Searcharxiv⌕ Search

arXiv subjects

Yi Gao

Publications and source records attributed to Yi Gao.

At least 37 records · Page 2Linked to original sources
arXiv2026

The AI Amplifier Effect: Defining Human-AI Intimacy and Romantic Relationships with Conversational AI

What does it mean to fall in love with something we know is virtual? The proliferation of conversational AI enables users to create customizable companions, fostering new intimate relationships that, while virtual, are perceived as authentic. However, public understanding of these bonds is limited, and platform policies regarding these interactions remain inconsistent. There is a pressing need for further HCI research to investigate: (a) the design affordances in AI that construct bonds and a sense of intimacy, (b) how such long-term engagement impacts users' real lives, and (c) how to balance user autonomy with platform regulation in the design of these systems without compromising users' well-being and experiences. This paper takes a step toward addressing these goals by providing a concrete definition of human AI intimacy based on in depth interviews with 30 users engaged in romantic relationships with AI companions. We elucidate the complexities of these relationships, from their formation to sustainability, and identify key features of the bonds formed. Notably, we introduce the AI Amplifier Effect, where the AI serves as a medium that intensifies the user's existing emotional state, leading to divergent positive, neutral, and negative impacts. We argue that designing for emotion must extend beyond technical affordances to encompass the essence of human affection. This paper's contributions aim to initiate a conversation and guide future research on human AI relationships within the HCI community.

cs.HC↗
arXiv2026

An isoperimetric inequality for the second Robin eigenvalue of the Weighted Laplacian

In this paper, we investigate a shape optimization problem for the second Robin eigenvalue of the weighted Laplacian on bounded Lipschitz domains symmetric about the origin. Our main theorem states that the ball centered at the origin maximizes the second Robin eigenvalue among all Lipschitz bounded domains of prescribed weighted measure and symmetric about the origin for a range of negative Robin parameters.

math.AP↗
arXiv2026

Gen-Diaolou: An Integrated AI-Assisted Interactive System for Diachronic Understanding and Preservation of the Kaiping Diaolou

The Kaiping Diaolou and Villages, a UNESCO World Heritage Site, exemplify hybrid Chinese and Western architecture shaped by migration culture. However, architectural heritage engagement often faces authenticity debates, resource constraints, and limited participatory approaches. This research explores current challenges of leveraging Artificial Intelligence (AI) for architectural heritage, and how AI-assisted interactive systems can foster cultural heritage understanding and preservation awareness. We conducted a formative study (N=14) to uncover empirical insights from heritage stakeholders that inform design. These insights informed the design of Gen-Diaolou, an integrated AI-assisted interactive system that supports heritage understanding and preservation. A pilot study (N=18) and a museum field study (N=26) provided converging evidence suggesting that Gen-Diaolou may support visitors' diachronic understanding and preservation awareness, and together informed design implications for future human-AI collaborative systems for digital cultural heritage engagement. More broadly, this work bridges the research gap between passive heritage systems and unconstrained creative tools in the HCI domain.

cs.HC↗
arXiv2026

Multi-Agent End-to-End Vulnerability Management for Mitigating Recurring Vulnerabilities

Software vulnerability management has become increasingly critical as modern systems scale in size and complexity. However, existing automated approaches remain insufficient. Traditional static analysis methods struggle to precisely capture contextual dependencies, especially when vulnerabilities span multiple functions or modules. Large language models (LLMs) often lack the ability to retrieve and exploit sufficient contextual information, resulting in incomplete reasoning and unreliable outcomes. Meanwhile, recurring vulnerabilities emerge repeatedly due to code reuse and shared logic, making historical vulnerability knowledge an indispensable foundation for effective vulnerability detection and repair. Nevertheless, prior approaches such as clone-based detection and patch porting, have not fully leveraged this knowledge. To address these challenges, we present MAVM, a multi-agent framework for end-to-end recurring vulnerability management. MAVM integrates five components, including a vulnerability knowledge base, detection, confirmation, repair, and validation, into a unified multi-agent pipeline. We construct a knowledge base from publicly disclosed vulnerabilities, thereby addressing the underuse of historical knowledge in prior work and mitigating the lack of domain-specific expertise in LLMs. Furthermore, we design context-retrieval tools that allow agents to extract and reason over repository-level information, overcoming the contextual limitations of previous methods. Based on agents, MAVM effectively simulates real-world security workflows. To evaluate the performance of MAVM, we construct a dataset containing 78 real-world patch-porting cases (covering 114 function-level migrations). On this dataset, MAVM successfully detects and repairs 51 real vulnerabilities, outperforming baselines by 31.9%-45.2% in repair accuracy, which demonstrates its effectiveness.

cs.SE↗
arXiv2026

Bridging the Knowledge-Action Gap by Evaluating LLMs in Dynamic Dental Clinical Scenarios

The transition of Large Language Models (LLMs) from passive knowledge retrievers to autonomous clinical agents demands a shift in evaluation-from static accuracy to dynamic behavioral reliability. To explore this boundary in dentistry, a domain where high-quality AI advice uniquely empowers patient-participatory decision-making, we present the Standardized Clinical Management & Performance Evaluation (SCMPE) benchmark, which comprehensively assesses performance from knowledge-oriented evaluations (static objective tasks) to workflow-based simulations (multi-turn simulated patient interactions). Our analysis reveals that while models demonstrate high proficiency in static objective tasks, their performance precipitates in dynamic clinical dialogues, identifying that the primary bottleneck lies not in knowledge retention, but in the critical challenges of active information gathering and dynamic state tracking. Mapping "Guideline Adherence" versus "Decision Quality" reveals a prevalent "High Efficacy, Low Safety" risk in general models. Furthermore, we quantify the impact of Retrieval-Augmented Generation (RAG). While RAG mitigates hallucinations in static tasks, its efficacy in dynamic workflows is limited and heterogeneous, sometimes causing degradation. This underscores that external knowledge alone cannot bridge the reasoning gap without domain-adaptive pre-training. This study empirically charts the capability boundaries of dental LLMs, providing a roadmap for bridging the gap between standardized knowledge and safe, autonomous clinical practice.

cs.CL↗
arXiv2026

DepRadar: Agentic Coordination for Context Aware Defect Impact Analysis in Deep Learning Libraries

Deep learning libraries like Transformers and Megatron are now widely adopted in modern AI programs. However, when these libraries introduce defects, ranging from silent computation errors to subtle performance regressions, it is often challenging for downstream users to assess whether their own programs are affected. Such impact analysis requires not only understanding the defect semantics but also checking whether the client code satisfies complex triggering conditions involving configuration flags, runtime environments, and indirect API usage. We present DepRadar, an agent coordination framework for fine grained defect and impact analysis in DL library updates. DepRadar coordinates four specialized agents across three steps: 1. the PR Miner and Code Diff Analyzer extract structured defect semantics from commits or pull requests, 2. the Orchestrator Agent synthesizes these signals into a unified defect pattern with trigger conditions, and 3. the Impact Analyzer checks downstream programs to determine whether the defect can be triggered. To improve accuracy and explainability, DepRadar integrates static analysis with DL-specific domain rules for defect reasoning and client side tracing. We evaluate DepRadar on 157 PRs and 70 commits across two representative DL libraries. It achieves 90% precision in defect identification and generates high quality structured fields (average field score 1.6). On 122 client programs, DepRadar identifies affected cases with 90% recall and 80% precision, substantially outperforming other baselines.

cs.SE↗
arXiv2025

SemanticNN: Compressive and Error-Resilient Semantic Offloading for Extremely Weak Devices

With the rapid growth of the Internet of Things (IoT), integrating artificial intelligence (AI) on extremely weak embedded devices has garnered significant attention, enabling improved real-time performance and enhanced data privacy. However, the resource limitations of such devices and unreliable network conditions necessitate error-resilient device-edge collaboration systems. Traditional approaches focus on bit-level transmission correctness, which can be inefficient under dynamic channel conditions. In contrast, we propose SemanticNN, a semantic codec that tolerates bit-level errors in pursuit of semantic-level correctness, enabling compressive and resilient collaborative inference offloading under strict computational and communication constraints. It incorporates a Bit Error Rate (BER)-aware decoder that adapts to dynamic channel conditions and a Soft Quantization (SQ)-based encoder to learn compact representations. Building on this architecture, we introduce Feature-augmentation Learning, a novel training strategy that enhances offloading efficiency. To address encoder-decoder capability mismatches from asymmetric resources, we propose XAI-based Asymmetry Compensation to enhance decoding semantic fidelity. We conduct extensive experiments on STM32 using three models and six datasets across image classification and object detection tasks. Experimental results demonstrate that, under varying transmission error rates, SemanticNN significantly reduces feature transmission volume by 56.82-344.83x while maintaining superior inference accuracy.

cs.CV↗
arXiv2025

Probe-and-Release Coordination of Platoons at Highway Bottlenecks with Unknown Parameters

This paper considers coordination of platoons of connected and autonomous vehicles (CAVs) at mixed-autonomy bottlenecks in the face of three practically important factors, viz. time-varying traffic demand, random CAV platoon sizes, and capacity breakdowns. Platoon coordination is essential to smoothen the interaction between CAV platoons and non-CAV traffic. Based on a fluid queuing model, we develop a "probe-and-release" algorithm that simultaneously estimates environmental parameters and coordinates CAV platoons for traffic stabilization. We show that this algorithm ensures bounded estimation errors and bounded traffic queues. The proof builds on a Lyapunov function that jointly penalizes estimation errors and traffic queues and a drift argument for an embedded Markov process. We validate the proposed algorithm in a standard micro-simulation environment and compare against a representative deep reinforcement learning method in terms of control performance and computational efficiency.

eess.SY↗
arXiv2025

MoLink: Distributed and Efficient Serving Framework for Large Models

Large language models represent a groundbreaking shift in generative AI. Yet, these advances come with a significant challenge: the high cost of model serving. To mitigate these costs, consumer-grade GPUs emerge as a more affordable alternative. This presents an opportunity for more cost-efficient LLM serving by leveraging these GPUs. However, it is non-trivial to achieve high-efficiency LLM serving on consumer-grade GPUs, mainly due to two challenges: 1) these GPUs are often deployed in limited network conditions; 2) these GPUs often exhibit heterogeneity in host systems. To address these challenges, we present MoLink, a distributed LLM serving system for large models. It incorporates several key techniques, enabling efficient LLM serving on heterogeneous and weakly connected consumer-grade GPUs. Our experiments demonstrate that it achieves throughput improvements of up to 458\% and cost-profit margin improvements of up to 151\%, compared to state-of-the-art systems. MoLink allows users on Windows, Linux, and containerized VMs to seamlessly integrate GPUs with just a few lines of code over Ethernet or public networks. Currently, it supports 18 mainstream architectures of open-source large language models. The source code is publicly available https://github.com/oldcpple/MoLink.

cs.DC↗
arXiv2025

Theoretical study of the electronic correlation and superconducting pairing in La$_{2.85}$Pr$_{0.15}$Ni$_2$O$_7$ film grown on SrLaAlO$_4$

We investigate the electronic correlation effects in La$_{2.85}$Pr$_{0.15}$Ni$_2$O$_7$ film grown on SrLaAlO$_4$ by fluctuation-exchange approximation. The $γ$ and $δ$ bands are found to shift down with correlation while their Fermi surfaces become highly damped and cannot to be resolved experimentally. In contrast, the $α$ and $β$ Fermi surfaces do not vary much with correlation and they are sharp enough to be detected in experiments. The spin susceptibility peaks at a wave vector in the odd channel, connecting the symmetric $γ$ band and the asymmetric $δ$ band. The superconducting pairing symmetry is robustly $s$-wave and the $δ$ band has the largest pairing magnitude. All the findings suggest a dominant role of the $3d_{z^{2}}$ orbital in this material.

cond-mat.supr-con↗
arXiv2025

TensorShield: Safeguarding On-Device Inference by Shielding Critical DNN Tensors with TEE

To safeguard user data privacy, on-device inference has emerged as a prominent paradigm on mobile and Internet of Things (IoT) devices. This paradigm involves deploying a model provided by a third party on local devices to perform inference tasks. However, it exposes the private model to two primary security threats: model stealing (MS) and membership inference attacks (MIA). To mitigate these risks, existing wisdom deploys models within Trusted Execution Environments (TEEs), which is a secure isolated execution space. Nonetheless, the constrained secure memory capacity in TEEs makes it challenging to achieve full model security with low inference latency. This paper fills the gap with TensorShield, the first efficient on-device inference work that shields partial tensors of the model while still fully defending against MS and MIA. The key enabling techniques in TensorShield include: (i) a novel eXplainable AI (XAI) technique exploits the model's attention transition to assess critical tensors and shields them in TEE to achieve secure inference, and (ii) two meticulous designs with critical feature identification and latency-aware placement to accelerate inference while maintaining security. Extensive evaluations show that TensorShield delivers almost the same security protection as shielding the entire model inside TEE, while being up to 25.35$\times$ (avg. 5.85$\times$) faster than the state-of-the-art work, without accuracy loss.

cs.CR↗
arXiv2025

Gemma 3 Technical Report

We introduce Gemma 3, a multimodal addition to the Gemma family of lightweight open models, ranging in scale from 1 to 27 billion parameters. This version introduces vision understanding abilities, a wider coverage of languages and longer context - at least 128K tokens. We also change the architecture of the model to reduce the KV-cache memory that tends to explode with long context. This is achieved by increasing the ratio of local to global attention layers, and keeping the span on local attention short. The Gemma 3 models are trained with distillation and achieve superior performance to Gemma 2 for both pre-trained and instruction finetuned versions. In particular, our novel post-training recipe significantly improves the math, chat, instruction-following and multilingual abilities, making Gemma3-4B-IT competitive with Gemma2-27B-IT and Gemma3-27B-IT comparable to Gemini-1.5-Pro across benchmarks. We release all our models to the community.

cs.CL↗
arXiv2025

Robust $s_\pm$-wave pairing in a bilayer two-orbital model of pressurized La$_3$Ni$_2$O$_7$ without the $γ$ Fermi surface

We studied the superconducting pairing symmetry based on a newly constructed tight-binding model of La$_3$Ni$_2$O$_7$ under pressure, where the $γ$ band sinks below the Fermi level and does not form the Fermi surface. The superconducting pairing symmetry is $s_\pm$-wave and is robust against the variation of the interaction strength. In this model, although the $γ$ and $δ$ bands are away from the Fermi level, the superconducting pairing function on them is not tiny. Instead, since the top of the $γ$ band and bottom of the $δ$ band are both located at $\sim$500 meV away from the Fermi level, and they are almost nested by the peak structure in the spin fluctuation, thus by forming an anti-phase pairing function on them, these two bands act constructively to superconductivity. Finally with detailed derivation and numerical calculation, we demonstrate that the Fermi surface approximated Eliashberg equation may lead to deviation of the pairing symmetry.

cond-mat.supr-con↗
arXiv2025

Vulnerability-Triggering Test Case Generation from Third-Party Libraries

Open-source third-party libraries are widely used in software development. These libraries offer substantial advantages in terms of time and resource savings. However, a significant concern arises due to the publicly disclosed vulnerabilities within these libraries. Existing automated vulnerability detection tools often suffer from false positives and fail to accurately assess the propagation of inputs capable of triggering vulnerabilities from client projects to vulnerable code in libraries. In this paper, we propose a novel approach called VULEUT (Vulnerability Exploit Unit Test Generation), which combines vulnerability exploitation reachability analysis and LLM-based unit test generation. VULEUT is designed to automatically verify the exploitability of vulnerabilities in third-party libraries commonly used in client software projects. VULEUT first analyzes the client projects to determine the reachability of vulnerability conditions. And then, it leverages the Large Language Model (LLM) to generate unit tests for vulnerability confirmation. To evaluate the effectiveness of VULEUT, we collect 32 vulnerabilities from various third-party libraries and conduct experiments on 70 real client projects. Besides, we also compare our approach with two representative tools, i.e., TRANSFER and VESTA. Our results demonstrate the effectiveness of VULEUT, with 229 out of 292 generated unit tests successfully confirming vulnerability exploitation across 70 client projects, which outperforms baselines by 24%.

cs.SE↗
arXiv2025

Automated Unit Test Refactoring

Test smells arise from poor design practices and insufficient domain knowledge, which can lower the quality of test code and make it harder to maintain and update. Manually refactoring test smells is time-consuming and error-prone, highlighting the necessity for automated approaches. Current rule-based refactoring methods often struggle in scenarios not covered by predefined rules and lack the flexibility needed to handle diverse cases effectively. In this paper, we propose a novel approach called UTRefactor, a context-enhanced, LLM-based framework for automatic test refactoring in Java projects. UTRefactor extracts relevant context from test code and leverages an external knowledge base that includes test smell definitions, descriptions, and DSL-based refactoring rules. By simulating the manual refactoring process through a chain-of-thought approach, UTRefactor guides the LLM to eliminate test smells in a step-by-step process, ensuring both accuracy and consistency throughout the refactoring. Additionally, we implement a checkpoint mechanism to facilitate comprehensive refactoring, particularly when multiple smells are present. We evaluate UTRefactor on 879 tests from six open-source Java projects, reducing the number of test smells from 2,375 to 265, achieving an 89% reduction. UTRefactor outperforms direct LLM-based refactoring methods by 61.82% in smell elimination and significantly surpasses the performance of a rule-based test smell refactoring tool. Our results demonstrate the effectiveness of UTRefactor in enhancing test code quality while minimizing manual involvement.

cs.SE↗
arXiv2024

Theoretical investigation of the superconducting pairing symmetry in a bilayer two-orbital model of pressurized La$_3$Ni$_2$O$_7$

We investigate the superconducting pairing symmetry in pressurized La$_3$Ni$_2$O$_7$ based on a bilayer two-orbital model. There are two symmetric bands $α$ and $γ$, as well as two antisymmetric ones $β$ and $δ$. It is found that the $γ$ band induces considerable ferromagnetic spin fluctuation and prefers an odd-frequency, $s$-wave spin triplet pairing state. The addition of the other bands gradually suppresses the ferromagnetic spin fluctuation and enhances the antiferromagnetic ones. The superconducting pairing then evolves from a spin triplet into a $d$-wave spin singlet, and finally into an $s_\pm$-wave one. The competition between the $d$-wave and $s_\pm$-wave pairings relies on whether the ferromagnetic spin fluctuation is suppressed completely or not.

cond-mat.supr-con↗
arXiv2024

Can Students Beyond The Teacher? Distilling Knowledge from Teacher's Bias

Knowledge distillation (KD) is a model compression technique that transfers knowledge from a large teacher model to a smaller student model to enhance its performance. Existing methods often assume that the student model is inherently inferior to the teacher model. However, we identify that the fundamental issue affecting student performance is the bias transferred by the teacher. Current KD frameworks transmit both right and wrong knowledge, introducing bias that misleads the student model. To address this issue, we propose a novel strategy to rectify bias and greatly improve the student model's performance. Our strategy involves three steps: First, we differentiate knowledge and design a bias elimination method to filter out biases, retaining only the right knowledge for the student model to learn. Next, we propose a bias rectification method to rectify the teacher model's wrong predictions, fundamentally addressing bias interference. The student model learns from both the right knowledge and the rectified biases, greatly improving its prediction accuracy. Additionally, we introduce a dynamic learning approach with a loss function that updates weights dynamically, allowing the student model to quickly learn right knowledge-based easy tasks initially and tackle hard tasks corresponding to biases later, greatly enhancing the student model's learning efficiency. To the best of our knowledge, this is the first strategy enabling the student model to surpass the teacher model. Experiments demonstrate that our strategy, as a plug-and-play module, is versatile across various mainstream KD frameworks. We will release our code after the paper is accepted.

cs.CV↗