Chao Zhang

Learning to Detect Anomalies in Surveillance Video

2015 · IEEE Signal Processing Letters

Detecting anomalies in surveillance videos, that is, finding events or objects with low probability of occurrence, is a practical and challenging research topic in computer vision community. In this paper, we put forward a novel …

Study on Modeling of Local Intelligent Traffic Scheduling

2017

In order to alleviate the congestion of urban traffic roads, reduce the tension of congestion and reduce travel time, through traffic simulation, we can predict and evaluate the actual traffic situation, so as to layout …

BOND: BERT-Assisted Open-Domain Named Entity Recognition with Distant Supervision

2020

We study the open-domain named entity recognition (NER) problem under distant supervision. The distant supervision, though does not require large amounts of manual annotations, yields highly incomplete and noisy distant labels via external knowledge bases. …

Trade or Trick?

2022

The prosperity of the cryptocurrency ecosystem drives the need for digital asset trading platforms. Uniswap, as the most prominent cryptocurrency decentralized exchange (DEX), is continuing to attract scammers, with fraudulent cryptocurrencies flooding in the ecosystem. …

Automatic Generation of Adversarial Readable Chinese Texts

2022 · IEEE Transactions on Dependable and Secure Computing

Natural language processing (NLP) models are known vulnerable to adversarial examples, similar to image processing models. Studying adversarial texts is an essential step to improve the robustness of NLP models. However, existing studies mainly focus …

Experimental beating the standard quantum limit under non-markovian dephasing environment

2022 · arXiv (Cornell University)

Entanglement enhanced quantum metrology has been well investigated for beating the standard quantum limit (SQL). However, the metrological advantage of entangled states becomes much more elusive in the presence of noise. Under strictly Markovian dephasing …

Neighborhood-Regularized Self-Training for Learning with Few Labels

2023 · arXiv (Cornell University)

Training deep neural networks (DNNs) with limited supervision has been a popular research topic as it can significantly alleviate the annotation burden. Self-training has been successfully applied in semi-supervised learning tasks, but one drawback of …

Video Surveillance for Indoor Office Environment Based on Object-Level Anomaly Detection

2023 · Journal of Physics Conference Series

Abstract Traditional methods of Abnormal Behavior Detection (ABD) process the surveillance video on frame-level, which ignores object-level abnormal behavior patterns. To address the problem, this paper presents Object-Level Anomaly Detection model (OLAD), which aims to …

Tapping into Permutation Symmetry for Improved Detection of k-Symmetric Extensions

2023 · Entropy

Symmetric extensions are essential in quantum mechanics, providing a lens through which to investigate the correlations of entangled quantum systems and to address challenges like the quantum marginal problem. Though semi-definite programming (SDP) is a …

How Far Have We Gone in Vulnerability Detection Using Large Language Models

2023 · arXiv (Cornell University)

As software becomes increasingly complex and prone to vulnerabilities, automated vulnerability detection is critically important, yet challenging. Given the significant successes of large language models (LLMs) in various tasks, there is growing anticipation of their …

Competitive-Driven Learning for Image Ordinal Classification

2023

Image ordinal classification (IOC) assigns a discrete yet ordinal scalar label to an input image, such as age estimation. For IOC, it is non-trivial to effectively incorporate the ordinal information of inputs into the classification …

Bayesian Example Selection Improves In-Context Learning for Speech, Text, and Visual Modalities

2024 · arXiv (Cornell University)

Large language models (LLMs) can adapt to new tasks through in-context learning (ICL) based on a few examples presented in dialogue history without any model parameter update. Despite such convenience, the performance of ICL heavily …

TextToucher: Fine-Grained Text-to-Touch Generation

2024 · arXiv (Cornell University)

Tactile sensation plays a crucial role in the development of multi-modal large models and embodied intelligence. To collect tactile data with minimal cost as possible, a series of studies have attempted to generate tactile images …

LoRC: Low-Rank Compression for LLMs KV Cache with a Progressive Compression Strategy

2024 · arXiv (Cornell University)

The Key-Value (KV) cache is a crucial component in serving transformer-based autoregressive large language models (LLMs), enabling faster inference by storing previously computed KV vectors. However, its memory consumption scales linearly with sequence length and …

Speaker Adaptation for Quantised End-to-End ASR Models

2024 · arXiv (Cornell University)

End-to-end models have shown superior performance for automatic speech recognition (ASR). However, such models are often very large in size and thus challenging to deploy on resource-constrained edge devices. While quantisation can reduce model sizes, …

Simplified Multiplayer Battle Game-inspired Optimizer with Diverse Search Strategies

2024

We propose two modifications to the standard multiplayer battle game-inspired optimizer (MBGO) to simplify its search framework and enhance its performance. Specifically, the first modification changes the original serial two-stage search to a parallel approach, …

Weakly-Supervised Neural Text Classification

2018

Deep neural networks are gaining increasing popularity for the classic text classification task, due to their strong expressive power and less requirement for feature engineering. Despite such attractiveness, neural text classification models suffer from the …

Spherical Text Embedding

2019 · arXiv (Cornell University)

Unsupervised text embedding has shown great power in a wide range of NLP tasks. While text embeddings are typically learned in the Euclidean space, directional similarity is often more effective in tasks such as word …

PanGu-$α$: Large-scale Autoregressive Pretrained Chinese Language Models with Auto-parallel Computation

2021 · arXiv (Cornell University)

Large-scale Pretrained Language Models (PLMs) have become the new paradigm for Natural Language Processing (NLP). PLMs with hundreds of billions parameters such as GPT-3 have demonstrated strong performances on natural language understanding and generation with …

Large language models for generative information extraction: a survey

2024 · Frontiers of Computer Science

Abstract Information Extraction (IE) aims to extract structural knowledge from plain natural language texts. Recently, generative Large Language Models (LLMs) have demonstrated remarkable capabilities in text understanding and generation. As a result, numerous works have …