arXiv subjects

Lei Pan

Publications and source records attributed to Lei Pan.

At least 37 records · Page 2Linked to original sources

arXiv2025

ProKG-Dial: Progressive Multi-Turn Dialogue Construction with Domain Knowledge Graphs

Current large language models (LLMs) excel at general NLP tasks but often lack domain specific precision in professional settings. Building a high quality domain specific multi turn dialogue dataset is essential for developing specialized conversational systems. However, existing methods such as manual annotation, simulated human LLM interactions, and role based LLM dialogues are resource intensive or suffer from limitations in dialogue quality and domain coverage. To address these challenges, we introduce ProKG Dial, a progressive framework for constructing knowledge intensive multi turn dialogue datasets using domain specific knowledge graphs (KGs). ProKG Dial leverages the structured nature of KGs to encode complex domain knowledge and relationships, providing a solid foundation for generating meaningful and coherent dialogues. Specifically, ProKG Dial begins by applying community detection to partition the KG into semantically cohesive subgraphs. For each subgraph, the framework incrementally generates a series of questions and answers centered around a target entity, ensuring relevance and coverage. A rigorous filtering step is employed to maintain high dialogue quality. We validate ProKG Dial on a medical knowledge graph by evaluating the generated dialogues in terms of diversity, semantic coherence, and entity coverage. Furthermore, we fine tune a base LLM on the resulting dataset and benchmark it against several baselines. Both automatic metrics and human evaluations demonstrate that ProKG Dial substantially improves dialogue quality and domain specific performance, highlighting its effectiveness and practical utility.

cs.AI↗

arXiv2025

Multi-turn Natural Language to Graph Query Language Translation

In recent years, research on transforming natural language into graph query language (NL2GQL) has been increasing. Most existing methods focus on single-turn transformation from NL to GQL. In practical applications, user interactions with graph databases are typically multi-turn, dynamic, and context-dependent. While single-turn methods can handle straightforward queries, more complex scenarios often require users to iteratively adjust their queries, investigate the connections between entities, or request additional details across multiple dialogue turns. Research focused on single-turn conversion fails to effectively address multi-turn dialogues and complex context dependencies. Additionally, the scarcity of high-quality multi-turn NL2GQL datasets further hinders the progress of this field. To address this challenge, we propose an automated method for constructing multi-turn NL2GQL datasets based on Large Language Models (LLMs) , and apply this method to develop the MTGQL dataset, which is constructed from a financial market graph database and will be publicly released for future research. Moreover, we propose three types of baseline methods to assess the effectiveness of multi-turn NL2GQL translation, thereby laying a solid foundation for future research.

cs.AI↗

arXiv2025

NeRFs in Robotics: A Survey

Detailed and realistic 3D environment representations have been a long-standing goal in the fields of computer vision and robotics. The recent emergence of neural implicit representations has introduced significant advances to these domains, enabling numerous novel capabilities. Among these, Neural Radiance Fields (NeRFs) have gained considerable attention because of their considerable representational advantages, such as simplified mathematical models, low memory footprint, and continuous scene representations. In addition to computer vision, NeRFs have demonstrated significant potential in robotics. Thus, we present this survey to provide a comprehensive understanding of NeRFs in the field of robotics. By exploring the advantages and limitations of NeRF as well as its current applications and future potential, we aim to provide an overview of this promising area of research. Our survey is divided into two main sections: \textit{Applications of NeRFs in Robotics} and \textit{Advances for NeRFs in Robotics}, from the perspective of how NeRF enters the field of robotics. In the first section, we introduce and analyze some works that have been or could be used in robotics for perception and interaction tasks. In the second section, we show some works related to improving NeRF's own properties, which are essential for deploying NeRFs in robotics. In the discussion section of the review, we summarize the existing challenges and provide valuable future research directions.

cs.RO↗

arXiv2025

Reducing Tool Hallucination via Reliability Alignment

Large Language Models (LLMs) have expanded their capabilities beyond language generation to interact with external tools, enabling automation and real-world applications. However, tool hallucinations, where models either select inappropriate tools or misuse them, pose significant challenges, leading to erroneous task execution, increased computational costs, and reduced system reliability. To systematically address this issue, we define and categorize tool hallucinations into two main types, tool selection hallucination and tool usage hallucination. To evaluate and mitigate these issues, we introduce RelyToolBench, which integrates specialized test cases and novel metrics to assess hallucination-aware task success and efficiency. Finally, we propose Relign, a reliability alignment framework that expands the tool-use action space to include indecisive actions, allowing LLMs to defer tool use, seek clarification, or adjust tool selection dynamically. Through extensive experiments, we demonstrate that Relign significantly reduces tool hallucinations, improves task reliability, and enhances the efficiency of LLM tool interactions.

cs.CL↗

arXiv2025

Expedited thermalization dynamics in incommensurate systems

We study the thermalization dynamics of a quantum system embedded in an incommensurate potential and coupled to a Markovian thermal reservoir. The dephasing induced by the bath drives the system toward an infinite-temperature steady state, erasing all initial information-including signatures of localization. We find that initially localized states can relax to the homogeneous steady state faster than delocalized states. Moreover, low-temperature initial states thermalize to infinite temperature more rapidly than high-temperature states -- a phenomenon reminiscent of the Mpemba effect, in which hotter liquids freeze faster than colder ones. The slowest relaxation mode in the Liouvillian spectrum plays a critical role in the expedited thermalization for localized or cold initial states. Our results reveal that the combination of disordered structure and environmental dissipation may lead to non-trivial thermalization behavior, which advances both the conceptual framework of the Mpemba effect and the theoretical understanding of nonequilibrium processes in dissipative disordered systems.

quant-ph↗

arXiv2025

Task-Specific Data Selection for Instruction Tuning via Monosemantic Neuronal Activations

Instruction tuning improves the ability of large language models (LLMs) to follow diverse human instructions, but achieving strong performance on specific target tasks remains challenging. A critical bottleneck is selecting the most relevant data to maximize task-specific performance. Existing data selection approaches include unstable influence-based methods and more stable distribution alignment methods, the latter of which critically rely on the underlying sample representation. In practice, most distribution alignment methods, from shallow features (e.g., BM25) to neural embeddings (e.g., BGE, LLM2Vec), may fail to capture how the model internally processes samples. To bridge this gap, we adopt a model-centric strategy in which each sample is represented by its neuronal activation pattern in the model, directly reflecting internal computation. However, directly using raw neuron activations leads to spurious similarity between unrelated samples due to neuron polysemanticity, where a single neuron may respond to multiple, unrelated concepts. To address this, we employ sparse autoencoders to disentangle polysemantic activations into sparse, monosemantic representations, and introduce a dedicated similarity metric for this space to better identify task-relevant data. Comprehensive experiments across multiple instruction datasets, models, tasks, and selection ratios show that our approach consistently outperforms existing data selection baselines in both stability and task-specific performance.

cs.LG↗

arXiv2025

Dissipation induced localization-delocalization transition in a flat band

The interplay between dissipation and localization in quantum systems has garnered significant attention due to its potential to manipulate transport properties and induce phase transitions. In this work, we explore the dissipation-induced extended-localized transition in a flat band model, where the system's asymptotic state can be controlled by tailored dissipative operators. By analyzing the steady-state density matrix and dissipative dynamics, we demonstrate that dissipation is able to drive the system to states dominated by either extended or localized modes, irrespective of the initial conditions. The control mechanism relies on the phase properties of the dissipative operators, which selectively favor specific eigenstates of the Hamiltonian. Our findings reveal that dissipation can be harnessed to induce transitions between extended and localized phases, offering a novel approach to manipulate quantum transport in flat band systems. This work not only deepens our understanding of dissipation-induced phenomena in flat band systems but also provides a new avenue for controlling quantum states in open systems.

quant-ph↗

arXiv2025

Dissipation induced transition between delocalization and localization in the three-dimensional Anderson model

We investigate the probable delocalization-localization transition in open quantum systems with disorder. The disorder can induce localization in isolated quantum systems and it is generally recognized that localization is fragile under the action of dissipations from the external environment due to its interfering nature. Recent work [Y. Liu, et al, Phys. Rev. Lett. 132, 216301 (2024).] found that a one-dimensional quasiperiodic system can be driven into the localization phase by a tailored local dissipation where a dissipation-induced delocalized-localized transition is proposed. Based on this, we consider a more realistic system and show that a dissipation-induced transition between delocalization and localization appears in the three-dimensional (3D) Anderson model. By tuning local dissipative operators acting on nearest neighboring sites, we find that the system can relax to localized states dominated steady state instead of the choice of initial conditions and dissipation strengths. Moreover, we can also realize a delocalized states predominated steady-state from a localized initial state by using a kind of dissipation operators acting on next nearest neighboring sites. Our results enrich the applicability of dissipation-induced localization and identify the transition between delocalized and localized phases in 3D disordered systems.

cond-mat.dis-nn↗

arXiv2025

Alignment for Efficient Tool Calling of Large Language Models

Recent advancements in tool learning have enabled large language models (LLMs) to integrate external tools, enhancing their task performance by expanding their knowledge boundaries. However, relying on tools often introduces tradeoffs between performance, speed, and cost, with LLMs sometimes exhibiting overreliance and overconfidence in tool usage. This paper addresses the challenge of aligning LLMs with their knowledge boundaries to make more intelligent decisions about tool invocation. We propose a multi objective alignment framework that combines probabilistic knowledge boundary estimation with dynamic decision making, allowing LLMs to better assess when to invoke tools based on their confidence. Our framework includes two methods for knowledge boundary estimation, consistency based and absolute estimation, and two training strategies for integrating these estimates into the model decision making process. Experimental results on various tool invocation scenarios demonstrate the effectiveness of our framework, showing significant improvements in tool efficiency by reducing unnecessary tool usage.

cs.CL↗

arXiv2025

RL-GSBridge: 3D Gaussian Splatting Based Real2Sim2Real Method for Robotic Manipulation Learning

Sim-to-Real refers to the process of transferring policies learned in simulation to the real world, which is crucial for achieving practical robotics applications. However, recent Sim2real methods either rely on a large amount of augmented data or large learning models, which is inefficient for specific tasks. In recent years, with the emergence of radiance field reconstruction methods, especially 3D Gaussian splatting, it has become possible to construct realistic real-world scenes. To this end, we propose RL-GSBridge, a novel real-to-sim-to-real framework which incorporates 3D Gaussian Splatting into the conventional RL simulation pipeline, enabling zero-shot sim-to-real transfer for vision-based deep reinforcement learning. We introduce a mesh-based 3D GS method with soft binding constraints, enhancing the rendering quality of mesh models. Then utilizing a GS editing approach to synchronize the rendering with the physics simulator, RL-GSBridge could reflect the visual interactions of the physical robot accurately. Through a series of sim-to-real experiments, including grasping and pick-and-place tasks, we demonstrate that RL-GSBridge maintains a satisfactory success rate in real-world task completion during sim-to-real transfer. Furthermore, a series of rendering metrics and visualization results indicate that our proposed mesh-based 3D GS reduces artifacts in unstructured objects, demonstrating more realistic rendering performance.

cs.RO↗

arXiv2025

Robustness of quantum many-body scars in the presence of Markovian bath

A generic closed quantum many-body system will inevitably tend to thermalization, whose local information encoded in the initial state eventually scrambles into the full space, known as quantum ergodicity. A paradigmatic exception in closed quantum systems for strong ergodicity breaking is known as many-body localization, where strong disorder-induced localization prevents the occurrence of thermalization. It is generally recognized that a localized quantum system would be delocalized under dissipation induced by the environment. However, this consequence recently has received challenges where an exotic dissipation-induced localization mechanism is proposed, and transitions between localized and extended phases are found. In this Letter, we promote this mechanism to systems for weak ergodicity breaking hosting quantum many-body scars (QMBS). We find that the system relaxes to a steady state dominated by QMBS, and the dissipative dynamics exhibit dynamic revivals by suitably preparing an initial state. We point out an experimental realization of the controlled dissipation with a cold atomic setup. This makes the signature of ergodicity breaking visible over dissipative dynamics and offers potential possibilities for experimentally preparing stable QMBS with associated coherent dynamics.

cond-mat.dis-nn↗

arXiv2024

Generic non-Hermitian mobility edges in a class of duality-breaking quasicrystals

We provide approximate solutions for the mobility edge (ME) that demarcates localized and extended states within a specific class of one-dimensional non-Hermitian (NH) quasicrystals. These NH quasicrystals exhibit a combination of nonreciprocal hopping terms and complex quasiperiodic on-site potentials. Our analytical approach is substantiated by rigorous numerical calculations, demonstrating significant accuracy. Furthermore, our ansatz closely agrees with the established limiting cases of the NH Aubry-Andr{é}-Harper (AAH) and Ganeshan-Pixley-Das Sarma (GPD) models, which have exact results, thereby enhancing its credibility. Additionally, we have examined their dynamic properties and discovered distinct behaviors in different regimes. Our research provides a practical methodology for estimating the position of MEs in a category of NH quasicrystals that break duality.

cond-mat.dis-nn↗

arXiv2024

Smart energy management: process structure-based hybrid neural networks for optimal scheduling and economic predictive control in integrated systems

Integrated energy systems (IESs) are complex systems consisting of diverse operating units spanning multiple domains. To address its operational challenges, we propose a physics-informed hybrid time-series neural network (NN) surrogate to predict the dynamic performance of IESs across multiple time scales. This neural network-based modeling approach develops time-series multi-layer perceptrons (MLPs) for the operating units and integrates them with prior process knowledge about system structure and fundamental dynamics. This integration forms three hybrid NNs (long-term, slow, and fast MLPs) that predict the entire system dynamics across multiple time scales. Leveraging these MLPs, we design an NN-based scheduler and an NN-based economic model predictive control (NEMPC) framework to meet global operational requirements: rapid electrical power responsiveness to operators requests, adequate cooling supply to customers, and increased system profitability, while addressing the dynamic time-scale multiplicity present in IESs. The proposed day-ahead scheduler is formulated using the ReLU network-based MLP, which effectively represents IES performance under a broad range of conditions from a long-term perspective. The scheduler is then exactly recast into a mixed-integer linear programming problem for efficient evaluation. The real-time NEMPC, based on slow and fast MLPs, comprises two sequential distributed control agents: a slow NEMPC for the cooling-dominant subsystem with slower transient responses and a fast NEMPC for the power-dominant subsystem with faster responses. Extensive simulations demonstrate that the developed scheduler and NEMPC schemes outperform their respective benchmark scheduler and controller by about 25% and 40%. Together, they enhance overall system performance by over 70% compared to benchmark approaches.

eess.SY↗

arXiv2024

Exact anomalous mobility edges in one-dimensional non-Hermitian quasicrystals

Recent research has made significant progress in understanding localization transitions and mobility edges (MEs) that separate extended and localized states in non-Hermitian (NH) quasicrystals. Here we focus on studying critical states and anomalous MEs, which identify the boundaries between critical and localized states within two distinct NH quasiperiodic models. Specifically, the first model is a quasiperiodic mosaic lattice with both nonreciprocal hopping term and on-site potential. In contrast, the second model features an unbounded quasiperiodic on-site potential and nonreciprocal hopping. Using Avila's global theory, we analytically derive the Lyapunov exponent and exact anomalous MEs. To confirm the emergence of the robust critical states in both models, we conduct a numerical multifractal analysis of the wave functions and spectrum analysis of level spacing. Furthermore, we investigate the transition between real and complex spectra and the topological origins of the anomalous MEs. Our results may shed light on exploring the critical states and anomalous MEs in NH quasiperiodic systems.

cond-mat.dis-nn↗

arXiv2024

Probing the Robustness of Vision-Language Pretrained Models: A Multimodal Adversarial Attack Approach

Vision-language pretraining (VLP) with transformers has demonstrated exceptional performance across numerous multimodal tasks. However, the adversarial robustness of these models has not been thoroughly investigated. Existing multimodal attack methods have largely overlooked cross-modal interactions between visual and textual modalities, particularly in the context of cross-attention mechanisms. In this paper, we study the adversarial vulnerability of recent VLP transformers and design a novel Joint Multimodal Transformer Feature Attack (JMTFA) that concurrently introduces adversarial perturbations in both visual and textual modalities under white-box settings. JMTFA strategically targets attention relevance scores to disrupt important features within each modality, generating adversarial samples by fusing perturbations and leading to erroneous model predictions. Experimental results indicate that the proposed approach achieves high attack success rates on vision-language understanding and reasoning downstream tasks compared to existing baselines. Notably, our findings reveal that the textual modality significantly influences the complex fusion processes within VLP transformers. Moreover, we observe no apparent relationship between model size and adversarial robustness under our proposed attacks. These insights emphasize a new dimension of adversarial robustness and underscore potential risks in the reliable deployment of multimodal AI systems.

cs.CV↗

arXiv2024

Localization and mobility edges in non-Hermitian continuous quasiperiodic systems

The mobility edge (ME) is a fundamental concept in the Anderson localized systems, which marks the energy separating extended and localized states. Although the ME and localization phenomena have been extensively studied in non-Hermitian (NH) quasiperiodic tight-binding models, they remain limited to NH continuum systems. Here, we investigate the ME and localization properties of a one-dimensional (1D) NH quasiperiodic continuous system, which is described by a Schr{ö}dinger equation with an imaginary vector potential and an incommensurable one-site potential. We find that the ME is located in the real spectrum and falls between the localized and extended states. Additionally, we show that under the periodic boundary condition, the energy spectrum always exhibits an open curve representing high-energy extended electronic states characterized by a non-zero integer winding number. This complex spectrum topology is closely connected with the non-Hermitian skin effect (NHSE) observed under open boundary conditions, where the eigenstates of the bulk bands accumulate at the boundaries. Furthermore, we analyze the critical behavior of the localization transition and obtain critical potential amplitude accompanied by the universal critical exponent $ν\simeq 1/3$. Our study provides valuable inspiration for exploring MEs and localization behaviors in NH quasiperiodic continuous systems.

cond-mat.dis-nn↗

arXiv2024

Dissipation induced ergodic-nonergodic transitions in finite-height mosaic Wannier-Stark lattices

Recent research has observed the occurrence of pseudo-mobility edge (ME) within a modulated mosaic model incorporating the Wannier-Stark potential. This pseudo-ME, which signifies the critical energy that distinguishes between ergodic and weakly ergodic, or weakly ergodic and nonergodic states, is a crucial concept in comprehending the transport and localization phenomena in Wannier-Stark systems. Here we investigate the influence of dissipation on a finite-height mosaic Wannier-Stark lattice that features such pseudo-MEs by computing the steady state density matrix. Our findings indicate that particular dissipation can steer the system into specific states, regardless of its initial state, predominantly characterized by either ergodic or nonergodic states. This suggests that dissipation can be harnessed as a novel method for inducing transitions between these states and manipulating particle localization behaviors in disorder-free systems.

cond-mat.mes-hall↗

arXiv2024

Liouvillian skin effect in a one-dimensional open many-body quantum system with generalized boundary conditions

Non-Hermitian skin effect (NHSE), namely that eigenstates of non-Hermitian Hamiltonains are localized at one boundary in the open boundary condition, attracts great interest recently.In this paper, we investigate the skin effect in one-dimensional dissipative quantum many-body systems, which we call the Liouvillian skin effect (LSE). We rigorously identify the existence of LSE for generalized boundary conditions by solving the Liouvillian superoperator of an exactly solvable model with the advantage of Bethe ansatz. The LSE is sensitive to boundary conditions where the signature is reflected in eigenfunctions of the system. We confirm that the LSE is fragile to a tiny co-flow boundary hopping with non-Hermitian current but can survive for a counter-flow boundary hopping in the thermodynamic limit. Our work provides a prototypical example of exactly solvable dissipative quantum many-body lattice systems exhibiting LSE for generalized boundary conditions. It can be further extended to other integrable open quantum many-body models.

quant-ph↗