXiv (Cornell University)

Automatically constructing taxonomy finds many applications in e-commerce and web search. One critical challenge is as data and business scope grow in real applications, new concepts are emerging and needed to be added to the …

"Listen, Understand and Translate": Triple Supervision Decouples End-to-end Speech-to-text Translation

2020 · arXiv (Cornell University)

An end-to-end speech-to-text translation (ST) takes audio in a source language and outputs the text in a target language. Existing methods are limited by the amount of parallel corpus. Can we build a system to …

Secoco: Self-Correcting Encoding for Neural Machine Translation

2021 · arXiv (Cornell University)

This paper presents Self-correcting Encoding (Secoco), a framework that effectively deals with input noise for robust neural machine translation by introducing self-correcting predictors. Different from previous robust approaches, Secoco enables NMT to explicitly correct noisy …

E-KAR: A Benchmark for Rationalizing Natural Language Analogical Reasoning

2022 · Findings of the Association for Computational Linguistics: ACL 2022

The ability to recognize analogies is fundamental to human cognition. Existing benchmarks to test word analogy do not reveal the underneath process of analogical reasoning of neural models. Holding the belief that models capable of …

Converge to the Truth: Factual Error Correction via Iterative Constrained Editing

2023 · Proceedings of the AAAI Conference on Artificial Intelligence

Given a possibly false claim sentence, how can we automatically correct it with minimal editing? Existing methods either require a large number of pairs of false and corrected claims for supervised training or do not …

Communication Efficient Federated Learning for Multilingual Neural Machine Translation with Adapter

2023

Federated Multilingual Neural Machine Translation (Fed-MNMT) has emerged as a promising paradigm for institutions with limited language resources. This approach allows multiple institutions to act as clients and train a unified model through model synchronization, …

Progression Cognition Reinforcement Learning With Prioritized Experience for Multi-Vehicle Pursuit

2024 · IEEE Transactions on Intelligent Transportation Systems

Multi-vehicle pursuit (MVP) such as autonomous police vehicles pursuing suspects is important but very challenging due to its mission and safety-critical nature. While multi-agent reinforcement learning (MARL) algorithms have been proposed for MVP in structured …

EasyInstruct: An Easy-to-use Instruction Processing Framework for Large Language Models

2024 · arXiv (Cornell University)

In recent years, instruction tuning has gained increasing attention and emerged as a crucial technique to enhance the capabilities of Large Language Models (LLMs). To construct high-quality instruction datasets, many instruction processing approaches have been …

DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models

2024 · arXiv (Cornell University)

The visual projector, which bridges the vision and language modalities and facilitates cross-modal alignment, serves as a crucial component in MLLMs. However, measuring the effectiveness of projectors in vision-language alignment remains under-explored, which currently can …

Improving Factual Error Correction for Abstractive Summarization via Data Distillation and Conditional-generation Cloze

2024 · arXiv (Cornell University)

Improving factual consistency in abstractive summarization has been a focus of current research. One promising approach is the post-editing method. However, previous works have yet to make sufficient use of factual factors in summaries and …

Improving Factual Error Correction for Abstractive Summarization via Data Distillation and Conditional-Generation Cloze

2025 · IEEE Transactions on Audio Speech and Language Processing

Improving factual consistency in abstractive summarization has been a focus of recent research. One promising approach is the post-editing method. However, previous works have yet to make sufficient use of factual factors in summaries and …

CLGB-Net: fusion network for identifying local and global information of lesions in digital mammography images

2025 · Frontiers in Oncology

Worldwide, breast cancer ranks among the cancers with the highest incidence rate. Early diagnosis is crucial to improve the survival rate of patients. Digital Mammography (DM) is widely used for breast cancer diagnosis. The disadvantage …

CMU’s IWSLT 2025 Simultaneous Speech Translation System

2025

This paper presents CMU's submission to the IWSLT 2025 Simultaneous Speech Translation (SST) task for translating unsegmented English speech into Chinese and German text in a streaming manner.Our end-to-end speechto-text system integrates a chunkwise causal …

Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis

2025

In the quest for artificial general intelligence, Multi-modal Large Language Models (MLLMs) have emerged as a focal point in recent advancements. However, the predominant focus remains on developing their capabilities in static image understanding. The …

QiMeng-TensorOp: One-Line Prompt is Enough for High-Performance Tensor Operator Generation with Hardware Primitives

2025

Computation-intensive tensor operators constitute over 90% of the computations in Large Language Models (LLMs) and Deep Neural Networks. Automatically and efficiently generating high-performance tensor operators with hardware primitives is crucial for diverse and ever-evolving hardware …

LONGCODEU: Benchmarking Long-Context Language Models on Long Code Understanding

2025 · arXiv (Cornell University)

Current advanced long-context language models offer great potential for real-world software engineering applications. However, progress in this critical domain remains hampered by a fundamental limitation: the absence of a rigorous evaluation framework for long code …

Generating Fluent Adversarial Examples for Natural Languages

2019

Efficiently building an adversarial attacker for natural language processing (NLP) tasks is a real challenge. Firstly, as the sentence space is discrete, it is difficult to make small perturbations along the direction of gradients. Secondly, …

Dynamically Fused Graph Network for Multi-hop Reasoning

2019

Text-based question answering (TBQA) has been studied extensively in recent years. Most existing approaches focus on finding the answer to a question within a single paragraph. However, many difficult questions require multiple supporting evidence from …

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 …

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, …

Generate Neural Template Explanations for Recommendation

2020

Personalized recommender systems are important to assist user decision-making in the era of information overload. Meanwhile, explanations of the recommendations further help users to better understand the recommended items so as to make informed choices, …

Double Graph Based Reasoning for Document-level Relation Extraction

2020

Document-level relation extraction aims to extract relations among entities within a document. Different from sentence-level relation extraction, it requires reasoning over multiple sentences across paragraphs. In this paper, we propose Graph Aggregation-and-Inference Network (GAIN), a …

Contrastive Learning for Many-to-many Multilingual Neural Machine Translation

2021

Xiao Pan, Mingxuan Wang, Liwei Wu, Lei Li. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

A Survey on In-context Learning

2022 · arXiv (Cornell University)

With the increasing capabilities of large language models (LLMs), in-context learning (ICL) has emerged as a new paradigm for natural language processing (NLP), where LLMs make predictions based on contexts augmented with a few examples. …