arXiv subjects

Qi Li

Publications and source records attributed to Qi Li.

At least 217 records · Page 12Linked to original sources

arXiv2025

Multi-Level Collaboration in Model Merging

Parameter-level model merging is an emerging paradigm in multi-task learning with significant promise. Previous research has explored its connections with prediction-level model ensembling-commonly viewed as the upper bound for merging-to reveal the potential of achieving performance consistency between the two. However, this observation relies on certain preconditions, such as being limited to two models, using ViT-based models, and all models are fine-tuned from the same pre-trained checkpoint. To further understand the intrinsic connections between model merging and model ensembling, this paper explores an interesting possibility: If these restrictions are removed, can performance consistency still be achieved between merging and ensembling? To answer this question, we first theoretically establish a performance correlation between merging and ensembling. We find that even when previous restrictions are not met, there is still a way for model merging to attain a near-identical and superior performance similar to that of ensembling. To verify whether our findings are practical, we introduce a validation framework termed Neural Ligand (NeuLig). The learning process of NeuLig is meticulously designed with a specialized loss function supported by theoretical foundations. Experimental results demonstrate the robust resilience of NeuLig in terms of both model scale and the number of collaborating models. For instance, for the case involving 5 CLIP-ViT-B/32 models, parameter-level merging achieves the same performance as prediction-level ensembling (merging: 95.44% vs. ensembling: 95.46%).

cs.LG↗

arXiv2025

GraphBridge: Towards Arbitrary Transfer Learning in GNNs

Graph neural networks (GNNs) are conventionally trained on a per-domain, per-task basis. It creates a significant barrier in transferring the acquired knowledge to different, heterogeneous data setups. This paper introduces GraphBridge, a novel framework to enable knowledge transfer across disparate tasks and domains in GNNs, circumventing the need for modifications to task configurations or graph structures. Specifically, GraphBridge allows for the augmentation of any pre-trained GNN with prediction heads and a bridging network that connects the input to the output layer. This architecture not only preserves the intrinsic knowledge of the original model but also supports outputs of arbitrary dimensions. To mitigate the negative transfer problem, GraphBridge merges the source model with a concurrently trained model, thereby reducing the source bias when applied to the target domain. Our method is thoroughly evaluated across diverse transfer learning scenarios, including Graph2Graph, Node2Node, Graph2Node, and graph2point-cloud. Empirical validation, conducted over 16 datasets representative of these scenarios, confirms the framework's capacity for task- and domain-agnostic transfer learning within graph-like data, marking a significant advancement in the field of GNNs. Code is available at https://github.com/jujulili888/GraphBridge.

cs.LG↗

arXiv2025

Na'vi or Knave: Jailbreaking Language Models via Metaphorical Avatars

Metaphor serves as an implicit approach to convey information, while enabling the generalized comprehension of complex subjects. However, metaphor can potentially be exploited to bypass the safety alignment mechanisms of Large Language Models (LLMs), leading to the theft of harmful knowledge. In our study, we introduce a novel attack framework that exploits the imaginative capacity of LLMs to achieve jailbreaking, the J\underline{\textbf{A}}ilbreak \underline{\textbf{V}}ia \underline{\textbf{A}}dversarial Me\underline{\textbf{TA}} -pho\underline{\textbf{R}} (\textit{AVATAR}). Specifically, to elicit the harmful response, AVATAR extracts harmful entities from a given harmful target and maps them to innocuous adversarial entities based on LLM's imagination. Then, according to these metaphors, the harmful target is nested within human-like interaction for jailbreaking adaptively. Experimental results demonstrate that AVATAR can effectively and transferablly jailbreak LLMs and achieve a state-of-the-art attack success rate across multiple advanced LLMs. Our study exposes a security risk in LLMs from their endogenous imaginative capabilities. Furthermore, the analytical study reveals the vulnerability of LLM to adversarial metaphors and the necessity of developing defense methods against jailbreaking caused by the adversarial metaphor. \textcolor{orange}{ \textbf{Warning: This paper contains potentially harmful content from LLMs.}}

cs.CL↗

arXiv2025

An Engorgio Prompt Makes Large Language Model Babble on

Auto-regressive large language models (LLMs) have yielded impressive performance in many real-world tasks. However, the new paradigm of these LLMs also exposes novel threats. In this paper, we explore their vulnerability to inference cost attacks, where a malicious user crafts Engorgio prompts to intentionally increase the computation cost and latency of the inference process. We design Engorgio, a novel methodology, to efficiently generate adversarial Engorgio prompts to affect the target LLM's service availability. Engorgio has the following two technical contributions. (1) We employ a parameterized distribution to track LLMs' prediction trajectory. (2) Targeting the auto-regressive nature of LLMs' inference process, we propose novel loss functions to stably suppress the appearance of the token, whose occurrence will interrupt the LLM's generation process. We conduct extensive experiments on 13 open-sourced LLMs with parameters ranging from 125M to 30B. The results show that Engorgio prompts can successfully induce LLMs to generate abnormally long outputs (i.e., roughly 2-13$\times$ longer to reach 90%+ of the output length limit) in a white-box scenario and our real-world experiment demonstrates Engergio's threat to LLM service with limited computing resources. The code is released at: https://github.com/jianshuod/Engorgio-prompt.

cs.CR↗

arXiv2025

Crime Forecasting: A Spatio-temporal Analysis with Deep Learning Models

This study uses deep-learning models to predict city partition crime counts on specific days. It helps police enhance surveillance, gather intelligence, and proactively prevent crimes. We formulate crime count prediction as a spatiotemporal sequence challenge, where both input data and prediction targets are spatiotemporal sequences. In order to improve the accuracy of crime forecasting, we introduce a new model that combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks. We conducted a comparative analysis to access the effects of various data sequences, including raw and binned data, on the prediction errors of four deep learning forecasting models. Directly inputting raw crime data into the forecasting model causes high prediction errors, making the model unsuitable for real - world use. The findings indicate that the proposed CNN-LSTM model achieves optimal performance when crime data is categorized into 10 or 5 groups. Data binning can enhance forecasting model performance, but poorly defined intervals may reduce map granularity. Compared to dividing into 5 bins, binning into 10 intervals strikes an optimal balance, preserving data characteristics and surpassing raw data in predictive modelling efficacy.

cs.LG↗

arXiv2025

What causes the ultraviolet extinction bump at the cosmic dawn?

The enigmatic ultraviolet (UV) extinction bump at 2175 Angstrom, the strongest spectroscopic absorption feature superimposed on the interstellar extinction curve, has recently been detected at the cosmic dawn by the James Webb Space Telescope (JWST) in JADES-GS-z6-0, a distant galaxy at redshift z=6.71, corresponding to a cosmic age of just 800 million years after the Big Bang. Although small graphite grains have historically long been suggested as the carrier of the 2175 Angstrom extinction bump and graphite grains are expected to have already been pervasive in the early Universe, in this work we demonstrate that small graphite grains are not responsible for the UV extinction bump seen at the cosmic dawn in JADES-GS-z6-0, as the extinction bump arising from small graphite grains is too broad and peaks at wavelengths that are too short to be consistent with what is seen in JADES-GS-z6-0.

astro-ph.GA↗

arXiv2025

Feature Explosion: a generic optimization strategy for outlier detection algorithms

Outlier detection tasks aim at discovering potential issues or opportunities and are widely used in cybersecurity, financial security, industrial inspection, etc. To date, thousands of outlier detection algorithms have been proposed. Clearly, in real-world scenarios, such a large number of algorithms is unnecessary. In other words, a large number of outlier detection algorithms are redundant. We believe the root cause of this redundancy lies in the current highly customized (i.e., non-generic) optimization strategies. Specifically, when researchers seek to improve the performance of existing outlier detection algorithms, they have to design separate optimized versions tailored to the principles of each algorithm, leading to an ever-growing number of outlier detection algorithms. To address this issue, in this paper, we introduce the explosion from physics into the outlier detection task and propose a generic optimization strategy based on feature explosion, called OSD (Optimization Strategy for outlier Detection algorithms). In the future, when improving the performance of existing outlier detection algorithms, it will be sufficient to invoke the OSD plugin without the need to design customized optimized versions for them. We compared the performances of 14 outlier detection algorithms on 24 datasets before and after invoking the OSD plugin. The experimental results show that the performances of all outlier detection algorithms are improved on almost all datasets. In terms of average accuracy, OSD make these outlier detection algorithms improve by 15% (AUC), 63.7% (AP).

cs.LG↗

arXiv2025

Deep Learning Library Testing: Definition, Methods and Challenges

In recent years, software systems powered by deep learning (DL) techniques have significantly facilitated people's lives in many aspects. As the backbone of these DL systems, various DL libraries undertake the underlying optimization and computation. However, like traditional software, DL libraries are not immune to bugs, which can pose serious threats to users' personal property and safety. Studying the characteristics of DL libraries, their associated bugs, and the corresponding testing methods is crucial for enhancing the security of DL systems and advancing the widespread application of DL technology. This paper provides an overview of the testing research related to various DL libraries, discusses the strengths and weaknesses of existing methods, and provides guidance and reference for the application of the DL library. This paper first introduces the workflow of DL underlying libraries and the characteristics of three kinds of DL libraries involved, namely DL framework, DL compiler, and DL hardware library. It then provides definitions for DL underlying library bugs and testing. Additionally, this paper summarizes the existing testing methods and tools tailored to these DL libraries separately and analyzes their effectiveness and limitations. It also discusses the existing challenges of DL library testing and outlines potential directions for future research.

cs.SE↗

arXiv2025

Parallel Coding for Orthogonal Delay-Doppler Division Multiplexing

This paper proposes a novel parallel coding transmission strategy and an iterative detection and decoding receiver signal processing technique for orthogonal delay-Doppler division multiplexing (ODDM) modulation. Specifically, the proposed approach employs a parallel channel encoding (PCE) scheme that consists of multiple short-length codewords for each delay-Doppler multicarrier (DDMC) symbol. Building upon such a PCE transmission framework, we then introduce an iterative detection and decoding algorithm incorporating a successive decoding feedback (SDF) technique, which enables instant information exchange between the detector and decoder for each DDMC symbol. To characterize the error performance of the proposed scheme, we perform density evolution analysis considering the finite blocklength effects. Our analysis results, coupled with extensive simulations, demonstrate that the proposed PCE scheme with the SDF algorithm not only showcases a better overall performance but also requires much less decoding complexity to implement, compared to the conventional benchmark scheme that relies on a single long channel code for coding the entire ODDM frame.

cs.IT↗

arXiv2025

Solid-state Synapse Based on Magnetoelectrically Coupled Memristor

Brain-inspired computing architectures attempt to emulate the computations performed in the neurons and the synapses in human brain. Memristors with continuously tunable resistances are ideal building blocks for artificial synapses. Through investigating the memristor behaviors in a La0.7Sr0.3MnO3/BaTiO3/La0.7Sr0.3MnO3 multiferroic tunnel junction, it was found that the ferroelectric domain dynamics characteristics are influenced by the relative magnetization alignment of the electrodes, and the interfacial spin polarization is manipulated continuously by ferroelectric domain reversal, enriching our understanding of the magnetoelectric coupling fundamentally. This creates a functionality that not only the resistance of the memristor but also the synaptic plasticity form can be further manipulated, as demonstrated by the spike-timing-dependent plasticity investigations. Density functional theory calculations are carried out to describe the obtained magnetoelectric coupling, which is probably related to the Mn-Ti intermixing at the interfaces. The multiple and controllable plasticity characteristic in a single artificial synapse, to resemble the synaptic morphological alteration property in a biological synapse, will be conducive to the development of artificial intelligence.

cond-mat.mtrl-sci↗

arXiv2025

The geometric impact of the quantum Hall interface on a cone

Recently, quantum Hall interface has become a popular subject of research; distinct from that of the quantum Hall edge, which is constrained by external background confinement, the interface has the freedom to move, likely towards a string-like state. In disk geometry, it was known that the interface energy has an extra correction due to its curvature which depends on the size of the disk. In this work, we analytically calculate the energy of the integer quantum Hall interface on a cone surface which has the advantage that its curvature is more easily adjustable. By tuning the length and curvature of the interface by the cone angle parameter $β$, we analyze the dependence of the quantum Hall interface energy on the curvature and verify this geometric correction. Moreover, we find that the tip of the cone geometry has an extra contribution to the energy that reflects on the $u_2,u_4$ term.

cond-mat.str-el↗

arXiv2025

Towards Robust Multi-tab Website Fingerprinting

Website fingerprinting enables an eavesdropper to determine which websites a user is visiting over an encrypted connection. State-of-the-art website fingerprinting (WF) attacks have demonstrated effectiveness even against Tor-protected network traffic. However, existing WF attacks have critical limitations on accurately identifying websites in multi-tab browsing sessions, where the holistic pattern of individual websites is no longer preserved, and the number of tabs opened by a client is unknown a priori. In this paper, we propose ARES, a novel WF framework natively designed for multi-tab WF attacks. ARES formulates the multi-tab attack as a multi-label classification problem and solves it using the novel Transformer-based models. Specifically, ARES extracts local patterns based on multi-level traffic aggregation features and utilizes the improved self-attention mechanism to analyze the correlations between these local patterns, effectively identifying websites. We implement a prototype of ARES and extensively evaluate its effectiveness using our large-scale datasets collected over multiple months. The experimental results illustrate that ARES achieves optimal performance in several realistic scenarios. Further, ARES remains robust even against various WF defenses.

cs.CR↗

arXiv2025

Multiscale simulation of rarefied polyatomic gas flow via DIG method

A novel multiscale numerical method is developed to accelerate direct simulation Monte Carlo (DSMC) simulations for polyatomic gases with internal energy. This approach applies the general synthetic iterative scheme to stochastic simulations, removing the inherent restrictions on spatial grid size in DSMC and boosting the evolution of the particle distribution towards a steady state. Firstly, the proposed method intermittently couples the standard DSMC solver with the solution of steady-state macroscopic synthetic equations derived from the kinetic equation for polyatomic gases. These synthetic equations, encompassing accurate constitutive relations and relaxation rates, ensure applicability across the entire flow regime. Secondly, the particle distribution within the DSMC framework is adjusted to align with the solution of the macroscopic equations, significantly accelerating convergence to the steady-state solution. Finally, an adaptive treatment is implemented for the constitutive relations. Higher-order terms extracted from DSMC are applied only in rarefied regions where the local Knudsen number exceeds a predefined threshold. The fast convergence and asymptotic-preserving properties of the proposed method are demonstrated through several numerical simulations, in which it achieves remarkable speedups compared to standard DSMC, particularly in the near-continuum regime.

physics.comp-ph↗

arXiv2025

Edge-Dependent Step-Flow Growth Mechanism in $β$-Ga$_{2}$O$_{3}$ (100) Facet at the Atomic Level

Homoepitaxial step-flow growth of high-quality $β$-Ga$_{2}$O$_{3}$ thin films is essential for the advancement of high-performance Ga$_{2}$O$_{3}$-based devices. In this work, the step-flow growth mechanism of $β$-Ga$_{2}$O$_{3}$ (100) facet is explored by machine-learning molecular dynamics simulations and density functional theory calculations. Our results reveal that Ga adatoms and Ga-O adatom pairs, with their high mobility, are the primary atomic species responsible for efficient surface migration on the (100) facet. The asymmetric monoclinic structure of $β$-Ga$_{2}$O$_{3}$ induces a distinct two-stage Ehrlich-Schwoebel barrier for Ga adatoms at the [00$\overline{1}$] step edge, contributing to the suppression of double-step and hillock formation. Furthermore, a miscut towards [00$\overline{1}$] does not induce the nucleation of stable twin boundaries, whereas a miscut towards [001] leads to the spontaneous formation of twin boundaries. This research provides meaningful insights not only for high-quality $β$-Ga$_{2}$O$_{3}$ homoepitaxy but also the step-flow growth mechanism of other similar systems.

cond-mat.mtrl-sci↗

arXiv2025

OblivCDN: A Practical Privacy-preserving CDN with Oblivious Content Access

Content providers increasingly utilise Content Delivery Networks (CDNs) to enhance users' content download experience. However, this deployment scenario raises significant security concerns regarding content confidentiality and user privacy due to the involvement of third-party providers. Prior proposals using private information retrieval (PIR) and oblivious RAM (ORAM) have proven impractical due to high computation and communication costs, as well as integration challenges within distributed CDN architectures. In response, we present \textsf{OblivCDN}, a practical privacy-preserving system meticulously designed for seamless integration with the existing real-world Internet-CDN infrastructure. Our design strategically adapts Range ORAM primitives to optimise memory and disk seeks when accessing contiguous blocks of CDN content, both at the origin and edge servers, while preserving both content confidentiality and user access pattern hiding features. Also, we carefully customise several oblivious building blocks that integrate the distributed trust model into the ORAM client, thereby eliminating the computational bottleneck in the origin server and reducing communication costs between the origin server and edge servers. Moreover, the newly-designed ORAM client also eliminates the need for trusted hardware on edge servers, and thus significantly ameliorates the compatibility towards networks with massive legacy devices.In real-world streaming evaluations, OblivCDN} demonstrates remarkable performance, downloading a $256$ MB video in just $5.6$ seconds. This achievement represents a speedup of $90\times$ compared to a strawman approach (direct ORAM adoption) and a $366\times$ improvement over the prior art, OblivP2P.

cs.CR↗

arXiv2025

MBA-RAG: a Bandit Approach for Adaptive Retrieval-Augmented Generation through Question Complexity

Retrieval Augmented Generation (RAG) has proven to be highly effective in boosting the generative performance of language model in knowledge-intensive tasks. However, existing RAG framework either indiscriminately perform retrieval or rely on rigid single-class classifiers to select retrieval methods, leading to inefficiencies and suboptimal performance across queries of varying complexity. To address these challenges, we propose a reinforcement learning-based framework that dynamically selects the most suitable retrieval strategy based on query complexity. % our solution Our approach leverages a multi-armed bandit algorithm, which treats each retrieval method as a distinct ``arm'' and adapts the selection process by balancing exploration and exploitation. Additionally, we introduce a dynamic reward function that balances accuracy and efficiency, penalizing methods that require more retrieval steps, even if they lead to a correct result. Our method achieves new state of the art results on multiple single-hop and multi-hop datasets while reducing retrieval costs. Our code are available at https://github.com/FUTUREEEEEE/MBA .

cs.AI↗

arXiv2025

Diversity Optimization for Travelling Salesman Problem via Deep Reinforcement Learning

Existing neural methods for the Travelling Salesman Problem (TSP) mostly aim at finding a single optimal solution. To discover diverse yet high-quality solutions for Multi-Solution TSP (MSTSP), we propose a novel deep reinforcement learning based neural solver, which is primarily featured by an encoder-decoder structured policy. Concretely, on the one hand, a Relativization Filter (RF) is designed to enhance the robustness of the encoder to affine transformations of the instances, so as to potentially improve the quality of the found solutions. On the other hand, a Multi-Attentive Adaptive Active Search (MA3S) is tailored to allow the decoders to strike a balance between the optimality and diversity. Experimental evaluations on benchmark instances demonstrate the superiority of our method over recent neural baselines across different metrics, and its competitive performance against state-of-the-art traditional heuristics with significantly reduced computational time, ranging from $1.3\times$ to $15\times$ faster. Furthermore, we demonstrate that our method can also be applied to the Capacitated Vehicle Routing Problem (CVRP).

cs.LG↗

arXiv2024

Impact of Cognitive Load on Human Trust in Hybrid Human-Robot Collaboration

Human trust plays a crucial role in the effectiveness of human-robot collaboration. Despite its significance, the development and maintenance of an optimal trust level are obstructed by the complex nature of influencing factors and their mechanisms. This study investigates the effects of cognitive load on human trust within the context of a hybrid human-robot collaboration task. An experiment is conducted where the humans and the robot, acting as team members, collaboratively construct pyramids with differentiated levels of task complexity. Our findings reveal that cognitive load exerts diverse impacts on human trust in the robot. Notably, there is an increase in human trust under conditions of high cognitive load. Furthermore, the rewards for performance are substantially higher in tasks with high cognitive load compared to those with low cognitive load, and a significant correlation exists between human trust and the failure risk of performance in tasks with low and medium cognitive load. By integrating interdependent task steps, this research emphasizes the unique dynamics of hybrid human-robot collaboration scenarios. The insights gained not only contribute to understanding how cognitive load influences trust but also assist developers in optimizing collaborative target selection and designing more effective human-robot interfaces in such environments.

cs.RO↗