arXiv subjects

Shan Jiang

Publications and source records attributed to Shan Jiang.

At least 55 records · Page 3Linked to original sources

arXiv2025

An approach for API synthesis using large language models

APIs play a pivotal role in modern software development by enabling seamless communication and integration between various systems, applications, and services. Component-based API synthesis is a form of program synthesis that constructs an API by assembling predefined components from a library. Existing API synthesis techniques typically implement dedicated search strategies over bounded spaces of possible implementations, which can be very large and time consuming to explore. In this paper, we present a novel approach of using large language models (LLMs) in API synthesis. LLMs offer a foundational technology to capture developer insights and provide an ideal framework for enabling more effective API synthesis. We perform an experimental evaluation of our approach using 135 real-world programming tasks, and compare it with FrAngel, a state-of-the-art API synthesis tool. The experimental results show that our approach completes 133 of the tasks, and overall outperforms FrAngel. We believe LLMs provide a very useful foundation for tackling the problem of API synthesis, in particular, and program synthesis, in general.

cs.SE↗

arXiv2025

On the Effectiveness of Large Language Models in Writing Alloy Formulas

Declarative specifications have a vital role to play in developing safe and dependable software systems. Writing specifications correctly, however, remains particularly challenging. This paper presents a controlled experiment on using large language models (LLMs) to write declarative formulas in the well-known language Alloy. Our use of LLMs is three-fold. One, we employ LLMs to write complete Alloy formulas from given natural language descriptions (in English). Two, we employ LLMs to create alternative but equivalent formulas in Alloy with respect to given Alloy formulas. Three, we employ LLMs to complete sketches of Alloy formulas and populate the holes in the sketches by synthesizing Alloy expressions and operators so that the completed formulas accurately represent the desired properties (that are given in natural language). We conduct the experimental evaluation using 11 well-studied subject specifications and employ two popular LLMs, namely ChatGPT and DeepSeek. The experimental results show that the LLMs generally perform well in synthesizing complete Alloy formulas from input properties given in natural language or in Alloy, and are able to enumerate multiple unique solutions. Moreover, the LLMs are also successful at completing given sketches of Alloy formulas with respect to natural language descriptions of desired properties (without requiring test cases). We believe LLMs offer a very exciting advance in our ability to write specifications, and can help make specifications take a pivotal role in software development and enhance our ability to build robust software.

cs.SE↗

arXiv2025

Real-Time Neural-Enhancement for Online Cloud Gaming

Online Cloud gaming demands real-time, high-quality video transmission across variable wide-area networks (WANs). Neural-enhanced video transmission algorithms employing super-resolution (SR) for video quality enhancement have effectively challenged WAN environments. However, these SR-based methods require intensive fine-tuning for the whole video, making it infeasible in diverse online cloud gaming. To address this, we introduce River, a cloud gaming delivery framework designed based on the observation that video segment features in cloud gaming are typically repetitive and redundant. This permits a significant opportunity to reuse fine-tuned SR models, reducing the fine-tuning latency of minutes to query latency of milliseconds. To enable the idea, we design a practical system that addresses several challenges, such as model organization, online model scheduler, and transfer strategy. River first builds a content-aware encoder that fine-tunes SR models for diverse video segments and stores them in a lookup table. When delivering cloud gaming video streams online, River checks the video features and retrieves the most relevant SR models to enhance the frame quality. Meanwhile, if no existing SR model performs well enough for some video segments, River will further fine-tune new models and update the lookup table. Finally, to avoid the overhead of streaming model weight to the clients, River designs a prefetching strategy that predicts the models with the highest possibility of being retrieved. Our evaluation based on real video game streaming demonstrates River can reduce redundant training overhead by 44% and improve the Peak-Signal-to-Noise-Ratio by 1.81dB compared to the SOTA solutions. Practical deployment shows River meets real-time requirements, achieving approximately 720p 20fps on mobile devices.

cs.NI↗

arXiv2024

Generating executable oracles to check conformance of client code to requirements of JDK Javadocs using LLMs

Software testing remains the most widely used methodology for validating quality of code. However, effectiveness of testing critically depends on the quality of test suites used. Test cases in a test suite consist of two fundamental parts: (1) input values for the code under test, and (2) correct checks for the outputs it produces. These checks are commonly written as assertions, and termed test oracles. The last couple of decades have seen much progress in automated test input generation, e.g., using fuzzing and symbolic execution. However, automating test oracles remains a relatively less explored problem area. Indeed, a test oracle by its nature requires knowledge of expected behavior, which may only be known to the developer and may not not exist in a formal language that supports automated reasoning. Our focus in this paper is automation of test oracles for clients of widely used Java libraries, e.g., java.lang and java.util packages. Our key insight is that Javadocs that provide a rich source of information can enable automated generation of test oracles. Javadocs of the core Java libraries are fairly detailed documents that contain natural language descriptions of not only how the libraries behave but also how the clients must (not) use them. We use large language models as an enabling technology to embody our insight into a framework for test oracle automation, and evaluate it experimentally. Our experiments demonstrate that LLMs can generate oracles for checking normal and exceptional behaviors from Javadocs, with 98.8% of these oracles being compilable and 96.4% accurately reflecting intended properties. Even for the few incorrect oracles, errors are minor and can be easily corrected with the help of additional comment information generated by the LLMs.

cs.SE↗

arXiv2024

Read Anywhere Pointed: Layout-aware GUI Screen Reading with Tree-of-Lens Grounding

Graphical User Interfaces (GUIs) are central to our interaction with digital devices and growing efforts have been made to build models for various GUI understanding tasks. However, these efforts largely overlook an important GUI-referring task: screen reading based on user-indicated points, which we name the Screen Point-and-Read (ScreenPR) task. Currently, this task is predominantly handled by rigid accessible screen reading tools, in great need of new models driven by advancements in Multimodal Large Language Models (MLLMs). In this paper, we propose a Tree-of-Lens (ToL) agent, utilizing a novel ToL grounding mechanism, to address the ScreenPR task. Based on the input point coordinate and the corresponding GUI screenshot, our ToL agent constructs a Hierarchical Layout Tree. Based on the tree, our ToL agent not only comprehends the content of the indicated area but also articulates the layout and spatial relationships between elements. Such layout information is crucial for accurately interpreting information on the screen, distinguishing our ToL agent from other screen reading tools. We also thoroughly evaluate the ToL agent against other baselines on a newly proposed ScreenPR benchmark, which includes GUIs from mobile, web, and operating systems. Last but not least, we test the ToL agent on mobile GUI navigation tasks, demonstrating its utility in identifying incorrect actions along the path of agent execution trajectories. Code and data: https://screen-point-and-read.github.io

cs.CL↗

arXiv2024

Magnetostriction, piezomagnetism and domain nucleation in a kagome antiferromagnet

Whenever the elastic energy of a solid depends on magnetic field, there is a magnetostrictive response. Field-linear magnetostriction implies piezomagnetism and vice versa. Here, we show that Mn$_3$Sn, a non-collinear antiferromanget with Weyl nodes, hosts a large and almost perfectly linear magnetostriction even at room temperature. The longitudinal and transverse magnetostriction, with opposite signs and similar amplitude are restricted to the kagome planes and the out-of-plane response is negligibly small. By studying four different samples with different Mn:Sn ratios, we find a clear correlation between the linear magnetostriction, the spontaneous magnetization and the concentration of Sn vacancies. The recently reported piezomagnetic data fits in our picture. We show that linear magnetostriction and piezomagnetism are both driven by the field-induced in-plane twist of spins. A quantitative account of the experimental data requires the distortion of the spin texture by Sn vacancies. We find that the field-induced domain nucleation within the hysteresis loop corresponds to a phase transition. Within the hysteresis loop, a concomitant mesoscopic modulation of local strain and spin twist angles, leading to twisto-magnetic stripes, arises as a result of the competition between elastic and magnetic energies.

cond-mat.str-el↗

arXiv2024

Muffin or Chihuahua? Challenging Multimodal Large Language Models with Multipanel VQA

Multipanel images, commonly seen as web screenshots, posters, etc., pervade our daily lives. These images, characterized by their composition of multiple subfigures in distinct layouts, effectively convey information to people. Toward building advanced multimodal AI applications, such as agents that understand complex scenes and navigate through webpages, the skill of multipanel visual reasoning is essential, and a comprehensive evaluation of models in this regard is important. Therefore, we introduce Multipanel Visual Question Answering (MultipanelVQA), a novel benchmark comprising 6,600 triplets of questions, answers, and multipanel images that specifically challenge models in comprehending multipanel images. Our evaluation shows that questions in the MultipanelVQA benchmark pose significant challenges to the state-of-the-art Multimodal Large Language Models (MLLMs) tested, even though humans can attain approximately 99% accuracy on these questions. Distinctively, the MultipanelVQA benchmark features synthetically generated multipanel images specifically crafted to isolate and assess the impact of various factors, such as the layout, on MLLMs' multipanel image comprehension abilities. As a result, in addition to benchmarking the capabilities of MLLMs in understanding multipanel images, we analyze various factors of the multipanel image that affect MLLMs' performance with synthetic data and offer insights for enhancement. Code and data are released at https://sites.google.com/view/multipanelvqa/home.

cs.CV↗

arXiv2024

PRoDeliberation: Parallel Robust Deliberation for End-to-End Spoken Language Understanding

Spoken Language Understanding (SLU) is a critical component of voice assistants; it consists of converting speech to semantic parses for task execution. Previous works have explored end-to-end models to improve the quality and robustness of SLU models with Deliberation, however these models have remained autoregressive, resulting in higher latencies. In this work we introduce PRoDeliberation, a novel method leveraging a Connectionist Temporal Classification-based decoding strategy as well as a denoising objective to train robust non-autoregressive deliberation models. We show that PRoDeliberation achieves the latency reduction of parallel decoding (2-10x improvement over autoregressive models) while retaining the ability to correct Automatic Speech Recognition (ASR) mistranscriptions of autoregressive deliberation systems. We further show that the design of the denoising training allows PRoDeliberation to overcome the limitations of small ASR devices, and we provide analysis on the necessity of each component of the system.

cs.CL↗

arXiv2024

The polarisation fluctuation length scale shaping the superconducting dome of SrTiO$_3$

Superconducting domes, ubiquitous across a variety of quantum materials, are often understood as a window favorite for pairing opened by the fluctuations of competing orders. Yet, a quantitative understanding of how such a window closes is missing. Here, we show that inelastic neutron scattering, by quantifying a length scale associated with polar fluctuations, $\ell_0$, addresses this issue. We find that the superconducting dome of strontium titanate definitely ends when $\ell_0$ vanishes. Moreover, the product of $\ell_0$ and the Fermi wavevector peaks close to the maximum critical temperature. Thus, this superconducting dome stems from the competition between the increase of the density of states and the unavoidable collapse of the quantum paraelectric phase, both induced by doping. The successful quantitative account of both the peak and the end of the superconducting dome implies a central role in the pairing mechanism played by the soft ferro-electric mode and its hybridisation with the acoustic branch. Such a scenario may also be at work in other quantum paraelectric materials, either bulk or interfaces.

cond-mat.supr-con↗

arXiv2024

Multi-attention Associate Prediction Network for Visual Tracking

Classification-regression prediction networks have realized impressive success in several modern deep trackers. However, there is an inherent difference between classification and regression tasks, so they have diverse even opposite demands for feature matching. Existed models always ignore the key issue and only employ a unified matching block in two task branches, decaying the decision quality. Besides, these models also struggle with decision misalignment situation. In this paper, we propose a multi-attention associate prediction network (MAPNet) to tackle the above problems. Concretely, two novel matchers, i.e., category-aware matcher and spatial-aware matcher, are first designed for feature comparison by integrating self, cross, channel or spatial attentions organically. They are capable of fully capturing the category-related semantics for classification and the local spatial contexts for regression, respectively. Then, we present a dual alignment module to enhance the correspondences between two branches, which is useful to find the optimal tracking solution. Finally, we describe a Siamese tracker built upon the proposed prediction network, which achieves the leading performance on five tracking benchmarks, consisting of LaSOT, TrackingNet, GOT-10k, TNL2k and UAV123, and surpasses other state-of-the-art approaches.

cs.CV↗

arXiv2024

Resource-efficient Parallel Split Learning in Heterogeneous Edge Computing

Edge AI has been recently proposed to facilitate the training and deployment of Deep Neural Network (DNN) models in proximity to the sources of data. To enable the training of large models on resource-constraint edge devices and protect data privacy, parallel split learning is becoming a practical and popular approach. However, current parallel split learning neglects the resource heterogeneity of edge devices, which may lead to the straggler issue. In this paper, we propose EdgeSplit, a novel parallel split learning framework to better accelerate distributed model training on heterogeneous and resource-constraint edge devices. EdgeSplit enhances the efficiency of model training on less powerful edge devices by adaptively segmenting the model into varying depths. Our approach focuses on reducing total training time by formulating and solving a task scheduling problem, which determines the most efficient model partition points and bandwidth allocation for each device. We employ a straightforward yet effective alternating algorithm for this purpose. Comprehensive tests conducted with a range of DNN models and datasets demonstrate that EdgeSplit not only facilitates the training of large models on resource-restricted edge devices but also surpasses existing baselines in performance.

cs.DC↗

arXiv2023

Hard to Track Objects with Irregular Motions and Similar Appearances? Make It Easier by Buffering the Matching Space

We propose a Cascaded Buffered IoU (C-BIoU) tracker to track multiple objects that have irregular motions and indistinguishable appearances. When appearance features are unreliable and geometric features are confused by irregular motions, applying conventional Multiple Object Tracking (MOT) methods may generate unsatisfactory results. To address this issue, our C-BIoU tracker adds buffers to expand the matching space of detections and tracks, which mitigates the effect of irregular motions in two aspects: one is to directly match identical but non-overlapping detections and tracks in adjacent frames, and the other is to compensate for the motion estimation bias in the matching space. In addition, to reduce the risk of overexpansion of the matching space, cascaded matching is employed: first matching alive tracks and detections with a small buffer, and then matching unmatched tracks and detections with a large buffer. Despite its simplicity, our C-BIoU tracker works surprisingly well and achieves state-of-the-art results on MOT datasets that focus on irregular motions and indistinguishable appearances. Moreover, the C-BIoU tracker is the dominant component for our 2-nd place solution in the CVPR'22 SoccerNet MOT and ECCV'22 MOTComplex DanceTrack challenges. Finally, we analyze the limitation of our C-BIoU tracker in ablation studies and discuss its application scope.

cs.CV↗

arXiv2023

Assessing and Enhancing Robustness of Deep Learning Models with Corruption Emulation in Digital Pathology

Deep learning in digital pathology brings intelligence and automation as substantial enhancements to pathological analysis, the gold standard of clinical diagnosis. However, multiple steps from tissue preparation to slide imaging introduce various image corruptions, making it difficult for deep neural network (DNN) models to achieve stable diagnostic results for clinical use. In order to assess and further enhance the robustness of the models, we analyze the physical causes of the full-stack corruptions throughout the pathological life-cycle and propose an Omni-Corruption Emulation (OmniCE) method to reproduce 21 types of corruptions quantified with 5-level severity. We then construct three OmniCE-corrupted benchmark datasets at both patch level and slide level and assess the robustness of popular DNNs in classification and segmentation tasks. Further, we explore to use the OmniCE-corrupted datasets as augmentation data for training and experiments to verify that the generalization ability of the models has been significantly enhanced.

eess.IV↗

arXiv2023

Online Streaming Video Super-Resolution with Convolutional Look-Up Table

Online video streaming has fundamental limitations on the transmission bandwidth and computational capacity and super-resolution is a promising potential solution. However, applying existing video super-resolution methods to online streaming is non-trivial. Existing video codecs and streaming protocols (\eg, WebRTC) dynamically change the video quality both spatially and temporally, which leads to diverse and dynamic degradations. Furthermore, online streaming has a strict requirement for latency that most existing methods are less applicable. As a result, this paper focuses on the rarely exploited problem setting of online streaming video super resolution. To facilitate the research on this problem, a new benchmark dataset named LDV-WebRTC is constructed based on a real-world online streaming system. Leveraging the new benchmark dataset, we proposed a novel method specifically for online video streaming, which contains a convolution and Look-Up Table (LUT) hybrid model to achieve better performance-latency trade-off. To tackle the changing degradations, we propose a mixture-of-expert-LUT module, where a set of LUT specialized in different degradations are built and adaptively combined to handle different degradations. Experiments show our method achieves 720P video SR around 100 FPS, while significantly outperforms existing LUT-based methods and offers competitive performance compared to efficient CNN-based methods.

eess.IV↗

arXiv2023

T-square dependence of the electronic thermal resistivity in metallic strontium titanate

The temperature dependence of the phase space for electron-electron (e-e) collisions leads to a T-square contribution to electrical resistivity of metals. Umklapp scattering are identified as the origin of momentum loss due to e-e scattering in dense metals. However, in dilute metals like lightly doped strontium titanate, the origin of T-square electrical resistivity in absence of Umklapp events is yet to be pinned down. Here, by separating electron and phonon contributions to heat transport, we extract the electronic thermal resistivity in niobium-doped strontium titanate and show that it also displays a T-square temperature dependence. Its amplitude correlates with the T-square electrical resistivity. The Wiedemann-Franz law strictly holds in the zero-temperature limit, but not at finite-temperature, because the two T-square prefactors are different by a factor of $\approx 3$, like in other Fermi liquids. Recalling the case of $^3$He, we argue that T-square thermal resistivity does not require Umklapp events. The approximate recovery of the Wiedemann-Franz law in presence of disorder would account for a T-square electrical resistivity without Umklapp.

cond-mat.str-el↗

arXiv2023

Glass-like thermal conductivity and narrow insulating gap of EuTiO$_3$

Crystals and glasses differ by the amplitude and the temperature dependence of their thermal conductivity. However, there are crystals known to display glass-like thermal conductivity. Here, we show that EuTiO$_3$, a quantum paraelectric known to order antiferromagnetically at 5.5 K, is one such system. The temperature dependence of resistivity and Seebeck coefficient yield an insulating band gap of $\sim 0.22$ eV. Thermal conductivity is drastically reduced. Its amplitude and temperature dependence are akin to what is seen in amorphous silica. Comparison with non-magnetic perovskite solids, SrTiO$_3$, KTaO$_3$, and EuCoO$_3$, shows that what impedes heat transport are $4f$ spins at Eu$^{2+}$ sites, which couple to phonons well above the ordering temperature. Thus, in this case, superexchange and valence fluctuations, not magnetic frustration, are the drivers of the glass-like thermal conductivity.

cond-mat.mtrl-sci↗

arXiv2023

A Unified Multi-view Multi-person Tracking Framework

Although there is a significant development in 3D Multi-view Multi-person Tracking (3D MM-Tracking), current 3D MM-Tracking frameworks are designed separately for footprint and pose tracking. Specifically, frameworks designed for footprint tracking cannot be utilized in 3D pose tracking, because they directly obtain 3D positions on the ground plane with a homography projection, which is inapplicable to 3D poses above the ground. In contrast, frameworks designed for pose tracking generally isolate multi-view and multi-frame associations and may not be robust to footprint tracking, since footprint tracking utilizes fewer key points than pose tracking, which weakens multi-view association cues in a single frame. This study presents a Unified Multi-view Multi-person Tracking framework to bridge the gap between footprint tracking and pose tracking. Without additional modifications, the framework can adopt monocular 2D bounding boxes and 2D poses as the input to produce robust 3D trajectories for multiple persons. Importantly, multi-frame and multi-view information are jointly employed to improve the performance of association and triangulation. The effectiveness of our framework is verified by accomplishing state-of-the-art performance on the Campus and Shelf datasets for 3D pose tracking, and by comparable results on the WILDTRACK and MMPTRACK datasets for 3D footprint tracking.

cs.CV↗

arXiv2023

Multifunctional fiber-based optoacoustic emitter for non-genetic bidirectional neural communication

A bidirectional brain interface with both "write" and "read" functions can be an important tool for fundamental studies and potential clinical treatments for neurological diseases. Here we report a miniaturized multifunctional fiber based optoacoustic emitter (mFOE) that first integrates simultaneous non-genetic optoacoustic stimulation for "write" and electrophysiology recording of neural circuits for "read". The non-genetic feature addresses the challenges of the viral transfection required by optogenetics in primates and human. The orthogonality between optoacoustic waves and electrical field provides a solution to avoid the interference between electrical stimulation and recording. We first validated the non-genetic stimulation function of the mFOE in rat cultured neurons using calcium imaging. In vivo application of mFOE for successful simultaneous optoacoustic stimulation and electrical recording of brain activities was confirmed in mouse hippocampus in both acute and chronical applications up to 1 month. Minimal brain tissue damage has been confirmed after these applications. The capability of non-genetic neural stimulation and recording enabled by mFOE opens up new possibilities for the investigation of neural circuits and brings new insights into the study of ultrasound neurostimulation.

q-bio.NC↗