arXiv subjects

Tao Jiang

Publications and source records attributed to Tao Jiang.

At least 91 records · Page 5Linked to original sources

arXiv2025

Tutorial on Large Language Model-Enhanced Reinforcement Learning for Wireless Networks

Reinforcement Learning (RL) has shown remarkable success in enabling adaptive and data-driven optimization for various applications in wireless networks. However, classical RL suffers from limitations in generalization, learning feedback, interpretability, and sample efficiency in dynamic wireless environments. Large Language Models (LLMs) have emerged as a transformative Artificial Intelligence (AI) paradigm with exceptional capabilities in knowledge generalization, contextual reasoning, and interactive generation, which have demonstrated strong potential to enhance classical RL. This paper serves as a comprehensive tutorial on LLM-enhanced RL for wireless networks. We propose a taxonomy to categorize the roles of LLMs into four critical functions: state perceiver, reward designer, decision-maker, and generator. Then, we review existing studies exploring how each role of LLMs enhances different stages of the RL pipeline. Moreover, we provide a series of case studies to illustrate how to design and apply LLM-enhanced RL in low-altitude economy networking, vehicular networks, and space-air-ground integrated networks. Finally, we conclude with a discussion on potential future directions for LLM-enhanced RL and offer insights into its future development in wireless networks.

cs.NI↗

arXiv2025

A transfer learning approach for automatic conflicts detection in software requirement sentence pairs based on dual encoders

Software Requirement Document (RD) typically contain tens of thousands of individual requirements, and ensuring consistency among these requirements is critical for the success of software engineering projects. Automated detection methods can significantly enhance efficiency and reduce costs; however, existing approaches still face several challenges, including low detection accuracy on imbalanced data, limited semantic extraction due to the use of a single encoder, and suboptimal performance in cross-domain transfer learning. To address these issues, this paper proposes a Transferable Software Requirement Conflict Detection Framework based on SBERT and SimCSE, termed TSRCDF-SS. First, the framework employs two independent encoders, Sentence-BERT (SBERT) and Simple Contrastive Sentence Embedding (SimCSE), to generate sentence embeddings for requirement pairs, followed by a six-element concatenation strategy. Furthermore, the classifier is enhanced by a two-layer fully connected feedforward neural network (FFNN) with a hybrid loss optimization strategy that integrates a variant of Focal Loss, domain-specific constraints, and a confidence-based penalty term. Finally, the framework synergistically integrates sequential and cross-domain transfer learning. Experimental results demonstrate that the proposed framework achieves a 10.4% improvement in both macro-F1 and weighted-F1 scores in in-domain settings, and an 11.4% increase in macro-F1 in cross-domain scenarios.

cs.SE↗

arXiv2025

Multi-agent In-context Coordination via Decentralized Memory Retrieval

Large transformer models, trained on diverse datasets, have demonstrated impressive few-shot performance on previously unseen tasks without requiring parameter updates. This capability has also been explored in Reinforcement Learning (RL), where agents interact with the environment to retrieve context and maximize cumulative rewards, showcasing strong adaptability in complex settings. However, in cooperative Multi-Agent Reinforcement Learning (MARL), where agents must coordinate toward a shared goal, decentralized policy deployment can lead to mismatches in task alignment and reward assignment, limiting the efficiency of policy adaptation. To address this challenge, we introduce Multi-agent In-context Coordination via Decentralized Memory Retrieval (MAICC), a novel approach designed to enhance coordination by fast adaptation. Our method involves training a centralized embedding model to capture fine-grained trajectory representations, followed by decentralized models that approximate the centralized one to obtain team-level task information. Based on the learned embeddings, relevant trajectories are retrieved as context, which, combined with the agents' current sub-trajectories, inform decision-making. During decentralized execution, we introduce a novel memory mechanism that effectively balances test-time online data with offline memory. Based on the constructed memory, we propose a hybrid utility score that incorporates both individual- and team-level returns, ensuring credit assignment across agents. Extensive experiments on cooperative MARL benchmarks, including Level-Based Foraging (LBF) and SMAC (v1/v2), show that MAICC enables faster adaptation to unseen tasks compared to existing methods. Code is available at https://github.com/LAMDA-RL/MAICC.

cs.MA↗

arXiv2025

Back to Ear: Perceptually Driven High Fidelity Music Reconstruction

Variational Autoencoders (VAEs) are essential for large-scale audio tasks like diffusion-based generation. However, existing open-source models often neglect auditory perceptual aspects during training, leading to weaknesses in phase accuracy and stereophonic spatial representation. To address these challenges, we propose εar-VAE, an open-source music signal reconstruction model that rethinks and optimizes the VAE training paradigm. Our contributions are threefold: (i) A K-weighting perceptual filter applied prior to loss calculation to align the objective with auditory perception. (ii) Two novel phase losses: a Correlation Loss for stereo coherence, and a Phase Loss using its derivatives--Instantaneous Frequency and Group Delay--for precision. (iii) A new spectral supervision paradigm where magnitude is supervised by all four Mid/Side/Left/Right components, while phase is supervised only by the LR components. Experiments show εar-VAE at 44.1kHz substantially outperforms leading open-source models across diverse metrics, showing particular strength in reconstructing high-frequency harmonics and the spatial characteristics.

cs.SD↗

arXiv2025

TianHui: A Domain-Specific Large Language Model for Diverse Traditional Chinese Medicine Scenarios

Domain-specific LLMs in TCM face limitations in research settings due to constrained adaptability, insufficient evaluation datasets, and limited computational resources. This study presents TianHui, a specialized TCM LLM built through contextual data integration and domain knowledge fusion. We constructed a large-scale TCM corpus (0.97GB unsupervised data + 611,312 QA pairs) and employed a two-stage training strategy with QLoRA, DeepSpeed Stage 2, and Flash Attention 2. Evaluation on 12 benchmarks showed TianHui ranked top-three in all metrics for six datasets (APQ, TCMCD, HFR, HCCA, DHPE, TLAW) and achieved top results in the other six (TCMEE, APR, GCPMI, TCMKQA, TCMRC, ADTG). Optimal configuration was identified as LoRA rank=128, alpha=256, epoch=4, dropout=0.2, max length=2048. TianHui enables systematic preservation and scalable application of TCM knowledge. All resources are open-sourced.

cs.CL↗

arXiv2025

Phase estimation via photon subtraction at the output of the hybrid interferometer

The hybrid interferometer integrating an optical parametric amplifier and a beam splitter has the potential to outperform the SU(1,1) interferometer. However, photon loss remains a critical limitation for practical implementation. To address this challenge, we propose a quantum metrology scheme utilizing multi-photon subtraction at the output and replacing the conventional 50:50 beam splitter with a variable beam splitter to enhance robustness against photon loss. We employ a coherent state and a vacuum state as inputs and perform homodyne detection. Our results show that the selection of input modes significantly affects phase estimation, and optimizing the beam splitter's transmittance is crucial for maximizing phase sensitivity in lossy conditions. Furthermore, photon subtraction markedly improves phase sensitivity, quantum Fisher information, and robustness against noise. Our scheme achieves sensitivities beyond the Heisenberg limit even under 20% photon loss.

quant-ph↗

arXiv2025

DepthVLA: Enhancing Vision-Language-Action Models with Depth-Aware Spatial Reasoning

Vision-Language-Action (VLA) models have recently shown impressive generalization and language-guided manipulation capabilities. However, their performance degrades on tasks requiring precise spatial reasoning due to limited spatial reasoning inherited from Vision-Language Models (VLMs). Existing VLAs rely on extensive action-data pretraining to ground VLMs in 3D space, which reduces training efficiency and is still insufficient for accurate spatial understanding. In this work, we present DepthVLA, a simple yet effective VLA architecture that explicitly incorporates spatial awareness through a pretrained depth prediction module. DepthVLA adopts a mixture-of-transformers design that unifies a VLM, a depth transformer, and an action expert with fully shared attentions, forming an end-to-end model with enhanced spatial reasoning. Extensive evaluations in both real-world and simulated environments show that DepthVLA outperforms state-of-the-art approaches, achieving 78.5% vs. 65.0% progress in real-world tasks, 94.9% vs. 93.6% in the LIBERO simulator, and 74.8% vs. 58.8% in the Simpler simulator. Our code will be made publicly available.

cs.CV↗

arXiv2025

Tree Posets: Supersaturation, Enumeration, and Randomness

We develop a powerful tool for embedding any tree poset $P$ of height $k$ in the Boolean lattice which allows us to solve several open problems in the area. We show that: * If $H$ is a family in $B_n$ with $|H|\ge (q-1+\varepsilon){n\choose \lfloor n/2\rfloor}$ for some $q\ge k$, then $H$ contains on the order of as many induced copies of $P$ as is contained in the $q$ middle layers of the Boolean lattice. This generalizes results of Bukh and of Boehnlein and Jiang which guaranteed a single such copy in non-induced and induced settings respectively. * The number of induced $P$-free families of $B_n$ is $2^{(k-1+o(1)){n\choose \lfloor n/2\rfloor}}$, strengthening recent independent work of Balogh, Garcia, Wigal who obtained the same bounds in the non-induced setting. * The largest induced $P$-free subset of a $p$-random subset of $B_n$ for $p\gg n^{-1}$ has size at most $(k-1+o(1))p{n\choose \lfloor n/2\rfloor}$, generalizing previous work of Balogh, Mycroft, and Treglown and of Collares and Morris for the case when $P$ is a chain. All three results are asymptotically tight and give affirmative answers to general conjectures of Gerbner, Nagy, Patkós, and Vizer in the case of tree posets.

math.CO↗

arXiv2025

Measuring and Controlling the Spectral Bias for Self-Supervised Image Denoising

Current self-supervised denoising methods for paired noisy images typically involve mapping one noisy image through the network to the other noisy image. However, after measuring the spectral bias of such methods using our proposed Image Pair Frequency-Band Similarity, it suffers from two practical limitations. Firstly, the high-frequency structural details in images are not preserved well enough. Secondly, during the process of fitting high frequencies, the network learns high-frequency noise from the mapped noisy images. To address these challenges, we introduce a Spectral Controlling network (SCNet) to optimize self-supervised denoising of paired noisy images. First, we propose a selection strategy to choose frequency band components for noisy images, to accelerate the convergence speed of training. Next, we present a parameter optimization method that restricts the learning ability of convolutional kernels to high-frequency noise using the Lipschitz constant, without changing the network structure. Finally, we introduce the Spectral Separation and low-rank Reconstruction module (SSR module), which separates noise and high-frequency details through frequency domain separation and low-rank space reconstruction, to retain the high-frequency structural details of images. Experiments performed on synthetic and real-world datasets verify the effectiveness of SCNet.

cs.CV↗

arXiv2025

Sanitize Your Responses: Mitigating Privacy Leakage in Large Language Models

As Large Language Models (LLMs) achieve remarkable success across a wide range of applications, such as chatbots and code copilots, concerns surrounding the generation of harmful content have come increasingly into focus. Despite significant advances in aligning LLMs with safety and ethical standards, adversarial prompts can still be crafted to elicit undesirable responses. Existing mitigation strategies are predominantly based on post-hoc filtering, which introduces substantial latency or computational overhead, and is incompatible with token-level streaming generation. In this work, we introduce Self-Sanitize, a novel LLM-driven mitigation framework inspired by cognitive psychology, which emulates human self-monitor and self-repair behaviors during conversations. Self-Sanitize comprises a lightweight Self-Monitor module that continuously inspects high-level intentions within the LLM at the token level via representation engineering, and a Self-Repair module that performs in-place correction of harmful content without initiating separate review dialogues. This design allows for real-time streaming monitoring and seamless repair, with negligible impact on latency and resource utilization. Given that privacy-invasive content has often been insufficiently focused in previous studies, we perform extensive experiments on four LLMs across three privacy leakage scenarios. The results demonstrate that Self-Sanitize achieves superior mitigation performance with minimal overhead and without degrading the utility of LLMs, offering a practical and robust solution for safer LLM deployments. Our code is available at the following link: https://github.com/wjfu99/LLM_Self_Sanitize

cs.CL↗

arXiv2025

Dirac monopole magnets in non-Hermitian systems

We theoretically establish that non-Hermitian perturbations induce a topological transformation of point-like Dirac monopoles into extended monopole distributions, characterized by distinct charge configurations emergent from three distinct Berry connection forms. Using piecewise adiabatic evolution, we confirm the validity of these configurations through observations of complex geometric phases. Most critically, we find a quantitative relation $Δϕ_d = Δϕ_g$, which quantifies how cumulative minute energy differences ($Δϕ_d$) manifest as geometric phase shifts ($Δϕ_g$) uniquely in non-Hermitian systems. We further propose a scheme leveraging soliton dynamics in dissipative two-component Bose-Einstein condensates, enabling direct measurement of these topological signatures. These results establish a milestone for understanding Dirac monopole charge distributions and measuring complex geometric phases in non-Hermitian systems, with far-reaching implications for topological quantum computing and non-Hermitian photonics.

quant-ph↗

arXiv2025

Galaxea Open-World Dataset and G0 Dual-System VLA Model

We present Galaxea Open-World Dataset, a large-scale, diverse collection of robot behaviors recorded in authentic human living and working environments. All demonstrations are gathered using a consistent robotic embodiment, paired with precise subtask-level language annotations to facilitate both training and evaluation. Building on this dataset, we introduce G0, a dual-system framework that couples a Vision-Language Model (VLM) for multimodal planning with a Vision-Language-Action (VLA) model for fine-grained execution. G0 is trained using a three-stage curriculum: cross-embodiment pre-training, single-embodiment pre-training, and task-specific post-training. A comprehensive benchmark spanning tabletop manipulation, few-shot learning, and long-horizon mobile manipulation, demonstrates the effectiveness of our approach. In particular, we find that the single-embodiment pre-training stage, together with the Galaxea Open-World Dataset, plays a critical role in achieving strong performance.

cs.RO↗

arXiv2025

A note on zeros of derivatives of ultraspherical Bessel functions

For any fixed $ν\ge 0, δ\in \mathbb R$ and $x>0$, we investigate the positive zeros of the derivatives $j'_{ν,δ}(x)$ and $y'_{ν,δ}(x)$, where \begin{equation*} j_{ν,δ}(x)=x^{-δ}J_ν(x)\quad\text{and} \quad y_{ν,δ}(x)=x^{-δ}Y_ν(x). \end{equation*} We derive asymptotic expansions for their $k$-th positive zeros as $k\rightarrow \infty$.

math.CA↗

arXiv2025

Structural contribution to light-induced gap suppression in Ta$_2$NiSe$_5$

An excitonic insulator is a material that hosts an exotic ground state, where an energy gap opens due to spontaneous condensation of bound electron-hole pairs. Ta$_2$NiSe$_5$ is a promising candidate for this type of material, but the coexistence of a structural phase transition with the gap opening has led to a long-standing debate regarding the origin of the insulating gap. Here we employ MeV ultrafast electron diffraction to obtain quantitative insights into the atomic displacements in Ta$_2$NiSe$_5$ following photoexcitation, which has been overlooked in previous time-resolved spectroscopy studies. In conjunction with first-principles calculations using the measured atomic displacements, we find that the structural change can largely account for the photoinduced reduction in the energy gap without considering excitonic effects. Our work illustrates the importance of a quantitative reconstruction of individual atomic pathways during nonequilibrium phase transitions, paving the way for a mechanistic understanding of a diverse array of phase transitions in correlated materials where lattice dynamics can play a pivotal role.

cond-mat.mtrl-sci↗

arXiv2025

Deep Learning For Point Cloud Denoising: A Survey

Real-world environment-derived point clouds invariably exhibit noise across varying modalities and intensities. Hence, point cloud denoising (PCD) is essential as a preprocessing step to improve downstream task performance. Deep learning (DL)-based PCD models, known for their strong representation capabilities and flexible architectures, have surpassed traditional methods in denoising performance. To our best knowledge, despite recent advances in performance, no comprehensive survey systematically summarizes the developments of DL-based PCD. To fill the gap, this paper seeks to identify key challenges in DL-based PCD, summarizes the main contributions of existing methods, and proposes a taxonomy tailored to denoising tasks. To achieve this goal, we formulate PCD as a two-step process: outlier removal and surface noise restoration, encompassing most scenarios and requirements of PCD. Additionally, we compare methods in terms of similarities, differences, and respective advantages. Finally, we discuss research limitations and future directions, offering insights for further advancements in PCD.

cs.CV↗

arXiv2025

Super-resolution femtosecond electron diffraction reveals electronic and nuclear dynamics at conical intersections

Conical intersections play a pivotal role in excited-state quantum dynamics. Capturing transient molecular structures near conical intersections remains challenging due to the rapid timescales and subtle structural changes involved. We overcome this by combining the enhanced temporal resolution of mega-electron-volt ultrafast electron diffraction with a super-resolution real-space inversion algorithm, enabling visualization of nuclear and electronic motions at conical intersections with sub-angstrom resolution, surpassing the diffraction limit. We apply this technique to the textbook example of the ring-opening reaction of 1,3-cyclohexadiene, which proceeds through two conical intersections within 100 femtoseconds. The super-resolved transient structures near conical intersections reveal a C-C bond length difference of less than 0.4 angstrom and an approximately 30-femtosecond traversal time of the nuclear wave packet between them. These findings establish super-resolution ultrafast scattering as a transformative tool for uncovering quantum dynamics in molecules and open new avenues for studying light-matter interactions at the most fundamental level.

physics.chem-ph↗

arXiv2025

Regularization and asymmetric extremal numbers of subdivisions

Given a real $μ\geq 1$, a graph $H$ is $μ$-almost-regular if $Δ(H)\leq μδ(H)$. The celebrated regularization theorem of Erdős and Simonovits states that for every real $0<\varepsilon<1$ there exists a real $μ=μ(\varepsilon)$ such that every $n$-vertex graph $G$ with $Ω(n^{1+\varepsilon})$ edges contains an $m$-vertex $μ$-almost-regular subgraph $H$ with $Ω(m^{1+\varepsilon})$ edges for some $n^{\varepsilon\frac{1-\varepsilon}{1+\varepsilon}}\leq m\leq n$. We develop an enhanced version of it in which the subgraph $H$ also has average degree at least $Ω(\frac{d(G)}{\log n})$, where $d(G)$ is the average degree of $G$. We then give a bipartite analogue of the enhanced regularization theorem. Using the bipartite regularization theorem, we establish upper bounds on the maximum number of edges in a bipartite graph with part sizes $m$ and $n$ that does not contain a $2k$-subdivision of $K_{s,t}$ or $2k$-multi-subdivisions of $K_p$, thus extending the corresponding work of Janzer to the bipartite setting for even subdivisions. We show these upper bounds are tight up to a constant factor for infinitely many pairs $(m,n)$. The problem for estimating the maximum number of edges in a bipartite graph with part sizes $m$ and $n$ that does not contain a $(2k+1)$-subdivision of $K_{s,t}$ remains open.

math.CO↗

arXiv2025

A comparative analysis of plasmonic and dielectric metasurface sensing platforms powered by bound states in the continuum

Nanophotonic platforms based on surface-enhanced infrared absorbance spectroscopy (SEIRAS) have emerged as an effective tool for molecular detection. Sensitive nanophotonic sensors with robust resonant modes and amplified electromagnetic near fields are essential for spectroscopy, especially in lossy environments. Metasurfaces driven by bound state in the continuum (BICs) have unlocked a powerful platform for molecular detection due to their exceptional spectral selectivity. While plasmonic BIC metasurfaces are preferred for molecular spectroscopy due to their high surface fields, enhancing the interaction with analytes, dielectric BICs have become popular due to their high-quality factors and, thus high sensitivity. However, their sensing performance has largely been demonstrated in air, neglecting the intrinsic infrared (IR) losses found in common solvents. This study evaluates the suitability of plasmonic versus dielectric platforms for in-situ molecular spectroscopy. Here, the sensing performance of plasmonic (gold) and dielectric (silicon) metasurfaces is assessed across liquid environments with varying losses resembling typical solvents. The results show that dielectric metasurfaces excel in dry conditions, while plasmonic BIC metasurfaces outperform them in lossy solvents, with a distinct crossover point where both show similar performance. Our results provide a framework for selecting the optimal metasurface material platform for SEIRAS studies based on environmental conditions.

physics.optics↗