arXiv subjects

Cheng Qiu

Publications and source records attributed to Cheng Qiu.

5 recordsLinked to original sources

arXiv2024

Leaving Some Facial Features Behind

Facial expressions are crucial to human communication, offering insights into emotional states. This study examines how specific facial features influence emotion classification, using facial perturbations on the Fer2013 dataset. As expected, models trained on data with the removal of some important facial feature experienced up to an 85% accuracy drop when compared to baseline for emotions like happy and surprise. Surprisingly, for the emotion disgust, there seem to be slight improvement in accuracy for classifier after mask have been applied. Building on top of this observation, we applied a training scheme to mask out facial features during training, motivating our proposed Perturb Scheme. This scheme, with three phases-attention-based classification, pixel clustering, and feature-focused training, demonstrates improvements in classification accuracy. The experimental results obtained suggests there are some benefits to removing individual facial features in emotion recognition tasks.

cs.CV↗

arXiv2022

Paraphrase Identification with Deep Learning: A Review of Datasets and Methods

The rapid progress of Natural Language Processing (NLP) technologies has led to the widespread availability and effectiveness of text generation tools such as ChatGPT and Claude. While highly useful, these technologies also pose significant risks to the credibility of various media forms if they are employed for paraphrased plagiarism -- one of the most subtle forms of content misuse in scientific literature and general text media. Although automated methods for paraphrase identification have been developed, detecting this type of plagiarism remains challenging due to the inconsistent nature of the datasets used to train these methods. In this article, we examine traditional and contemporary approaches to paraphrase identification, investigating how the under-representation of certain paraphrase types in popular datasets, including those used to train Large Language Models (LLMs), affects the ability to detect plagiarism. We introduce and validate a new refined typology for paraphrases (ReParaphrased, REfined PARAPHRASE typology definitions) to better understand the disparities in paraphrase type representation. Lastly, we propose new directions for future research and dataset development to enhance AI-based paraphrase detection.

cs.CL↗

arXiv2021

An even-load-distribution design for composite bolted joints using a novel circuit model and artificial neural networks

Due to the brittle feature of carbon fiber reinforced plastic laminates, mechanical multi-joint within these composite components show uneven load distribution for each bolt, which weaken the strength advantage of composite laminates. In order to reduce this defect and achieve the goal of even load distribution in mechanical joints, we propose a machine learning-based framework as an optimization method. Since that the friction effect has been proven to be a significant factor in determining bolt load distribution, our framework aims at providing optimal parameters including bolt-hole clearances and tightening torques for a minimum unevenness of bolt load. A novel circuit model is established to generate data samples for the training of artificial networks at a relatively low computational cost. A database for all the possible inputs in the design space is built through the machine learning model. The optimal dataset of clearances and torques provided by the database is validated by both the finite element method, circuit model, and an experimental measurement based on the linear superposition principle, which shows the effectiveness of this general framework for the optimization problem. Then, our machine learning model is further compared and worked in collaboration with commonly used optimization algorithms, which shows the potential of greatly increasing computational efficiency for the inverse design problem.

cs.LG↗

arXiv2016

Atom-light superposition oscillation and Ramsey-like atom-light interferometer

Coherent wave splitting is crucial in interferometers. Normally, the waves after this splitting are of the same type. But recent progress in interaction between atom and light has led to the coherent conversion of photon to atomic excitation. This makes it possible to split an incoming light wave into a coherent superposition state of atom and light and paves the way for an interferometer made of different types of waves. Here we report on a Rabi-like coherent-superposition oscillation observed between atom and light and a coherent mixing of light wave with excited atomic spin wave in a Raman process. We construct a new kind of hybrid interferometer based on the atom-light coherent superposition state. Interference fringes are observed in both optical output intensity and atomic output in terms of the atomic spin wave strength when we scan either or both of the optical and atomic phases. Such a hybrid interferometer can be used to interrogate atomic states by optical detection and will find its applications in precision measurement and quantum control of atoms and light.

quant-ph↗

arXiv2012

Efficient Raman Frequency Conversion by Feedbacks of Pump and Stokes Fields

We experimentally demonstrate efficient Raman conversion to respective Stokes and anti-Stokes fields in both pulsed and continuous modes with a Rb-87 atomic vapor cell. The conversion efficiency is about 40-50% for the Stokes field and 20-30% for the anti-Stokes field, respectively. This conversion process is realized with feedback of both the Raman pump and the frequency-converted fields (Stokes or anti-Stokes). The experimental setup is very simple and can be applied easily to produce the light source with larger frequency difference using other Raman media. They may have wide applications in nonlinear optics, atomic physics, quantum optics and precise measurement.

physics.atom-ph↗