Lin Li

MARS: A multi-aspect Recommender system for Point-of-Interest

2015

With the pervasive use of GPS-enabled smart phones, location-based services, e.g., Location Based Social Networking (LBSN) have emerged . Point-of-Interests (POIs) Recommendation, as a typical component in LBSN, provides additional values to both customers and …

A New Approach to the Definition of Information Diffusion and Confusion of Boolean Transformation

2016

Mass, energy and information are the three important quantities in the physical world. The diffusion and confusion of information is the guiding ideology of cipher algorithm coding. Linear transformation is generally used to realize the …

An Improved Algorithm For Conceptual Semantic Similarity In Domain Ontology

2017

With the extensive application of ontology in many fields, the semantic similarity computation based on domain ontology has been a hot research topic. At present, taking advantage of the upper and lower layer structure of …

The design and implementation for automatic evaluation system of virtual experiment report

2017

This paper designs and implements an automatic evaluation system for experimental reports in the field of university computer virtual experiment. The evaluation type is divided into three types: the only answer type, the rule-related type …

When Speaker Recognition Meets Noisy Labels: Optimizations for Front-ends and Back-ends

2021

A typical speaker recognition system often involves two modules: a feature extractor front-end and a speaker identity back-end. Despite the superior performance that deep neural networks have achieved for the front-end, their success benefits from …

Entanglement filter with Rydberg atoms

2022 · arXiv (Cornell University)

Devices capable of deterministically manipulating the photonic entanglement are of paramount importance, since photons are the ideal messengers for quantum information. Here, we report a Rydberg-atom-based entanglement filter that preserves the desired photonic entangled state …

Unifying and Personalizing Weakly-supervised Federated Medical Image Segmentation via Adaptive Representation and Aggregation

2023 · arXiv (Cornell University)

Federated learning (FL) enables multiple sites to collaboratively train powerful deep models without compromising data privacy and security. The statistical heterogeneity (e.g., non-IID data and domain shifts) is a primary obstacle in FL, impairing the …

Sinc‐attention feature extraction for trivial‐event based speaker verification

2023 · Electronics Letters

Abstract Human speech contains trivial events that are non‐subjectively controlled and are limited by physical characteristics. These events are represented by a series of extremely short utterances, such as the ‘hmm’ phrase used for expressing …

Electro-optically programmable photonic circuits enabled by wafer-scale integration on thin-film lithium niobate

2023 · Physical Review Research

Programmable photonic circuits performing universal linear-optical transformations underpin vital functions in photonic quantum information processing, quantum-enhanced sensor networks, machine learning, and many other intriguing applications. Recent advances in photonic integrated circuits facilitate monolithic integration of …

OODRobustBench: a Benchmark and Large-Scale Analysis of Adversarial Robustness under Distribution Shift

2023 · arXiv (Cornell University)

Existing works have made great progress in improving adversarial robustness, but typically test their method only on data from the same distribution as the training data, i.e. in-distribution (ID) testing. As a result, it is …

ASLseg: Adapting SAM in the Loop for Semi-supervised Liver Tumor Segmentation

2023 · arXiv (Cornell University)

Liver tumor segmentation is essential for computer-aided diagnosis, surgical planning, and prognosis evaluation. However, obtaining and maintaining a large-scale dataset with dense annotations is challenging. Semi-Supervised Learning (SSL) is a common technique to address these …

One Prompt Word is Enough to Boost Adversarial Robustness for Pre-trained Vision-Language Models

2024 · arXiv (Cornell University)

Large pre-trained Vision-Language Models (VLMs) like CLIP, despite having remarkable generalization ability, are highly vulnerable to adversarial examples. This work studies the adversarial robustness of VLMs from the novel perspective of the text prompt instead …

Multivariate Fourier Distribution Perturbation: Domain Shifts with Uncertainty in Frequency Domain

2024

Diversifying training data techniques have achieved tremendous success in Domain Generalization (DG) tasks. The key to diversifying domain data is by increasing the types of domain styles. After investigating this issue from the perspective of …

Clinlabomics‐Enabled Blending Ensemble Learning for Low‐Cost Pan‐Cancer Detection and Classification Using Routine Clinical Laboratory Data

2025 · Advanced Intelligent Systems

The early detection and accurate diagnosis of cancer is of great significance for cancer patients. Artificial intelligences rapid advancement has boosted clinlabomics, bridging clinical lab data with intelligent systems. Inspired by clinlabomics, a liquid biopsy …

Specific Emitter Identification for Open-Set Satellite TT${\&}$C Signals Based on Multi-Task Learning

2025 · IEEE Transactions on Vehicular Technology

Specific emitter identification (SEI) has attracted wide attention in the field of radar and communication counter-measures. Different from traditional SEI techniques, identifying satellite individual suffers from composite modulation (CM) of telemetry, tracking, and command (TT&C) …