arXiv subjects

Di Wu

Publications and source records attributed to Di Wu.

At least 325 records · Page 18Linked to original sources

arXiv2025

Generative Multi-Agent Collaboration in Embodied AI: A Systematic Review

Embodied multi-agent systems (EMAS) have attracted growing attention for their potential to address complex, real-world challenges in areas such as logistics and robotics. Recent advances in foundation models pave the way for generative agents capable of richer communication and adaptive problem-solving. This survey provides a systematic examination of how EMAS can benefit from these generative capabilities. We propose a taxonomy that categorizes EMAS by system architectures and embodiment modalities, emphasizing how collaboration spans both physical and virtual contexts. Central building blocks, perception, planning, communication, and feedback, are then analyzed to illustrate how generative techniques bolster system robustness and flexibility. Through concrete examples, we demonstrate the transformative effects of integrating foundation models into embodied, multi-agent frameworks. Finally, we discuss challenges and future directions, underlining the significant promise of EMAS to reshape the landscape of AI-driven collaboration.

cs.MA↗

arXiv2025

BRIEF: Bridging Retrieval and Inference for Multi-hop Reasoning via Compression

Retrieval-augmented generation (RAG) can supplement large language models (LLMs) by integrating external knowledge. However, as the number of retrieved documents increases, the input length to LLMs grows linearly, causing a dramatic increase in latency and a degradation in long-context understanding. This is particularly serious for multi-hop questions that require a chain of reasoning across documents. To accelerate inference, reduce costs, and minimize distractions, this paper presents BRIEF (Bridging Retrieval and Inference through Evidence Fusion), a lightweight approach that performs query-aware multi-hop reasoning by compressing retrieved documents into highly dense textual summaries to integrate into in-context RAG. To enable learning compression for multi-hop reasoning, we curate synthetic data by extracting atomic propositions that encapsulate distinct factoids from the source documents to compose synthetic summaries. Based on our synthetic data built entirely by open-source models, BRIEF generates more concise summaries and enables a range of LLMs to achieve exceptional open-domain question answering (QA) performance. For example, on HotpotQA, BRIEF improves the compression rate by 2 times compared to the state-of-the-art baseline, while outperforming it by 3.00% EM and 4.16% F1 with Flan-UL2 as the reader model. It also generates more concise summaries than proprietary GPT-3.5, while demonstrating nearly identical QA performance.

cs.CL↗

arXiv2025

IU4Rec: Interest Unit-Based Product Organization and Recommendation for E-Commerce Platform

Most recommendation systems typically follow a product-based paradigm utilizing user-product interactions to identify the most engaging items for users. However, this product-based paradigm has notable drawbacks for Xianyu~\footnote{Xianyu is China's largest online C2C e-commerce platform where a large portion of the product are post by individual sellers}. Most of the product on Xianyu posted from individual sellers often have limited stock available for distribution, and once the product is sold, it's no longer available for distribution. This result in most items distributed product on Xianyu having relatively few interactions, affecting the effectiveness of traditional recommendation depending on accumulating user-item interactions. To address these issues, we introduce \textbf{IU4Rec}, an \textbf{I}nterest \textbf{U}nit-based two-stage \textbf{Rec}ommendation system framework. We first group products into clusters based on attributes such as category, image, and semantics. These IUs are then integrated into the Recommendation system, delivering both product and technological innovations. IU4Rec begins by grouping products into clusters based on attributes such as category, image, and semantics, forming Interest Units (IUs). Then we redesign the recommendation process into two stages. In the first stage, the focus is on recommend these Interest Units, capturing broad-level interests. In the second stage, it guides users to find the best option among similar products within the selected Interest Unit. User-IU interactions are incorporated into our ranking models, offering the advantage of more persistent IU behaviors compared to item-specific interactions. Experimental results on the production dataset and online A/B testing demonstrate the effectiveness and superiority of our proposed IU-centric recommendation approach.

cs.IR↗

arXiv2025

Beyond Dataset Watermarking: Model-Level Copyright Protection for Code Summarization Models

Code Summarization Model (CSM) has been widely used in code production, such as online and web programming for PHP and Javascript. CSMs are essential tools in code production, enhancing software development efficiency and driving innovation in automated code analysis. However, CSMs face risks of exploitation by unauthorized users, particularly in an online environment where CSMs can be easily shared and disseminated. To address these risks, digital watermarks offer a promising solution by embedding imperceptible signatures within the models to assert copyright ownership and track unauthorized usage. Traditional watermarking for CSM copyright protection faces two main challenges: 1) dataset watermarking methods require separate design of triggers and watermark features based on the characteristics of different programming languages, which not only increases the computation complexity but also leads to a lack of generalization, 2) existing watermarks based on code style transformation are easily identifiable by automated detection, demonstrating poor concealment. To tackle these issues, we propose ModMark , a novel model-level digital watermark embedding method. Specifically, by fine-tuning the tokenizer, ModMark achieves cross-language generalization while reducing the complexity of watermark design. Moreover, we employ code noise injection techniques to effectively prevent trigger detection. Experimental results show that our method can achieve 100% watermark verification rate across various programming languages' CSMs, and the concealment and effectiveness of ModMark can also be guaranteed.

cs.CR↗

arXiv2025

How to Learn in a Noisy World? Self-Correcting the Real-World Data Noise in Machine Translation

The massive amounts of web-mined parallel data contain large amounts of noise. Semantic misalignment, as the primary source of the noise, poses a challenge for training machine translation systems. In this paper, we first introduce a process for simulating misalignment controlled by semantic similarity, which closely resembles misaligned sentences in real-world web-crawled corpora. Under our simulated misalignment noise settings, we quantitatively analyze its impact on machine translation and demonstrate the limited effectiveness of widely used pre-filters for noise detection. This underscores the necessity of more fine-grained ways to handle hard-to-detect misalignment noise. With an observation of the increasing reliability of the model's self-knowledge for distinguishing misaligned and clean data at the token level, we propose self-correction, an approach that gradually increases trust in the model's self-knowledge to correct the training supervision. Comprehensive experiments show that our method significantly improves translation performance both in the presence of simulated misalignment noise and when applied to real-world, noisy web-mined datasets, across a range of translation tasks.

cs.CL↗

arXiv2025

MXMap: A Multivariate Cross Mapping Framework for Causal Discovery in Dynamical Systems

Convergent Cross Mapping (CCM) is a powerful method for detecting causality in coupled nonlinear dynamical systems, providing a model-free approach to capture dynamic causal interactions. Partial Cross Mapping (PCM) was introduced as an extension of CCM to address indirect causality in three-variable systems by comparing cross-mapping quality between direct cause-effect mapping and indirect mapping through an intermediate conditioning variable. However, PCM remains limited to univariate delay embeddings in its cross-mapping processes. In this work, we extend PCM to the multivariate setting, introducing multiPCM, which leverages multivariate embeddings to more effectively distinguish indirect causal relationships. We further propose a multivariate cross-mapping framework (MXMap) for causal discovery in dynamical systems. This two-phase framework combines (1) pairwise CCM tests to establish an initial causal graph and (2) multiPCM to refine the graph by pruning indirect causal connections. Through experiments on simulated data and the ERA5 Reanalysis weather dataset, we demonstrate the effectiveness of MXMap. Additionally, MXMap is compared against several baseline methods, showing advantages in accuracy and causal graph refinement.

cs.LG↗

arXiv2025

A Systematic Literature Review on Explainability for Machine/Deep Learning-based Software Engineering Research

The remarkable achievements of Artificial Intelligence (AI) algorithms, particularly in Machine Learning (ML) and Deep Learning (DL), have fueled their extensive deployment across multiple sectors, including Software Engineering (SE). However, due to their black-box nature, these promising AI-driven SE models are still far from being deployed in practice. This lack of explainability poses unwanted risks for their applications in critical tasks, such as vulnerability detection, where decision-making transparency is of paramount importance. This paper endeavors to elucidate this interdisciplinary domain by presenting a systematic literature review of approaches that aim to improve the explainability of AI models within the context of SE. The review canvasses work appearing in the most prominent SE & AI conferences and journals, and spans 108 papers across 23 unique SE tasks. Based on three key Research Questions (RQs), we aim to (1) summarize the SE tasks where XAI techniques have shown success to date; (2) classify and analyze different XAI techniques; and (3) investigate existing evaluation approaches. Based on our findings, we identified a set of challenges remaining to be addressed in existing studies, together with a set of guidelines highlighting potential opportunities we deemed appropriate and important for future work.

cs.SE↗

arXiv2025

Tollmien-Schlichting waves near neutral stable curve

In this paper, we study the linear stability of boundary layer flows over a flat plate. Tollmien, Schlichting, Lin et al. found that there exists a neutral curve, which consists of two branches: lower branch $α_{low}(Re)$ and upper branch $α_{up}(Re)$. Here, $α$ is the wave number and $Re$ is the Reynolds number. For any $α\in(α_{low},α_{up})$, there exist unstable modes known as Tollmien-Schlichting (T-S) waves to the linearized Navier-Stokes system. These waves play a key role during the early stage of boundary layer transition. In a breakthrough work (Duke math Jour, 165(2016)), Grenier, Guo, and Nguyen provided a rigorous construction of the unstable T-S waves. In this paper, we confirm the existence of the neutral stable curve. To achieve this, we develop a more delicate method for solving the Orr-Sommerfeld equation by borrowing some ideas from the triple-deck theory. This approach allows us to construct the T-S waves in a neighborhood of the neutral curve.

math.AP↗

arXiv2025

Light rings and shadows of static black holes in effective quantum gravity II: A new solution without Cauchy horizons

Among the three known types of static solutions proposed within the Hamiltonian constraint approach to effective quantum gravity (EQG), the first two have been extensively investigated, whereas the third type-which preserves general covariance, is free of Cauchy horizons, and was only recently obtained-remains relatively unexplored. This solution can describe a black hole with an event horizon for certain parameter ranges, or a horizonless compact object beyond those ranges. In this paper, we focus on the third type and show that its light rings feature both stable and unstable branches, and that the black hole shadow size grows with the quantum parameter-unlike in the first two types. However, when we account for both the shadow and the lensing ring, the overall behavior closely resembles that of the second type, in which an increasing quantum parameter leads to a larger portion of the lensing ring being occupied by the shadow. This feature can serve as a hallmark of black holes in EQG, offering a potential way to distinguish them from their GR counterparts. Remarkably, the parameter ranges under which the solution remains a black hole are highly consistent with the current observational constraints on black hole shadows, lending strong support to the classification of the third type of compact object in EQG as a black hole endowed with an event horizon.

gr-qc↗

arXiv2025

Fine-Tuned Language Models as Space Systems Controllers

Large language models (LLMs), or foundation models (FMs), are pretrained transformers that coherently complete sentences auto-regressively. In this paper, we show that LLMs can control simplified space systems after some additional training, called fine-tuning. We look at relatively small language models, ranging between 7 and 13 billion parameters. We focus on four problems: a three-dimensional spring toy problem, low-thrust orbit transfer, low-thrust cislunar control, and powered descent guidance. The fine-tuned LLMs are capable of controlling systems by generating sufficiently accurate outputs that are multi-dimensional vectors with up to 10 significant digits. We show that for several problems the amount of data required to perform fine-tuning is smaller than what is generally required of traditional deep neural networks (DNNs), and that fine-tuned LLMs are good at generalizing outside of the training dataset. Further, the same LLM can be fine-tuned with data from different problems, with only minor performance degradation with respect to LLMs trained for a single application. This work is intended as a first step towards the development of a general space systems controller.

cs.LG↗

arXiv2025

An Aspect Performance-aware Hypergraph Neural Network for Review-based Recommendation

Online reviews allow consumers to provide detailed feedback on various aspects of items. Existing methods utilize these aspects to model users' fine-grained preferences for specific item features through graph neural networks. We argue that the performance of items on different aspects is important for making precise recommendations, which has not been taken into account by existing approaches, due to lack of data. In this paper, we propose an aspect performance-aware hypergraph neural network (APH) for the review-based recommendation, which learns the performance of items from the conflicting sentiment polarity of user reviews. Specifically, APH comprehensively models the relationships among users, items, aspects, and sentiment polarity by systematically constructing an aspect hypergraph based on user reviews. In addition, APH aggregates aspects representing users and items by employing an aspect performance-aware hypergraph aggregation method. It aggregates the sentiment polarities from multiple users by jointly considering user preferences and the semantics of their sentiments, determining the weights of sentiment polarities to infer the performance of items on various aspects. Such performances are then used as weights to aggregate neighboring aspects. Experiments on six real-world datasets demonstrate that APH improves MSE, Precision@5, and Recall@5 by an average of 2.30%, 4.89%, and 1.60% over the best baseline. The source code and data are available at https://github.com/dianziliu/APH.

cs.IR↗

arXiv2025

OnionEval: An Unified Evaluation of Fact-conflicting Hallucination for Small-Large Language Models

Large Language Models (LLMs) are highly capable but require significant computational resources for both training and inference. Within the LLM family, smaller models (those with fewer than 10 billion parameters) also perform well across various tasks. However, these smaller models share similar limitations to their larger counterparts, including the tendency to hallucinate. Despite the existence of many benchmarks to evaluate hallucination in LLMs, few have specifically focused on small LLMs (SLLMs). Additionally, SLLMs show widely varying performance across different benchmarks. In this paper, we introduce OnionEval, a multi-layer structured framework with a specific metric called the context-influence score (CI), designed to effectively assess the fact-conflicting hallucination tendencies of small LLMs across different contextual levels. Our experimental results reveal a key feature of SLLMs: they excel in factual analysis but face challenges with context reasoning. Further investigation shows that a simple Chain-of-Thought strategy can significantly reduce these limitations, improving the practical usefulness of SLLMs in real-world applications.

cs.CL↗

arXiv2025

NoteLLM-2: Multimodal Large Representation Models for Recommendation

Large Language Models (LLMs) have demonstrated exceptional proficiency in text understanding and embedding tasks. However, their potential in multimodal representation, particularly for item-to-item (I2I) recommendations, remains underexplored. While leveraging existing Multimodal Large Language Models (MLLMs) for such tasks is promising, challenges arise due to their delayed release compared to corresponding LLMs and the inefficiency in representation tasks. To address these issues, we propose an end-to-end fine-tuning method that customizes the integration of any existing LLMs and vision encoders for efficient multimodal representation. Preliminary experiments revealed that fine-tuned LLMs often neglect image content. To counteract this, we propose NoteLLM-2, a novel framework that enhances visual information. Specifically, we propose two approaches: first, a prompt-based method that segregates visual and textual content, employing a multimodal In-Context Learning strategy to balance focus across modalities; second, a late fusion technique that directly integrates visual information into the final representations. Extensive experiments, both online and offline, demonstrate the effectiveness of our approach. Code is available at https://github.com/Applied-Machine-Learning-Lab/NoteLLM.

cs.IR↗

arXiv2025

Institutional Adoption and Correlation Dynamics: Bitcoin's Evolving Role in Financial Markets

Bitcoin, widely recognized as the first cryptocurrency, has shown increasing integration with traditional financial markets, particularly major U.S. equity indices, amid accelerating institutional adoption. This study examines how Bitcoin exchange-traded funds and corporate Bitcoin holdings affect correlations with the Nasdaq 100 and the S&P 500, using rolling-window correlation, static correlation coefficients, and an event-study framework on daily data from 2018 to 2025.Correlation levels intensified following key institutional milestones, with peaks reaching 0.87 in 2024, and they vary across market regimes. These trends suggest that Bitcoin has transitioned from an alternative asset toward a more integrated financial instrument, carrying implications for portfolio diversification, risk management, and systemic stability. Future research should further investigate regulatory and macroeconomic factors shaping these evolving relationships.

q-fin.PM↗

arXiv2025

Study on a Fast Solver for Combined Field Integral Equations of 3D Conducting Bodies Based on Graph Neural Networks

In this paper, we present a graph neural networks (GNNs)-based fast solver (GraphSolver) for solving combined field integral equations (CFIEs) of 3D conducting bodies. Rao-Wilton-Glisson (RWG) basis functions are employed to discretely and accurately represent the geometry of 3D conducting bodies. A concise and informative graph representation is then constructed by treating each RWG function as a node in the graph, enabling the flow of current between nodes. With the transformed graphs, GraphSolver is developed to directly predict real and imaginary parts of the x, y and z components of the surface current densities at each node (RWG function). Numerical results demonstrate the efficacy of GraphSolver in solving CFIEs for 3D conducting bodies with varying levels of geometric complexity, including basic 3D targets, missile-shaped targets, and airplane-shaped targets.

cs.LG↗

arXiv2025

Nonreciprocal Optical Routing in Multi-port Magneto-Optical Devices on Silicon

Nonreciprocal optical devices are key components in photonic integrated circuits for light reflection blocking and routing. Most reported silicon integrated nonreciprocal optical devices to date were unit devices. To allow complex signal routing between multi-ports in photonic networks, multi-port magneto-optical (MO) nonreciprocal photonic devices are desired. In this study, we report experimental demonstration of a silicon integrated 5*5 multiport nonreciprocal photonic device based on magneto-optical waveguides. By introducing different nonreciprocal phase shift effect to planar photonic waveguides, the device focuses light to different ports for both forward and backward propagation. The device shows designable nonreciprocal transmission between 5*5 ports, achieving 16 dB isolation ratio and -18 dB crosstalk.

physics.optics↗

arXiv2025

ARNet: Self-Supervised FG-SBIR with Unified Sample Feature Alignment and Multi-Scale Token Recycling

Fine-Grained Sketch-Based Image Retrieval (FG-SBIR) aims to minimize the distance between sketches and corresponding images in the embedding space. However, scalability is hindered by the growing complexity of solutions, mainly due to the abstract nature of fine-grained sketches. In this paper, we propose an effective approach to narrow the gap between the two domains. It mainly facilitates unified mutual information sharing both intra- and inter-samples, rather than treating them as a single feature alignment problem between modalities. Specifically, our approach includes: (i) Employing dual weight-sharing networks to optimize alignment within the sketch and image domain, which also effectively mitigates model learning saturation issues. (ii) Introducing an objective optimization function based on contrastive loss to enhance the model's ability to align features in both intra- and inter-samples. (iii) Presenting a self-supervised Multi-Scale Token Recycling (MSTR) Module featured by recycling discarded patch tokens in multi-scale features, further enhancing representation capability and retrieval performance. Our framework achieves excellent results on CNN- and ViT-based backbones. Extensive experiments demonstrate its superiority over existing methods. We also introduce Cloths-V1, the first professional fashion sketch-image dataset, utilized to validate our method and will be beneficial for other applications.

cs.CV↗

arXiv2024

Liouville theorems for harmonic 1-forms on gradient Ricci solitons

We prove that there is no nontrivial $L^2$-integrable harmonic 1-form on noncompact complete gradient steady Ricci solitons or noncompact complete gradient shrinking Kähler-Ricci solitons. As an application, it can be used to distinguish certain flat vector bundles that arise from fundamental group representations into $SL(r,\mathbb{C})$.

math.DG↗