arXiv subjects

Xinyi Shen

Publications and source records attributed to Xinyi Shen.

10 recordsLinked to original sources

arXiv2026

A Hybrid BLE-Wi-Fi Communication Architecture for Adaptive Imaging in Wireless Capsule Endoscopy

Wireless capsule endoscopy (WCE) is fundamentally constrained by limited wireless bandwidth, resulting in low imaging resolution and frame rate, which can cause motion blur and missed lesions. Although adaptive frame-rate schemes have been explored to accommodate transient gastrointestinal (GI) motility, these approaches typically require sacrificing image resolution. The use of higher-frequency communication bands is further limited by increased tissue attenuation. To address these challenges, we propose a hybrid Bluetooth Low Energy (BLE) and WiFi communication architecture that combines the low-power operation of BLE with the high data throughput of WiFi. We systematically evaluate the performance of BLE and WiFi under tissue-mimicking conditions by measuring throughput, received signal strength indicator (RSSI), and power consumption. The results demonstrate that amplified BLE with an adaptive transmission power control strategy provides a stable frame rate at low power consumption, while 2.4 GHz WiFi operating in station mode is the most suitable high-throughput communication configuration for WCE. Compared with WiFi, BLE reduces power consumption by approximately ten times, whereas WiFi achieves up to ten times higher throughput. To reconcile these complementary trade-offs, we further introduce a hybrid system with a frame-boundary-synchronized switching mechanism to ensure lossless data transmission during BLE and WiFi transitions. Experimental results show that the switching latency from BLE to WiFi is approximately 92.66 ms, which is longer than the WiFi-to-BLE switching latency of 15.49 ms when transmitting 10 kB image payloads. Overall, the proposed hybrid BLE and WiFi system enables robust, lossless, and energy-efficient mode switching, supports adaptive imaging, and advances the development of next-generation autonomous WCE platforms.

eess.SY↗

arXiv2026

Reevaluating Bluetooth Low Energy for Ingestible Electronics

Bluetooth Low Energy (BLE) is widely used in wearable devices but has rarely been considered for ingestible electronics because of severe tissue attenuation at 2.4 GHz. In this work, we systematically reevaluate BLE for ingestible applications by benchmarking it against representative Sub-GHz schemes in terms of tissue attenuation, power consumption, throughput, reliability, bidirectional communication, latency, and system integration. Water-phantom and in vivo measurements confirm stronger attenuation at 2.4 GHz, by approximately 40 dB; however, front-end-module (FEM)-assisted BLE compensates for this loss while maintaining practical received signal levels and energy efficiency. For ingestible sensing applications requiring below 100 kbps, BLE with a 20 dBm transmission power achieves lower power consumption than the evaluated Sub-GHz alternatives. BLE also provides protocol-level reliability through retransmission and in-order packet delivery, whereas Sub-GHz links can suffer packet loss under high attenuation. In addition, BLE supports bidirectional communication with limited throughput degradation, while the evaluated Sub-GHz implementation shows a substantial reduction. End-to-end measurements further demonstrate lower BLE latency because of its native compatibility with modern computing infrastructure. Finally, BLE offers advantages in antenna form factor, ecosystem support, and system integration. These results demonstrate that properly configured BLE is a practical communication solution for next-generation ingestible electronics.

eess.SY↗

arXiv2025

MoEcho: Exploiting Side-Channel Attacks to Compromise User Privacy in Mixture-of-Experts LLMs

The transformer architecture has become a cornerstone of modern AI, fueling remarkable progress across applications in natural language processing, computer vision, and multimodal learning. As these models continue to scale explosively for performance, implementation efficiency remains a critical challenge. Mixture of Experts (MoE) architectures, selectively activating specialized subnetworks (experts), offer a unique balance between model accuracy and computational cost. However, the adaptive routing in MoE architectures, where input tokens are dynamically directed to specialized experts based on their semantic meaning inadvertently opens up a new attack surface for privacy breaches. These input-dependent activation patterns leave distinctive temporal and spatial traces in hardware execution, which adversaries could exploit to deduce sensitive user data. In this work, we propose MoEcho, discovering a side channel analysis based attack surface that compromises user privacy on MoE based systems. Specifically, in MoEcho, we introduce four novel architectural side channels on different computing platforms, including Cache Occupancy Channels and Pageout+Reload on CPUs, and Performance Counter and TLB Evict+Reload on GPUs, respectively. Exploiting these vulnerabilities, we propose four attacks that effectively breach user privacy in large language models (LLMs) and vision language models (VLMs) based on MoE architectures: Prompt Inference Attack, Response Reconstruction Attack, Visual Inference Attack, and Visual Reconstruction Attack. MoEcho is the first runtime architecture level security analysis of the popular MoE structure common in modern transformers, highlighting a serious security and privacy threat and calling for effective and timely safeguards when harnessing MoE based models for developing efficient large scale AI services.

cs.CR↗

arXiv2024

An Empirical Study on Context Length for Open-Domain Dialog Generation

Transformer-based open-domain dialog models have become increasingly popular in recent years. These models typically represent context as a concatenation of a dialog history. However, there is no criterion to decide how many utterances should be kept adequate in a context. We try to figure out how the choice of context length affects the model. We experiment on three questions from coarse to fine: (i) Does longer context help model training? (ii) Is it necessary to change the training context length when dealing with dialogs of different context lengths? (iii) Do different dialog samples have the same preference for context length? Our experimental results show that context length, an often overlooked setting, deserves attention when implementing Transformer-based dialog models.

cs.CL↗

arXiv2024

SciKnowEval: Evaluating Multi-level Scientific Knowledge of Large Language Models

Large language models (LLMs) are playing an increasingly important role in scientific research, yet there remains a lack of comprehensive benchmarks to evaluate the breadth and depth of scientific knowledge embedded in these models. To address this gap, we introduce SciKnowEval, a large-scale dataset designed to systematically assess LLMs across five progressive levels of scientific understanding: memory, comprehension, reasoning, discernment, and application. SciKnowEval comprises 28K multi-level questions and solutions spanning biology, chemistry, physics, and materials science. Using this benchmark, we evaluate 20 leading open-source and proprietary LLMs. The results show that while proprietary models often achieve state-of-the-art performance, substantial challenges remain -- particularly in scientific reasoning and real-world application. We envision SciKnowEval as a standard benchmark for evaluating scientific capabilities in LLMs and as a catalyst for advancing more capable and reliable scientific language models.

cs.CL↗

arXiv2024

Local and Global Contexts for Conversation

The context in conversation is the dialog history crucial for multi-turn dialogue. Learning from the relevant contexts in dialog history for grounded conversation is a challenging problem. Local context is the most neighbor and more sensitive to the subsequent response, and global context is relevant to a whole conversation far beyond neighboring utterances. Currently, pretrained transformer models for conversation challenge capturing the correlation and connection between local and global contexts. We introduce a local and global conversation model (LGCM) for general-purpose conversation in open domain. It is a local-global hierarchical transformer model that excels at accurately discerning and assimilating the relevant contexts necessary for generating responses. It employs a local encoder to grasp the local context at the level of individual utterances and a global encoder to understand the broader context at the dialogue level. The seamless fusion of these locally and globally contextualized encodings ensures a comprehensive comprehension of the conversation. Experiments on popular datasets show that LGCM outperforms the existing conversation models on the performance of automatic metrics with significant margins.

cs.CL↗

arXiv2023

UADB: Unsupervised Anomaly Detection Booster

Unsupervised Anomaly Detection (UAD) is a key data mining problem owing to its wide real-world applications. Due to the complete absence of supervision signals, UAD methods rely on implicit assumptions about anomalous patterns (e.g., scattered/sparsely/densely clustered) to detect anomalies. However, real-world data are complex and vary significantly across different domains. No single assumption can describe such complexity and be valid in all scenarios. This is also confirmed by recent research that shows no UAD method is omnipotent. Based on above observations, instead of searching for a magic universal winner assumption, we seek to design a general UAD Booster (UADB) that empowers any UAD models with adaptability to different data. This is a challenging task given the heterogeneous model structures and assumptions adopted by existing UAD methods. To achieve this, we dive deep into the UAD problem and find that compared to normal data, anomalies (i) lack clear structure/pattern in feature space, thus (ii) harder to learn by model without a suitable assumption, and finally, leads to (iii) high variance between different learners. In light of these findings, we propose to (i) distill the knowledge of the source UAD model to an imitation learner (booster) that holds no data assumption, then (ii) exploit the variance between them to perform automatic correction, and thus (iii) improve the booster over the original UAD model. We use a neural network as the booster for its strong expressive power as a universal approximator and ability to perform flexible post-hoc tuning. Note that UADB is a model-agnostic framework that can enhance heterogeneous UAD models in a unified way. Extensive experiments on over 80 tabular datasets demonstrate the effectiveness of UADB.

cs.LG↗

arXiv2022

PVT: Point-Voxel Transformer for Point Cloud Learning

The recently developed pure Transformer architectures have attained promising accuracy on point cloud learning benchmarks compared to convolutional neural networks. However, existing point cloud Transformers are computationally expensive since they waste a significant amount of time on structuring the irregular data. To solve this shortcoming, we present Sparse Window Attention (SWA) module to gather coarse-grained local features from non-empty voxels, which not only bypasses the expensive irregular data structuring and invalid empty voxel computation, but also obtains linear computational complexity with respect to voxel resolution. Meanwhile, to gather fine-grained features about the global shape, we introduce relative attention (RA) module, a more robust self-attention variant for rigid transformations of objects. Equipped with the SWA and RA, we construct our neural architecture called PVT that integrates both modules into a joint framework for point cloud learning. Compared with previous Transformer-based and attention-based models, our method attains top accuracy of 94.0% on classification benchmark and 10x inference speedup on average. Extensive experiments also valid the effectiveness of PVT on part and semantic segmentation benchmarks (86.6% and 69.2% mIoU, respectively).

cs.CV↗

arXiv2022

PatchFormer: An Efficient Point Transformer with Patch Attention

The point cloud learning community witnesses a modeling shift from CNNs to Transformers, where pure Transformer architectures have achieved top accuracy on the major learning benchmarks. However, existing point Transformers are computationally expensive since they need to generate a large attention map, which has quadratic complexity (both in space and time) with respect to input size. To solve this shortcoming, we introduce Patch ATtention (PAT) to adaptively learn a much smaller set of bases upon which the attention maps are computed. By a weighted summation upon these bases, PAT not only captures the global shape context but also achieves linear complexity to input size. In addition, we propose a lightweight Multi-Scale aTtention (MST) block to build attentions among features of different scales, providing the model with multi-scale features. Equipped with the PAT and MST, we construct our neural architecture called PatchFormer that integrates both modules into a joint framework for point cloud learning. Extensive experiments demonstrate that our network achieves comparable accuracy on general point cloud learning tasks with 9.2x speed-up than previous point Transformers.

cs.CV↗

arXiv2020

Quantifying socioeconomic activities and weather effects on the global spread of COVID-19 epidemic

The COVID-19 has caused more than three million infections and over two hundred thousand deaths by April 20201. Limiting socioeconomic activities (SA) is among the most adopted governmental mitigating efforts to combat the transmission of the virus, though the degree varies dramatically among different regimes2. This study aims to quantify the contribution from the SA and weather conditions to the transmission of COVID-19 at global scale. Ruling out the unobservable factors including medical facilities and other control policies (MOC) through region-by-time fixed effects3,4, we show that the limiting SA has a leading contribution to lower the reproductive number by 18.3%, while weather conditions, including ultraviolet, relative humidity, and wind explain a smaller amount of variation. Temperature might have a non-monotonic impact on the transmission. We further show that in developed countries5 and China, the SA effect is more pronounced whereas the weather effect is significantly downplayed possibly because people tend to stay indoors most of the time with a controlled climate. We finally estimate the reduced reproductive number and the population spared from infections due to restricting SA at 40,964, 180,336, 174,494, in China, United States, and Europe respectively. From late January to mid-April, all regions, except for China, Australia, and south Korea show a steep upward trend of spared infections due to restricting SA. US and Europe, in particular, show far steeper upward trends of spared infections in the analyzed timeframe, signaling a greater risk of reopening the economy too soon.

q-bio.PE↗