Publications (13)
Exposing the Cracks: Vulnerabilities of Retrieval-Augmented LLM-based Machine Translation
Yanming Sun, Runzhe Zhan, Chi Seng Cheang +7
\textbf{RE}trieval-\textbf{A}ugmented \textbf{L}LM-based \textbf{M}achine \textbf{T}ranslation (REAL-MT) shows promise for knowledge-intensive tasks like idiomatic translation, but…
Manipulating reflection-type all-dielectric non-local metasurfaces via parity of particle number
Hao Song, Xuelian Zhang, Yanming Sun +1
Parity of particle number is a new degree of freedom for manipulating metasurface, while its influence on controlling non-local metasurfaces remains an unresolved and intriguing qu…
Substrate-engineered tunable bound states in the continuum and directional radiation in dielectric metasurfaces
Hao Song, Yanming Sun, Jian Li +2
Tunable bound states in the continuum (BICs) in metasurfaces offer powerful opportunities to control light-matter interactions, yet the role of out-of-plane symmetry breaking remai…
Enumerating Chemical Graphs with Mono-block 2-Augmented Tree Structure from Given Upper and Lower Bounds on Path Frequencies
Yuui Tamura, Yuhei Nishiyama, Chenxi Wang +4
We consider a problem of enumerating chemical graphs from given constraints concerning their structures, which has an important application to a novel method for the inverse QSAR/Q…
A Novel Method for Inference of Acyclic Chemical Compounds with Bounded Branch-height Based on Artificial Neural Networks and Integer Programming
Naveed Ahmed Azam, Jianshen Zhu, Yanming Sun +5
Analysis of chemical graphs is a major research topic in computational molecular biology due to its potential applications to drug design. One approach is inverse quantitative stru…
A Novel Thiophene-fused Ending Group Enabling an Excellent Small Molecule Acceptor for High-performance Fullerene-free Polymer Solar Cells with 11.8% Efficiency
Dongjun Xie, Tao Liu, Wei Gao +8
In this article, we designed and synthesized a novel small molecule acceptor of ITCPTC with thiophene-fused ending group by employing a new active methylene precursor of CPTCN. The…
Rotation-tuned single hexagonal air cavity assisting in third-harmonic generation via hybrid modes
Hao Song, Junmin Deng, Yu Chen +3
A fillable air cavity with a high quality (Q) factor and large-scale electric field confinement is highly desired in many optical applications. Yet, it remains challenging due to t…
Can LMs Generalize to Future Data? An Empirical Analysis on Text Summarization
Chi Seng Cheang, Hou Pong Chan, Derek F. Wong +5
Recent pre-trained language models (PLMs) achieve promising results in existing abstractive summarization datasets. However, existing summarization benchmarks overlap in time with…
G-IdiomAlign: A Gloss-Pivoted Benchmark for Cross-Lingual Idiom Alignment
Fengying Ye, Yanming Sun, Runzhe Zhan +3
Idioms are difficult to transfer across languages due to their non-compositionality and weak surface-form grounding, making literal mappings unreliable. We present G-IdiomAlign, a…
Beyond Simple Fusion: Adaptive Gated Fusion for Robust Multimodal Sentiment Analysis
Han Wu, Yanming Sun, Yunhe Yang +1
Multimodal sentiment analysis (MSA) leverages information fusion from diverse modalities (e.g., text, audio, visual) to enhance sentiment prediction. However, simple fusion techniq…
Brain Tumor Detection Based on a Novel and High-Quality Prediction of the Tumor Pixel Distributions
Yanming Sun, Chunyan Wang
In this paper, we propose a system to detect brain tumor in 3D MRI brain scans of Flair modality. It performs 2 functions: (a) predicting gray-level and locational distributions of…
Bound state in the continuum and polarization-insensitive electric mirror in low-contrast metasurface
Hao Song, Xuelian Zhang, Jian Wang +2
High-contrast refractive indexes are pivotal in dielectric metasurfaces for inducing various exotic phenomena, such as the bound state in the continuum (BIC) and electric mirror (E…
A Computation-Efficient CNN System for High-Quality Brain Tumor Segmentation
Yanming Sun, Chunyan Wang
The work presented in this paper is to propose a reliable high-quality system of Convolutional Neural Network (CNN) for brain tumor segmentation with a low computation requirement.…