arXiv subjects

Yi Du

Publications and source records attributed to Yi Du.

At least 37 records · Page 2Linked to original sources

arXiv2026

BadBlocks: Low-Cost and Stealthy Backdoor Attacks Tailored for Text-to-Image Diffusion Models

Despite the remarkable progress of diffusion models in image generation, recent studies reveal their vulnerability to backdoor attacks via covert visual or textual triggers. Although evolving defense mechanisms can detect most existing threats through visual inspection or feature analysis, we introduce BadBlocks-a novel, lightweight, and highly covert attack that challenges these safeguards. By selectively poisoning specific blocks within the UNet architecture while keeping other components intact, BadBlocks requires only 30% of the computational resources and 20% of the GPU time of conventional attacks, effectively democratizing backdoor injection on consumer-grade GPUs. Empirical evaluations demonstrate that BadBlocks achieves a high attack success rate with negligible perceptual quality loss, while successfully bypassing state-of-the-art defenses, particularly attention-based detection frameworks. Layer-level ablation studies further confirm that backdoor mapping does not require full-network fine-tuning, revealing the disparate vulnerability of different neural layers. Overall, BadBlocks significantly lowers the barrier for executing backdoor attacks, presenting a critical security risk. Our code is available at: https://github.com/paoche11/BadBlocks.

cs.CR↗

arXiv2026

G-DRAGON: Geospatial Reasoning and Dynamic Planning for Retrieval-Augmented Outdoor Navigation

Autonomous ground robots operating in large-scale outdoor environments require both robust long-range navigation and fine-grained ''last-mile'' exploration. Current advances in visual-language navigation (VLN) work well at short-range tasks, lacking geospatial grounding for long-distance missions. Some OpenStreetMap (OSM)-based methods relying on cloud-based Large Language Models (LLMs) are prone to factual hallucination and cannot conduct ''last-mile'' exploration based on human instruction. To address these challenges, we present G-DRAGON, a retrieval-augmented framework for outdoor, open-world navigation. This framework maps natural-language commands to versioned, local OSM entities via generative retrieval based on lightweight LLM, yielding accurate coordinates for global route planning. A high-level planning module bridges global topological routes with the SLAM system, projecting geospatial waypoints into the robot's navigable frame. For the ''last mile," the framework transitions to frontier-based exploration and open-set semantic voxel mapping to localize open-vocabulary targets. Experimental results in simulation demonstrate our framework outperforms state-of-the-art baselines. Furthermore, we validate the system in unseen real-world urban environments on an Unmanned Ground Vehicle (UGV), successfully completing person-search missions with trajectories of up to 500m.

cs.RO↗

arXiv2026

FOCAL-Attention for Heterogeneous Multi-Label Prediction

Heterogeneous graphs have attracted increasing attention for modeling multi-typed entities and relations in complex real-world systems. Multi-label node classification on heterogeneous graphs is challenging due to structural heterogeneity and the need to learn shared representations across multiple labels. Existing methods typically adopt either flexible attention mechanisms or meta-path constrained anchoring, but in heterogeneous multi-label prediction they often suffer from semantic dilution or coverage constraint. Both issues are further amplified under multi-label supervision. We present a theoretical analysis showing that as heterogeneous neighborhoods expand, the attention mass allocated to task-critical (primary) neighborhoods diminishes, and that meta-path constrained aggregation exhibits a dilemma: too few meta-paths intensify coverage constraint, while too many re-introduce dilution. To resolve this coverage-anchoring conflict, we propose FOCAL: Fusion Of Coverage and Anchoring Learning, with two components: coverage-oriented attention (COA) for flexible, unconstrained heterogeneous context aggregation, and anchoring-oriented attention (AOA) that restricts aggregation to meta-path-induced primary semantics. Our theoretical analysis and experimental results further indicates that FOCAL has a better performance than other state-of-the-art methods.

cs.LG↗

arXiv2026

Beyond Paper-to-Paper: Structured Profiling and Rubric Scoring for Paper-Reviewer Matching

As conference submission volumes continue to grow, accurately recommending suitable reviewers has become a challenge. Most existing methods follow a ``Paper-to-Paper'' matching paradigm, implicitly representing a reviewer by their publication history. However, effective reviewer matching requires capturing multi-dimensional expertise, and textual similarity to past papers alone is often insufficient. To address this gap, we propose P2R, a training-free framework that shifts from implicit paper-to-paper matching to explicit profile-based matching. P2R uses general-purpose LLMs to construct structured profiles for both submissions and reviewers, disentangling them into Topics, Methodologies, and Applications. Building on these profiles, P2R adopts a coarse-to-fine pipeline to balance efficiency and depth. It first performs hybrid retrieval that combines semantic and aspect-level signals to form a high-recall candidate pool, and then applies an LLM-based committee to evaluate candidates under strict rubrics, integrating both multi-dimensional expert views and a holistic Area Chair perspective. Experiments on NeurIPS, SIGIR, and SciRepEval show that P2R consistently outperforms state-of-the-art baselines. Ablation studies further verify the necessity of each component. Overall, P2R highlights the value of explicit, structured expertise modeling and offers practical guidance for applying LLMs to reviewer matching.

cs.IR↗

arXiv2026

GSMem: 3D Gaussian Splatting as Persistent Spatial Memory for Zero-Shot Embodied Exploration and Reasoning

Effective embodied exploration requires agents to accumulate and retain spatial knowledge over time. However, existing scene representations, such as discrete scene graphs or static view-based snapshots, lack \textit{post-hoc re-observability}. If an initial observation misses a target, the resulting memory omission is often irrecoverable. To bridge this gap, we propose \textbf{GSMem}, a zero-shot embodied exploration and reasoning framework built upon 3D Gaussian Splatting (3DGS). By explicitly parameterizing continuous geometry and dense appearance, 3DGS serves as a persistent spatial memory that endows the agent with \textit{Spatial Recollection}: the ability to render photorealistic novel views from optimal, previously unoccupied viewpoints. To operationalize this, GSMem employs a retrieval mechanism that simultaneously leverages parallel object-level scene graphs and semantic-level language fields. This complementary design robustly localizes target regions, enabling the agent to ``hallucinate'' optimal views for high-fidelity Vision-Language Model (VLM) reasoning. Furthermore, we introduce a hybrid exploration strategy that combines VLM-driven semantic scoring with a 3DGS-based coverage objective, balancing task-aware exploration with geometric coverage. Extensive experiments on embodied question answering and lifelong navigation demonstrate the robustness and effectiveness of our framework

cs.CV↗

arXiv2026

Vision to Geometry: 3D Spatial Memory for Sequential Embodied MLLM Reasoning and Exploration

Embodied agents are expected to assist humans by actively exploring unknown environments and reasoning about spatial contexts. When deployed in real life, agents often face sequential tasks where each new task follows the completion of the previous one and may include infeasible objectives, such as searching for non-existent objects. However, most existing research focuses on isolated goals, overlooking the core challenge of sequential tasks: the ability to reuse spatial knowledge accumulated from previous explorations to guide subsequent reasoning and exploration. In this work, we investigate this underexplored yet practically significant embodied AI challenge. Specifically, we propose 3DSPMR, a 3D SPatial Memory Reasoning framework that utilizes Field-of-View (FoV) coverage as an explicit geometric prior. By integrating FoV-based constraints, 3DSPMR significantly enhances an agent's memory, reasoning, and exploration capabilities across sequential tasks. To facilitate research in this area, we further introduce SEER-Bench, a novel Sequential Embodied Exploration and Reasoning Benchmark that spans two foundational tasks: Embodied Question Answering (EQA) and Embodied Multi-modal Navigation (EMN). SEER-Bench uniquely incorporates both feasible and infeasible tasks to provide a rigorous and comprehensive evaluation of agent performance. Extensive experiments verify that 3DSPMR achieves substantial performance gains on both sequential EQA and EMN tasks.

cs.CV↗

arXiv2026

Fast Task Planning with Neuro-Symbolic Relaxation

Real-world task planning requires long-horizon reasoning over large sets of objects with complex relationships and attributes, leading to a combinatorial explosion for classical symbolic planners. To prune the search space, recent methods prioritize searching on a simplified task only containing a few ``important" objects predicted by a neural network. However, such a simple neuro-symbolic (NeSy) integration risks omitting critical objects and wasting resources on unsolvable simplified tasks. To enable Fast and reliable planning, we introduce a NeSy relaxation strategy (Flax), combining neural importance prediction with symbolic expansion. Specifically, we first learn a graph neural network to predict object importance to create a simplified task and solve it with a symbolic planner. Then, we solve a rule-relaxed task to obtain a quick rough plan, and reintegrate all referenced objects into the simplified task to recover any overlooked but essential elements. Finally, we apply complementary rules to refine the updated task, keeping it both reliable and compact. Extensive experiments are conducted on both synthetic and real-world maze navigation benchmarks where a robot must traverse through a maze and interact with movable obstacles. The results show that Flax boosts the average success rate by 20.82\% and cuts mean wall-clock planning time by 17.65\% compared with the state-of-the-art NeSy baseline. We expect that Flax offers a practical path toward fast, scalable, long-horizon task planning in complex environments.

cs.RO↗

arXiv2025

Imperative Learning: A Self-supervised Neuro-Symbolic Learning Framework for Robot Autonomy

Data-driven methods such as reinforcement and imitation learning have achieved remarkable success in robot autonomy. However, their data-centric nature still hinders them from generalizing well to ever-changing environments. Moreover, labeling data for robotic tasks is often impractical and expensive. To overcome these challenges, we introduce a new self-supervised neuro-symbolic (NeSy) computational framework, imperative learning (IL), for robot autonomy, leveraging the generalization abilities of symbolic reasoning. The framework of IL consists of three primary components: a neural module, a reasoning engine, and a memory system. We formulate IL as a special bilevel optimization (BLO), which enables reciprocal learning over the three modules. This overcomes the label-intensive obstacles associated with data-driven approaches and takes advantage of symbolic reasoning concerning logical reasoning, physical principles, geometric analysis, etc. We discuss several optimization techniques for IL and verify their effectiveness in five distinct robot autonomy tasks including path planning, rule induction, optimal control, visual odometry, and multi-robot routing. Through various experiments, we show that IL can significantly enhance robot autonomy capabilities and we anticipate that it will catalyze further research across diverse domains.

cs.RO↗

arXiv2025

Geometry-dependent Ekman layer approximations on curved domains: L^{\infty} convergence

The Ekman boundary layer is a fundamental concept in fluid dynamics that describes fluid motion near boundaries affected by Earth's rotation. Most theoretical studies have simplified their analysis by assuming a planar boundary surface, resulting in limited exploration of structures with general smooth boundary conditions. Investigating the impact of boundary geometry in the Ekman boundary layer is essential, as initially suggested by J.L. Lions and further examined in Masmoudi's study [Comm. Pure Appl. Math. 53 (2000), 432-483] under small amplitude periodic boundary conditions. This paper clarifies how boundary geometry influences flow fields and characterizes its effects on near-boundary layer flow. We construct a class of multi-scale approximate solutions based on the boundary's geometric features and establish their convergence in the L^{\infty} framework. Our findings do not require a small-amplitude assumption, only an upper bound on the Gaussian curvature of the boundary surface. Notably, when the boundary is planar, our approach aligns with existing studies. Additionally, in the vanishing-viscosity limit, we derive a limiting-state system dependent on boundary geometric parameters. These contributions extend the theoretical understanding of boundary-layer interactions to general curved geometries and have possible applications in atmospheric, oceanic, and other geophysical flow contexts.

math-ph↗

arXiv2025

Coalescence of multiple topological orders in quasi-one-dimensional bismuth halide chains

Topology is being widely adopted to understand and to categorize quantum matter in modern physics. The nexus of topology orders, which engenders distinct quantum phases with benefits to both fundamental research and practical applications for future quantum devices, can be driven by topological phase transition through modulating intrinsic or extrinsic ordering parameters. The conjoined topology, however, is still elusive in experiments due to the lack of suitable material platforms. Here we use scanning tunneling microscopy, angle-resolved photoemission spectroscopy, and theoretical calculations to investigate the doping-driven band structure evolution of a quasi-one-dimensional material system, bismuth halide, which contains rare multiple band inversions in two time-reversal-invariant momenta. According to the unique bulk-boundary correspondence in topological matter, we unveil a composite topological phase, the coexistence of a strong topological phase and a high-order topological phase, evoked by the band inversion associated with topological phase transition in this system. Moreover, we reveal multiple-stage topological phase transitions by varying the halide element ratio: from high-order topology to weak topology, the unusual dual topology, and trivial/weak topology subsequently. Our results not only realize an ideal material platform with composite topology, but also provide an insightful pathway to establish abundant topological phases in the framework of band inversion theory.

cond-mat.mtrl-sci↗

arXiv2025

Field manipulation of Weyl modes in an ideal Dirac semimetal

The emergent Weyl modes with the broken time-reversal symmetry or inversion symmetry provide large Berry curvature and chirality to carriers, offering the realistic platforms to explore topology of electrons in three-dimensional systems. However, the reversal transition between different types of Weyl modes in a single material, which is of particular interest in the fundamental research in Weyl physics and potential application in spintronics, is scarcely achieved due to restriction of inborn symmetry in crystals. Here, by tuning the direction and strength of magnetic field in an ideal Dirac semimetal, Bi4(Br0.27I0.73)4, we report the realization of multiple Weyl modes, including gapped Weyl mode, Weyl nodal ring, and coupled Weyl mode by the magnetoresistivity measurements and electronic structure calculations. Specifically, under a magnetic field with broken mirror symmetry, anomalous Hall effect with step feature results from the large Berry curvature for the gapped Weyl mode. A prominent negative magnetoresistivity is observed at low magnetic field with preserved mirror symmetry and disappears at high magnetic field, which is correlated to the chiral anomaly and its annihilation of Weyl nodal ring, respectively. Our findings reveal distinct Weyl modes under the intertwined crystal symmetry and time-reversal breaking, laying the foundation of manipulating multiple Weyl modes in chiral spintronic network.

cond-mat.mes-hall↗

arXiv2025

Img2CAD: Reverse Engineering 3D CAD Models from Images through VLM-Assisted Conditional Factorization

Reverse engineering 3D computer-aided design (CAD) models from images is an important task for many downstream applications including interactive editing, manufacturing, architecture, robotics, etc. The difficulty of the task lies in vast representational disparities between the CAD output and the image input. CAD models are precise, programmatic constructs that involve sequential operations combining discrete command structure with continuous attributes, making it challenging to learn and optimize in an end-to-end fashion. Concurrently, input images introduce inherent challenges such as photometric variability and sensor noise, complicating the reverse engineering process. In this work, we introduce a novel approach that conditionally factorizes the task into two sub-problems. First, we leverage vision-language foundation models (VLMs), a finetuned Llama3.2, to predict the global discrete base structure with semantic information. Second, we propose TrAssembler that, conditioned on the discrete structure with semantics, predicts the continuous attribute values. To support the training of our TrAssembler, we further constructed an annotated CAD dataset of common objects from ShapeNet. Putting all together, our approach and data demonstrate significant first steps towards CAD-ifying images in the wild. Code and data can be found in https://github.com/qq456cvb/Img2CAD.

cs.CV↗

arXiv2025

Parasite: A Steganography-based Backdoor Attack Framework for Diffusion Models

Recently, the diffusion model has gained significant attention as one of the most successful image generation models, which can generate high-quality images by iteratively sampling noise. However, recent studies have shown that diffusion models are vulnerable to backdoor attacks, allowing attackers to enter input data containing triggers to activate the backdoor and generate their desired output. Existing backdoor attack methods primarily focused on target noise-to-image and text-to-image tasks, with limited work on backdoor attacks in image-to-image tasks. Furthermore, traditional backdoor attacks often rely on a single, conspicuous trigger to generate a fixed target image, lacking concealability and flexibility. To address these limitations, we propose a novel backdoor attack method called "Parasite" for image-to-image tasks in diffusion models, which not only is the first to leverage steganography for triggers hiding, but also allows attackers to embed the target content as a backdoor trigger to achieve a more flexible attack. "Parasite" as a novel attack method effectively bypasses existing detection frameworks to execute backdoor attacks. In our experiments, "Parasite" achieved a 0 percent backdoor detection rate against the mainstream defense frameworks. In addition, in the ablation study, we discuss the influence of different hiding coefficients on the attack results. You can find our code at https://anonymous.4open.science/r/Parasite-1715/.

cs.CV↗

arXiv2025

CITE: A Comprehensive Benchmark for Heterogeneous Text-Attributed Graphs on Catalytic Materials

Text-attributed graphs(TAGs) are pervasive in real-world systems,where each node carries its own textual features. In many cases these graphs are inherently heterogeneous, containing multiple node types and diverse edge types. Despite the ubiquity of such heterogeneous TAGs, there remains a lack of large-scale benchmark datasets. This shortage has become a critical bottleneck, hindering the development and fair comparison of representation learning methods on heterogeneous text-attributed graphs. In this paper, we introduce CITE - Catalytic Information Textual Entities Graph, the first and largest heterogeneous text-attributed citation graph benchmark for catalytic materials. CITE comprises over 438K nodes and 1.2M edges, spanning four relation types. In addition, we establish standardized evaluation procedures and conduct extensive benchmarking on the node classification task, as well as ablation experiments on the heterogeneous and textual properties of CITE. We compare four classes of learning paradigms, including homogeneous graph models, heterogeneous graph models, LLM(Large Language Model)-centric models, and LLM+Graph models. In a nutshell, we provide (i) an overview of the CITE dataset, (ii) standardized evaluation protocols, and (iii) baseline and ablation experiments across diverse modeling paradigms.

cs.LG↗

arXiv2025

Family Seiberg-Witten equation on Kahler surface and $π_i(\Symp)$ on multiple-point blow ups of Calabi-Yau surfaces

Let $ω$ be a Kahler form on $M$, which is a torus $T^4$, a $K3$ surface or an Enriques surface, let $M\#n\overline{\mathbb{CP}^2}$ be $n-$point Kahler blowup of $M$. Suppose that $κ=[ω]$ satisfies certain irrationality condition. Applying techniques related to deformation of complex objects, we extend the guage-theoretic invariant on closed Kahler suraces developed by Kronheimer\cite{Kronheimer1998} and Smirnov\cite{Smirnov2022}\cite{Smirnov2023}. As a result, we show that even dimensional higher homotopy groups of $\Symp(M\#n\overline{\mathbb{CP}^2},ω)$ are infinitely generated.

math.GT↗

arXiv2025

Disentangled Multi-span Evolutionary Network against Temporal Knowledge Graph Reasoning

Temporal Knowledge Graphs (TKGs), as an extension of static Knowledge Graphs (KGs), incorporate the temporal feature to express the transience of knowledge by describing when facts occur. TKG extrapolation aims to infer possible future facts based on known history, which has garnered significant attention in recent years. Some existing methods treat TKG as a sequence of independent subgraphs to model temporal evolution patterns, demonstrating impressive reasoning performance. However, they still have limitations: 1) In modeling subgraph semantic evolution, they usually neglect the internal structural interactions between subgraphs, which are actually crucial for encoding TKGs. 2) They overlook the potential smooth features that do not lead to semantic changes, which should be distinguished from the semantic evolution process. Therefore, we propose a novel Disentangled Multi-span Evolutionary Network (DiMNet) for TKG reasoning. Specifically, we design a multi-span evolution strategy that captures local neighbor features while perceiving historical neighbor semantic information, thus enabling internal interactions between subgraphs during the evolution process. To maximize the capture of semantic change patterns, we design a disentangle component that adaptively separates nodes' active and stable features, used to dynamically control the influence of historical semantics on future evolution. Extensive experiments conducted on four real-world TKG datasets show that DiMNet demonstrates substantial performance in TKG reasoning, and outperforms the state-of-the-art up to 22.7% in MRR.

cs.AI↗

arXiv2025

Emergence of Diverse Topological States in Ge Doped MnBi2Te4

As an ideal platform for studying interplays between symmetry, topology and magnetism, the magnetic topological insulator (MTI) MnBi2Te4 has attracted extensive attentions. However, its strong n-type intrinsic defects hinder the realizations of exotic phenomena. Stimulated by recent discoveries that Ge doping can efficiently tune the position of Fermi level, here we systematically investigate the band evolution and topological phase diagram with doping concentration from MTI MnBi2Te4 to strong topological insulator GeBi2Te4. Different from magnetically doped Bi2Se3, the topology here is determined by competition of two band inversions arising from band folding of two time-reversal invariant momenta between antiferromagnetic and nonmagnetic/ferromagnetic unit cells. By employing a band momentum mapping method, besides the known MTI phase, remarkably, we find two classes of magnetic Dirac semimetal phases at antiferromagnetic state, two classes of Weyl semimetal phases at ferromagnetic state, and an intermediate trivial state at different doping regions. Interestingly, the trivial state can be tuned into a Weyl phase with two coexisting band inversions and extraordinarily long Fermi arcs by a small strain. Our work reveals diverse topological states with intrinsic quantum phenomena can be achieved with great potential for designing future electronic devices.

cond-mat.mtrl-sci↗

arXiv2025

Distilling Closed-Source LLM's Knowledge for Locally Stable and Economic Biomedical Entity Linking

Biomedical entity linking aims to map nonstandard entities to standard entities in a knowledge base. Traditional supervised methods perform well but require extensive annotated data to transfer, limiting their usage in low-resource scenarios. Large language models (LLMs), especially closed-source LLMs, can address these but risk stability issues and high economic costs: using these models is restricted by commercial companies and brings significant economic costs when dealing with large amounts of data. To address this, we propose ``RPDR'', a framework combining closed-source LLMs and open-source LLMs for re-ranking candidates retrieved by a retriever fine-tuned with a small amount of data. By prompting a closed-source LLM to generate training data from unannotated data and fine-tuning an open-source LLM for re-ranking, we effectively distill the knowledge to the open-source LLM that can be deployed locally, thus avoiding the stability issues and the problem of high economic costs. We evaluate RPDR on two datasets, including one real-world dataset and one publicly available dataset involving two languages: Chinese and English. RPDR achieves 0.019 Acc@1 improvement and 0.036 Acc@1 improvement on the Aier dataset and the Ask A Patient dataset when the amount of training data is not enough. The results demonstrate the superiority and generalizability of the proposed framework.

cs.CL↗