arXiv subjects

Jin Yu

Publications and source records attributed to Jin Yu.

At least 19 recordsLinked to original sources

arXiv2026

Don't Contrast the Impossible: Region-Constrained Batching for Contrastive User Modeling on a Local Community Platform

Contrastive learning is widely used for user modeling in large-scale recommender systems, where standard in-batch negatives implicitly assume universal exposure that any user can be shown any item. On local community platforms such as Karrot, however, exposure is geographically constrained; many user-item pairs are impossible by design yet still treated as negatives during training, diluting the contrastive learning signal. We address this impossible negatives problem and propose Region-Constrained Batch Sampling (RCBS), a simple yet effective batching method that constructs region-homogeneous mini-batches so that users are contrasted primarily against items they could feasibly see. By replacing impossible negatives with feasible ones, RCBS naturally introduces harder and more informative negatives under realistic exposure constraints. With offline evaluations and online A/B tests, we show that RCBS consistently improves user representation quality and consequently enhances home feed ranking, retrieval, and display ads ranking. The resulting user embeddings have been deployed in production across various applications.

cs.IR↗

arXiv2026

Norm or Direction? Decoding Vision Mambas for High-Resolution Vision

Vision Mamba models replace quadratic self-attention with linear complexity selective state space models (SSMs), emerging as efficient visual backbones. However, MambaOut demonstrates that a Gated CNN block can match or exceed VMamba on image classification, questioning the necessity of SSMs for vision. This raises a fundamental question: do VMamba and MambaOut encode visual information differently at the representation level? To investigate, we apply cross model centered kernel alignment (CKA) analysis and find that VMamba's final stage blocks form representations distinctly different from both MambaOut and its own preceding blocks. We therefore focus on the final block features, decomposing each spatial token into magnitude and direction. MambaOut concentrates class-discriminative information in high-norm foreground tokens that align with Grad-CAM attribution. VMamba, by contrast, produces high-norm tokens predominantly in background regions, misaligned with Grad-CAM, yet preserves discriminative signals primarily in token directions. These observations reveal that the two models rely on different encoding strategies. We connect this difference to high-resolution classification and semantic segmentation. VMamba distributes logit support broadly across object regions, whereas MambaOut relies on sparse dominant tokens, a strategy that becomes less stable as token counts grow. Under full fine-tuning for segmentation, VMamba consistently outperforms MambaOut. These results suggest that VMamba's advantage in dense prediction stems not merely from the SSM mechanism or sequence length, but from how semantic evidence is organized across token magnitude, direction. Ultimately, we conclude that token magnitude and directional structure serve as critical axes for improving visual backbones, particularly under dense supervision.

cs.CV↗

arXiv2026

Tracking thermal transport in colloidal quantum dot films using in-situ time-resolved X-ray diffraction

Colloidal quantum dots (QDs) and their thin-films are increasingly used in electronic and photonic devices replacing traditional bulk semiconductors. However, thermal properties of the QDs are comparatively underexplored relative to device development efforts. This study shows the use of time-resolved X-ray diffraction as a contact-free method to probe the thermal response of QDs in device-like environments, providing in-situ insights for future thermal management strategies. Through the extraction of Debye-Waller Factors on a sub-nanosecond timescale, we use time-resolved X-ray diffraction to directly capture the heating and cooling of core/shell CdSe/CdS QDs following pulsed optical excitation. In a QD thin-film that actively provides optical gain, the thermal conductivity is found to be as low as 0.55 $\mathrm{W\,m^{-1}\,K^{-1}}$, because of the poor heat flow within close-packed QD solids. For QDs dispersed in liquids, interfacial thermal conductance is found to dominate the thermal relaxation with a conductance on the order of 15 $\mathrm{MW\,m^{-2}\,K^{-1}}$.

cond-mat.mes-hall↗

arXiv2025

Designing for Engaging Communication Between Parents and Young Adult Children Through Shared Music Experiences

This paper aims to foster social interaction between parents and young adult children living apart via music. Our approach transforms their music-listening moment into an opportunity to listen to the other's favorite songs and enrich interaction in their daily lives. To this end, we explore the current practice and needs of parent-child communication and the experience and perception of music-mediated interaction. Based on the findings, we developed DJ-Fam, a mobile application that enables parents and children to listen to their favorite songs and use them as conversation starters to foster parent-child interaction. From our deployment study with seven families over four weeks in South Korea, we show the potential of DJ-Fam to influence parent-child interaction and their mutual understanding and relationship positively. Specifically, DJ-Fam considerably increases the frequency of communication and diversifies the communication channels and topics, all of which are satisfactory to the participants.

cs.HC↗

arXiv2025

ALFRED: Ask a Large-language model For Reliable ECG Diagnosis

Leveraging Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) for analyzing medical data, particularly Electrocardiogram (ECG), offers high accuracy and convenience. However, generating reliable, evidence-based results in specialized fields like healthcare remains a challenge, as RAG alone may not suffice. We propose a Zero-shot ECG diagnosis framework based on RAG for ECG analysis that incorporates expert-curated knowledge to enhance diagnostic accuracy and explainability. Evaluation on the PTB-XL dataset demonstrates the framework's effectiveness, highlighting the value of structured domain expertise in automated ECG interpretation. Our framework is designed to support comprehensive ECG analysis, addressing diverse diagnostic needs with potential applications beyond the tested dataset.

cs.LG↗

arXiv2023

RGB-T Tracking Based on Mixed Attention

RGB-T tracking involves the use of images from both visible and thermal modalities. The primary objective is to adaptively leverage the relatively dominant modality in varying conditions to achieve more robust tracking compared to single-modality tracking. An RGB-T tracker based on mixed attention mechanism to achieve complementary fusion of modalities (referred to as MACFT) is proposed in this paper. In the feature extraction stage, we utilize different transformer backbone branches to extract specific and shared information from different modalities. By performing mixed attention operations in the backbone to enable information interaction and self-enhancement between the template and search images, it constructs a robust feature representation that better understands the high-level semantic features of the target. Then, in the feature fusion stage, a modality-adaptive fusion is achieved through a mixed attention-based modality fusion network, which suppresses the low-quality modality noise while enhancing the information of the dominant modality. Evaluation on multiple RGB-T public datasets demonstrates that our proposed tracker outperforms other RGB-T trackers on general evaluation metrics while also being able to adapt to longterm tracking scenarios.

cs.CV↗

arXiv2023

Automatic Detection of Reactions to Music via Earable Sensing

We present GrooveMeter, a novel system that automatically detects vocal and motion reactions to music via earable sensing and supports music engagement-aware applications. To this end, we use smart earbuds as sensing devices, which are already widely used for music listening, and devise reaction detection techniques by leveraging an inertial measurement unit (IMU) and a microphone on earbuds. To explore reactions in daily music-listening situations, we collect the first kind of dataset, MusicReactionSet, containing 926-minute-long IMU and audio data with 30 participants. With the dataset, we discover a set of unique challenges in detecting music listening reactions accurately and robustly using audio and motion sensing. We devise sophisticated processing pipelines to make reaction detection accurate and efficient. We present a comprehensive evaluation to examine the performance of reaction detection and system cost. It shows that GrooveMeter achieves the macro F1 scores of 0.89 for vocal reaction and 0.81 for motion reaction with leave-one-subject-out cross-validation. More importantly, GrooveMeter shows higher accuracy and robustness compared to alternative methods. We also show that our filtering approach reduces 50% or more of the energy overhead. Finally, we demonstrate the potential use cases through a case study.

cs.SD↗

arXiv2022

HQANN: Efficient and Robust Similarity Search for Hybrid Queries with Structured and Unstructured Constraints

The in-memory approximate nearest neighbor search (ANNS) algorithms have achieved great success for fast high-recall query processing, but are extremely inefficient when handling hybrid queries with unstructured (i.e., feature vectors) and structured (i.e., related attributes) constraints. In this paper, we present HQANN, a simple yet highly efficient hybrid query processing framework which can be easily embedded into existing proximity graph-based ANNS algorithms. We guarantee both low latency and high recall by leveraging navigation sense among attributes and fusing vector similarity search with attribute filtering. Experimental results on both public and in-house datasets demonstrate that HQANN is 10x faster than the state-of-the-art hybrid ANNS solutions to reach the same recall quality and its performance is hardly affected by the complexity of attributes. It can reach 99\% recall@10 in just around 50 microseconds On GLOVE-1.2M with thousands of attribute constraints.

cs.DB↗

arXiv2022

Vocalsound: A Dataset for Improving Human Vocal Sounds Recognition

Recognizing human non-speech vocalizations is an important task and has broad applications such as automatic sound transcription and health condition monitoring. However, existing datasets have a relatively small number of vocal sound samples or noisy labels. As a consequence, state-of-the-art audio event classification models may not perform well in detecting human vocal sounds. To support research on building robust and accurate vocal sound recognition, we have created a VocalSound dataset consisting of over 21,000 crowdsourced recordings of laughter, sighs, coughs, throat clearing, sneezes, and sniffs from 3,365 unique subjects. Experiments show that the vocal sound recognition performance of a model can be significantly improved by 41.9% by adding VocalSound dataset to an existing dataset as training material. In addition, different from previous datasets, the VocalSound dataset contains meta information such as speaker age, gender, native language, country, and health condition.

cs.SD↗

arXiv2021

Cross-domain User Preference Learning for Cold-start Recommendation

Cross-domain cold-start recommendation is an increasingly emerging issue for recommender systems. Existing works mainly focus on solving either cross-domain user recommendation or cold-start content recommendation. However, when a new domain evolves at its early stage, it has potential users similar to the source domain but with much fewer interactions. It is critical to learn a user's preference from the source domain and transfer it into the target domain, especially on the newly arriving contents with limited user feedback. To bridge this gap, we propose a self-trained Cross-dOmain User Preference LEarning (COUPLE) framework, targeting cold-start recommendation with various semantic tags, such as attributes of items or genres of videos. More specifically, we consider three levels of preferences, including user history, user content and user group to provide reliable recommendation. With user history represented by a domain-aware sequential model, a frequency encoder is applied to the underlying tags for user content preference learning. Then, a hierarchical memory tree with orthogonal node representation is proposed to further generalize user group preference across domains. The whole framework updates in a contrastive way with a First-In-First-Out (FIFO) queue to obtain more distinctive representations. Extensive experiments on two datasets demonstrate the efficiency of COUPLE in both user and content cold-start situations. By deploying an online A/B test for a week, we show that the Click-Through-Rate (CTR) of COUPLE is superior to other baselines used on Taobao APP. Now the method is serving online for the cross-domain cold micro-video recommendation.

cs.IR↗

arXiv2021

Stable Prediction on Graphs with Agnostic Distribution Shift

Graph is a flexible and effective tool to represent complex structures in practice and graph neural networks (GNNs) have been shown to be effective on various graph tasks with randomly separated training and testing data. In real applications, however, the distribution of training graph might be different from that of the test one (e.g., users' interactions on the user-item training graph and their actual preference on items, i.e., testing environment, are known to have inconsistencies in recommender systems). Moreover, the distribution of test data is always agnostic when GNNs are trained. Hence, we are facing the agnostic distribution shift between training and testing on graph learning, which would lead to unstable inference of traditional GNNs across different test environments. To address this problem, we propose a novel stable prediction framework for GNNs, which permits both locally and globally stable learning and prediction on graphs. In particular, since each node is partially represented by its neighbors in GNNs, we propose to capture the stable properties for each node (locally stable) by re-weighting the information propagation/aggregation processes. For global stability, we propose a stable regularizer that reduces the training losses on heterogeneous environments and thus warping the GNNs to generalize well. We conduct extensive experiments on several graph benchmarks and a noisy industrial recommendation dataset that is collected from 5 consecutive days during a product promotion festival. The results demonstrate that our method outperforms various SOTA GNNs for stable prediction on graphs with agnostic distribution shift, including shift caused by node labels and attributes.

cs.LG↗

arXiv2021

Double Resonant Tunable Second Harmonic Generation in Two-dimensional Layered Materials through Band Nesting

We proposed a mechanism to generate giant anisotropic second harmonic nonlinear response via double resonance effect, achieved through band nesting via electronic bandstructure engineering. The ideal band setup would be a triplet of nested bands separated by the fundamental resonance energy, $\hbar\omega$. We demonstrate theoretically that the proposed phenomenon can be realized in bilayer SnS by band tuning with perpendicular electrical bias, which maximizes the second harmonic susceptibility by several orders of magnitude. Moreover, the tunability of the polarization anisotropy can be useful for realizing novel polarization-sensitive devices.

physics.optics↗

arXiv2021

Distribution of ripples in graphene membrane

Intrinsic ripples with various configurations and sizes were reported to affect the physical and chemical properties of 2D materials. By performing molecular dynamics simulations and theoretical analysis, we use two geometric models of the ripple shape to explore numerically the distribution of ripples in graphene membrane. We focus on the ratio of ripple height to its diameter (t/D) which was recently shown to be the most relevant for chemical activity of graphene membranes. Our result demonstrates that the ripple density decreases as the coefficient t/D increases, in a qualitative agreement with the Boltzmann distribution derived analytically from the bending energy of the membrane. Our theoretical study provides also specific quantitative information on the ripple distribution in graphene and gives new insights applicable to other 2D materials.

cond-mat.mtrl-sci↗

arXiv2021

Two-Phase Dynamics of DNA Supercoiling based on DNA Polymer Physics

DNA supercoils are generated in genome regulation processes such as transcription and replication, and provide mechanical feedback to such processes. Under tension, DNA supercoil can present a coexistence state of plectonemic (P) and stretched (S) phases. Experiments have revealed the dynamic behaviors of plectoneme, e.g. diffusion, nucleation and hopping. To represent these dynamics with computational changes, we demonstrated first the fast dynamics on the DNA to reach torque equilibrium within the P and S phases, and then identified the two-phase boundaries as collective slow variables to describe the essential dynamics. According to the time scale separation demonstrated here, we accordingly developed a two-phase model on the dynamics of DNA supercoiling, which can capture physiologically relevant events across time scales of several orders of magnitudes. In this model, we systematically characterized the slow dynamics between the two phases, and compared the numerical results with that from the DNA polymer physics-based worm-like chain model. The supercoiling dynamics, including the nucleation, diffusion, and hopping of plectoneme, have been well represented and reproduced, using the two-phase dynamic model, at trivial computational costs. Our current developments, therefore, can be implemented to explore multi-scale physical mechanisms of the DNA supercoiling-dependent physiological processes.

cond-mat.soft↗

arXiv2021

Deep learning for detecting pulmonary tuberculosis via chest radiography: an international study across 10 countries

Tuberculosis (TB) is a top-10 cause of death worldwide. Though the WHO recommends chest radiographs (CXRs) for TB screening, the limited availability of CXR interpretation is a barrier. We trained a deep learning system (DLS) to detect active pulmonary TB using CXRs from 9 countries across Africa, Asia, and Europe, and utilized large-scale CXR pretraining, attention pooling, and noisy student semi-supervised learning. Evaluation was on (1) a combined test set spanning China, India, US, and Zambia, and (2) an independent mining population in South Africa. Given WHO targets of 90% sensitivity and 70% specificity, the DLS's operating point was prespecified to favor sensitivity over specificity. On the combined test set, the DLS's ROC curve was above all 9 India-based radiologists, with an AUC of 0.90 (95%CI 0.87-0.92). The DLS's sensitivity (88%) was higher than the India-based radiologists (75% mean sensitivity), p<0.001 for superiority; and its specificity (79%) was non-inferior to the radiologists (84% mean specificity), p=0.004. Similar trends were observed within HIV positive and sputum smear positive sub-groups, and in the South Africa test set. We found that 5 US-based radiologists (where TB isn't endemic) were more sensitive and less specific than the India-based radiologists (where TB is endemic). The DLS also remained non-inferior to the US-based radiologists. In simulations, using the DLS as a prioritization tool for confirmatory testing reduced the cost per positive case detected by 40-80% compared to using confirmatory testing alone. To conclude, our DLS generalized to 5 countries, and merits prospective evaluation to assist cost-effective screening efforts in radiologist-limited settings. Operating point flexibility may permit customization of the DLS to account for site-specific factors such as TB prevalence, demographics, clinical resources, and customary practice patterns.

eess.IV↗

arXiv2021

M6: A Chinese Multimodal Pretrainer

In this work, we construct the largest dataset for multimodal pretraining in Chinese, which consists of over 1.9TB images and 292GB texts that cover a wide range of domains. We propose a cross-modal pretraining method called M6, referring to Multi-Modality to Multi-Modality Multitask Mega-transformer, for unified pretraining on the data of single modality and multiple modalities. We scale the model size up to 10 billion and 100 billion parameters, and build the largest pretrained model in Chinese. We apply the model to a series of downstream applications, and demonstrate its outstanding performance in comparison with strong baselines. Furthermore, we specifically design a downstream task of text-guided image generation, and show that the finetuned M6 can create high-quality images with high resolution and abundant details.

cs.CL↗

arXiv2021

From Machine Learning to Transfer Learning in Laser-Induced Breakdown Spectroscopy: the Case of Rock Analysis for Mars Exploration

With the ChemCam instrument, laser-induced breakdown spectroscopy (LIBS) has successively contributed to Mars exploration by determining elemental compositions of the soil, crust and rocks. Two new lunched missions, Chinese Tianwen 1 and American Perseverance, will further increase the number of LIBS instruments on Mars after the planned landings in spring 2021. Such unprecedented situation requires a reinforced research effort on the methods of LIBS spectral data treatment. Although the matrix effects correspond to a general issue in LIBS, they become accentuated in the case of rock analysis for Mars exploration, because of the large variation of rock composition leading to the chemical matrix effect, and the difference in morphology between laboratory standard samples (in pressed pellet, glass or ceramics) used to establish calibration models and natural rocks encountered on Mars, leading to the physical matric effect. The chemical matrix effect has been tackled in the ChemCam project with large sets of laboratory standard samples offering a good representation of various compositions of Mars rocks. The present work deals with the physical matrix effect which is still expecting a satisfactory solution. The approach consists in introducing transfer learning in LIBS data treatment. For the specific case of total alkali-silica (TAS) classification of natural rocks, the results show a significant improvement of the prediction capacity of pellet sample-based models when trained together with suitable information from rocks in a procedure of transfer learning. The correct classification rate of rocks increases from 33.3% with a machine learning model to 83.3% with a transfer learning model.

physics.app-ph↗

arXiv2020

Deep Learning for Distinguishing Normal versus Abnormal Chest Radiographs and Generalization to Unseen Diseases

Chest radiography (CXR) is the most widely-used thoracic clinical imaging modality and is crucial for guiding the management of cardiothoracic conditions. The detection of specific CXR findings has been the main focus of several artificial intelligence (AI) systems. However, the wide range of possible CXR abnormalities makes it impractical to build specific systems to detect every possible condition. In this work, we developed and evaluated an AI system to classify CXRs as normal or abnormal. For development, we used a de-identified dataset of 248,445 patients from a multi-city hospital network in India. To assess generalizability, we evaluated our system using 6 international datasets from India, China, and the United States. Of these datasets, 4 focused on diseases that the AI was not trained to detect: 2 datasets with tuberculosis and 2 datasets with coronavirus disease 2019. Our results suggest that the AI system generalizes to new patient populations and abnormalities. In a simulated workflow where the AI system prioritized abnormal cases, the turnaround time for abnormal cases reduced by 7-28%. These results represent an important step towards evaluating whether AI can be safely used to flag cases in a general setting where previously unseen abnormalities exist.

eess.IV↗