arXiv subjects

Jun Tao

Publications and source records attributed to Jun Tao.

At least 19 recordsLinked to original sources

arXiv2026

Periodic orbits around a magnetically charged black hole in $f(R,T)$ gravity coupled with Euler-Heisenberg electrodynamics

We systematically investigate the zoom-whirl periodic orbits of a small compact object orbiting a magnetically charged black hole in $f(R, T)$ gravity coupled with Euler-Heisenberg nonlinear electrodynamics. We evaluate the impact of the magnetic charge $Q_m$ and the Euler-Heisenberg parameter $a$ on the characteristics of the innermost stable circular orbit (ISCO) and the marginally bound orbit (MBO). Particularly, we focus on the precession parameter q and systematically examine how the magnetic charge and other parameters affect the trajectories of periodic orbits. Using the numerical kludge method, we generate gravitational waveforms for these periodic orbits. The results presented in our work demonstrate that the magnetic charge can significantly modify not only the precession parameter but also the orbital trajectories of the periodic orbits. Moreover, changing the magnetic charge $Q_m$ could cause a significant phase shift in the gravitational waveforms, while other parameters exert a relatively weaker influence on the gravitational waves.

gr-qc↗

arXiv2026

Pruned Traffic Trees: Native Semantic Compression with a Protocol-Structured Model Family for Encrypted Traffic Classification

Deep learning has achieved strong performance in encrypted traffic classification (ETC), yet its computational cost limits deployment on resource-constrained network devices such as routers and middleboxes. Existing compression methods mainly operate on weights, channels, hidden representations, or predictions, but do not explicitly determine which protocol fields and structural contexts should remain. We propose Pruned Traffic Trees (PTT), a three-level protocol-structured model family that treats native protocol structures as compression units. PTT-Full learns protocol-structured representations and field salience from complete Protocol Tree Graphs (PTGs), with flow-level self-supervised learning and protocol-presence-aware sparse execution. The learned salience and TopK+$k$ closure construct Distilled PTGs (PTG-Ds) for PTT-Distilled, while PTT-Lite inherits this topology and reduces width through structure-aligned transfer and flow-level logits distillation. Under flow-disjoint and Strong Information Information (SII)-masked settings, PTT-Full achieves Macro-F1 scores of 0.9519 and 0.9416 on CSTNET-TLS1.3 and CipherSpectrum, while PTT-Lite retains 0.9325 and 0.9136 with 80.3\% and 61.3\% fewer parameters, 98.85\% and 98.78\% lower effective GFLOPs, and 8.75$\times$ and 8.46$\times$ CPU inference speedups. These results demonstrate that treating protocol structure itself as the compression object enables effective performance-efficiency trade-offs for lightweight ETC.

cs.NI↗

arXiv2026

Mitigating Proxy-Induced Traffic Drift in Website Fingerprinting via Model-Agnostic Traffic Tailoring

Website fingerprinting (WF) based on deep learning can effectively identify websites from encrypted traffic. However, users often rely on proxy protocols to bypass censorship, and the diversity of these protocols poses a major challenge, as WF models trained on traffic from one set of protocols perform poorly when evaluated on that from unseen protocols. We attribute this issue to proxy-induced feature drift, where traffic patterns of the same website vary with the proxy protocol, leading to discrepancies that WF models fail to capture and severe performance degradation. To tackle this issue, we propose PA3, a model-agnostic preprocessing framework to analyze and mitigate the proxy-induced drift. PA3 first fingerprints the protocol-specific drift. These fingerprints are then used to tailor the proxied traffic for feature alignment, which mitigates the drift and considerably improves the generalization of WF models on traffic from unseen protocols. Extensive evaluations demonstrate that PA3 substantially enhances generalization on unseen protocols with an average improvement of 0.12 in F1-score (roughly 27% relative), achieving up to a 0.41 absolute gain across models, which narrows the performance gap introduced by the drift. In the best case, PA3 enables WF models to obtain F1-scores above 0.96 on traffic from unseen protocols.

cs.NI↗

arXiv2026

From Scalars to Time Series: Rethinking Implicit Neural Representations for Time-Varying Volumetric Data

Implicit neural representations (INRs) for time-varying volumetric data are typically trained using dense sampling over spatiotemporal coordinates, where each observation corresponds to a single point in space and time. This coordinate-wise formulation requires extensive sampling during optimization, leading to high computational cost and inefficient use of temporal structure. In this work, we revisit this design choice and show that dense spatiotemporal sampling is not necessary for learning time-varying fields. Instead, we represent the data as a collection of spatially indexed time series and train INRs using sequence-level supervision over each spatial location, rather than coordinate-wise scalar samples. This reformulation eliminates the need for dense spatiotemporal sampling and instead learns each spatial location from its full temporal evolution in a structured manner. We demonstrate that this representation is compatible with a range of existing INR architectures and consistently improves reconstruction quality, while significantly reducing training cost. Furthermore, we show that this formulation can be combined with mixture-of-experts architectures, and that our MoE instantiation further improves reconstruction quality compared to both the base reformulation and existing MoE-based INR methods, providing a stronger capacity allocation under heterogeneous temporal dynamics.

cs.AI↗

arXiv2026

Finite Coherence in Gravitational Waves from Tidally Excited Axion Clouds

Axion clouds around rotating black holes form gravitational atoms whose tidal transitions can radiate gravitational waves in binaries. For strongly coupled Bohr crossings, transition radiation is governed by the outgoing two-level coherence, not by the transition probability alone. This coherence is suppressed both on the adiabatic branch and in the weak passage limit, but survives for intermediate sweep rates, producing a finite transition waveform and a localized orbital response. In more massive systems, fine and hyperfine transitions produce narrowband gravitational radiation and cumulative departures from vacuum binary waveforms. Coherent tidal crossings offer a gravitational-wave probe of axion-cloud dynamics.

gr-qc↗

arXiv2026

Revisiting the Lyth bound constraints on inflation from ACT DR6 results

The Lyth bound asserts that the field excursion of inflaton must be sub-Planckian, thereby imposing an upper bound on the amplitude of the tensor power spectrum in inflationary scenario. This bound is conventionally derived assuming a scale-invariant curvature power spectrum, i.e., $n_s = 1$. However, astrophysical observations confirm a red-tilted spectrum with $n_s < 1$. In light of recent results from the Atacama Cosmology Telescope (ACT) DR6, we revisit these constraints using the newly implied scalar spectral index of $n_s \simeq 0.9743$. Incorporating the ACT data yields a different upper bound on the tensor-to-scalar ratio $r$, which can potentially exclude inflationary scenarios previously robust under the original Lyth bound with $n_s = 1$. Our result highlights the urgent need to combine theoretical Lyth bound considerations with the most up-to-date astrophysical data.

astro-ph.CO↗

arXiv2026

Treat Traffic Like Trees: A Semantic-Preserving Hierarchical Graph-Based Expert Framework for Encrypted Traffic Analysis

Graph-based deep learning methods have been widely employed in encrypted traffic analysis to exploit latent correlations across different granularities. However, while complex preprocessing pipelines and sophisticated model structures often achieve strong performance, they may obscure inherent protocol semantics during representation learning. Moreover, the hierarchical structure of protocol layers and their corresponding fields, defined by protocol specifications and routinely utilized in manual traffic analysis, remains underexplored in existing learning frameworks. In this paper, we propose Protocol Tree Graph Attention with Mixture of Experts (PTGAMoE), a semantic-preserving hierarchical graph-based expert framework for encrypted traffic analysis. The field-based graph construction and expert committee design enable PTGAMoE to quantify the model's preferences for specific fields and protocols. Extensive experimental results on representative benchmark datasets under strict no-data-leakage settings demonstrate that PTGAMoE significantly outperforms state-of-the-art (SOTA) models. Furthermore, the semantic-preserving design provides interpretable insights into protocol-level feature importance and expert-level contributions, reflecting the model's decision-making logic in encrypted traffic classification tasks.

cs.NI↗

arXiv2026

The MiniMax-M2 Series: Mini Activations Unleashing Max Real-World Intelligence

We introduce the MiniMax-M2 series, a family of Mixture-of-Experts language models built around the principle that mini activations can unleash maximum real-world intelligence. The flagship M2 contains 229.9B total parameters with only 9.8B activated per token. Designed end-to-end for agentic deployment, the M2 series rests on three components: (i) agent-driven data pipelines producing large-scale, verifiable trajectories across agentic coding and agentic cowork, each grounded in an executable workspace and an artifact-aligned reward; (ii) Forge, a scalable agent-native RL system that adapts to long-horizon agent trajectories, paired with windowed-FIFO scheduling, prefix-tree merging, inference optimization, and a clean training-inference-agent decoupling that supports both white-box and black-box agents; (iii) the latest M2.7 checkpoint takes an early step toward self-evolution -- autonomously debugging training runs and modifying its own scaffold. Across M2 through M2.7, this combination translates a mini-activation footprint into frontier-tier performance on agentic coding, deep search, office-task, and reasoning benchmarks.

cs.AI↗

arXiv2026

Exceptional Points in Quasinormal Spectra of Hairy Black Holes

Exceptional points (EPs) in quasinormal mode (QNM) spectra are non-Hermitian degeneracies at which both the eigenvalues and eigenfunctions coalesce. In this paper, we identify an EP in the scalar QNM spectrum of hairy black holes in the Einstein-Maxwell-scalar theory by scanning the parameter space. We then investigate its implications for ringdown signals by extracting QNMs from time-domain waveforms. Our results show that an EP ansatz, which includes a resonant contribution containing a term linear in time, provides a more robust description of ringdown at the EP than the standard ansatz based on a superposition of independent damped modes. In particular, it captures the resonant contribution associated with spectral coalescence more naturally, and even when the waveform is fitted with the standard ansatz, the resulting fit may still exhibit characteristic features of the EP ansatz.

gr-qc↗

arXiv2026

STEP: Scientific Time-Series Encoder Pretraining via Cross-Domain Distillation

Scientific time series are central to scientific AI but are typically sparse, highly heterogeneous, and limited in scale, making unified representation learning particularly challenging. Meanwhile, foundation models pretrained on relevant time series domains such as audio, general time series, and brain signals contain rich knowledge, but their applicability to scientific signals remains underexplored. In this paper, we investigate the transferability and complementarity of foundation models from relevant time series domains, and study how to effectively leverage them to build a unified encoder for scientific time series. We first systematically evaluate relevant foundation models, showing the effectiveness of knowledge transfer to scientific tasks and their complementary strengths. Based on this observation, we propose STEP, a Scientific Time Series Encoder Pretraining framework via cross domain distillation. STEP introduces adaptive patching to handle extreme-length sequences and a statistics compensation scheme to accommodate diverse numerical scales. It further leverages cross-domain distillation to integrate knowledge from multiple foundation models into a unified encoder. By combining complementary representations across different domains, STEP learns general-purpose and transferable features tailored for scientific signals. Experiments on seven scientific time series tasks demonstrate that STEP provides both an effective structure and an effective pretraining paradigm, taking a STEP toward scientific time series representation learning.

cs.LG↗

arXiv2026

A Multi-task Large Reasoning Model for Molecular Science

Advancements in artificial intelligence for molecular science are necessitating a paradigm shift from purely data-driven predictions to knowledge-guided computational reasoning. Existing molecular models are predominantly proprietary, lacking general molecular intelligence and generalizability. This underscores the necessity for computational methods that can effectively integrate scientific logic with deep learning architectures. Here we introduce a multi-task large reasoning model designed to emulate the cognitive processes of molecular scientists through structured reasoning and reflection. Our approach incorporates multi-specialist modules to provide versatile molecular expertise and a chain-of-thought (CoT) framework enhanced by reinforcement learning infused with molecular knowledge, enabling structured and reflective reasoning. Systematic evaluations across 10 molecular tasks and 47 metrics demonstrate that our model achieves an average 50.3% improvement over the base architecture, outperforming over 20 state-of-the-art baselines, including ultra-large-parameter foundation models, despite using significantly fewer training data and computational resources. This validates that embedding explicit reasoning mechanisms enables high-efficiency learning, allowing smaller-scale models to surpass massive counterparts in both efficacy and interpretability. The practical utility of this computational framework was validated through a case study on the design of central nervous system (CNS) drug candidates, illustrating its capacity to bridge data-driven and knowledge-integrated approaches for intelligent molecular design.

cs.LG↗

arXiv2026

Enhanced Drug-drug Interaction Prediction Using Adaptive Knowledge Integration

Drug-drug interaction event (DDIE) prediction is crucial for preventing adverse reactions and ensuring optimal therapeutic outcomes. However, existing methods often face challenges with imbalanced datasets, complex interaction mechanisms, and poor generalization to unknown drug combinations. To address these challenges, we propose a knowledge augmentation framework that adaptively infuses prior drug knowledge into a large language model (LLM). This framework utilizes reinforcement learning techniques to facilitate adaptive knowledge extraction and synthesis, thereby efficiently optimizing the strategy space to enhance the accuracy of LLMs for DDIE predictions. As a result of few-shot learning, we achieved a notable improvement compared to the baseline. This approach establishes an effective framework for scientific knowledge learning for DDIE predictions.

cs.LG↗

arXiv2026

Orbital Dynamics and Gravitational Wave Signatures of Extreme Mass Ratio Inspirals in Galactic Dark Matter Halos

In astrophysics, extreme mass ratio inspiral (EMRI) systems, which consist of a central supermassive black hole and a stellar-mass compact object (SCO), are typically embedded in galactic dark matter (DM) halos. This dark matter environment inevitably affects the orbital dynamics of the SCO and the gravitational wave (GW) signals emitted by the system. In this work, we select two typical dark matter halo profiles -- the Navarro-Frenk-White (NFW) and Beta models -- to systematically investigate their specific impacts on the long-term orbital evolution of the SCO. By incorporating three dissipative mechanisms -- dynamical friction, accretion, and gravitational radiation reaction -- our results demonstrate that, compared to a pure vacuum medium, the presence of a dark matter halo significantly alters the trajectories of precessing orbits, the dynamical evolution of orbital parameters, and the waveforms and phases of the emitted gravitational waves. Due to the strong accretion effect within the NFW model, the energy flux exhibits a distinctive "cusp" feature, marking a reversal from net energy loss to gain at a specific semi-latus rectum, which is a phenomenon absent in the Beta model. Although short-term observations may not be sufficient to distinguish between the NFW and Beta models, their differences become evident over long-term orbital evolution. The gravitational waveforms computed using the NFW and Beta models exhibit a phase shift, which could be detectable in high-density DM environments. This phase shift becomes even more pronounced for higher eccentric orbits and longer observation times. These results offer a theoretical framework for probing environmental effects on EMRIs across different dark matter models using future space-based gravitational wave observatories.

gr-qc↗

arXiv2026

GenMRP: A Generative Multi-Route Planning Framework for Efficient and Personalized Real-Time Industrial Navigation

Existing industrial-scale navigation applications contend with massive road networks, typically employing two main categories of approaches for route planning. The first relies on precomputed road costs for optimal routing and heuristic algorithms for generating alternatives, while the second, generative methods, has recently gained significant attention. However, the former struggles with personalization and route diversity, while the latter fails to meet the efficiency requirements of large-scale real-time scenarios. To address these limitations, we propose GenMRP, a generative framework for multi-route planning. To ensure generation efficiency, GenMRP first introduces a skeleton-to-capillary approach that dynamically constructs a relevant sub-network significantly smaller than the full road network. Within this sub-network, routes are generated iteratively. The first iteration identifies the optimal route, while the subsequent ones generate alternatives that balance quality and diversity using the newly proposed correctional boosting approach. Each iteration incorporates road features, user historical sequences, and previously generated routes into a Link Cost Model to update road costs, followed by route generation using the Dijkstra algorithm. Extensive experiments show that GenMRP achieves state-of-the-art performance with high efficiency in both offline and online environments. To facilitate further research, we have publicly released the training and evaluation dataset. GenMRP has been fully deployed in a real-world navigation app, demonstrating its effectiveness and benefits.

cs.RO↗

arXiv2025

Periodic orbits and their gravitational waves in EMRIs: supermassive black hole affected by galactic dark matter halos

Periodic orbits exhibiting zoom-whirl behavior have become attractive topics for studying particle dynamics and gravitational wave emission in extreme-mass-ratio inspirals (EMRIs). This study systematically investigates periodic orbits around black holes and their gravitational wave radiation in three dark matter halo environments: NFW, Beta, and Moore models. The dark matter distribution in these models can be effectively incorporated using two parameters -- the dark matter characteristic mass and halo characteristic radius. Our results reveal that for a larger dark matter mass and a smaller characteristic radius, the shapes of the periodic orbits and the corresponding gravitational waveforms show more significant deviations from the Schwarzschild case. As the halo characteristic radius increases, the orbital shapes and waveform characteristics gradually converge with the Schwarzschild black hole results. Our results also suggest that the NFW and Beta models produce nearly indistinguishable results, while the Moore model shows distinct signatures compared with Beta/NFW models. Furthermore, calculations of characteristic strains in frequency spectra show that gravitational-wave signals associated with these periodic orbits lie above the sensitivity curves of LISA, TianQin, and Taiji, indicating their detectability in future space-based gravitational wave observatories. These findings deepen our understanding of dark matter halo effects on periodic motions and gravitational wave signatures. Keywords: GR black holes; Gravitational waves in GR and beyond: theory; dark matter theory; astrophysical black holes

gr-qc↗

arXiv2025

LG-CD: Enhancing Language-Guided Change Detection through SAM2 Adaptation

Remote Sensing Change Detection (RSCD) typically identifies changes in land cover or surface conditions by analyzing multi-temporal images. Currently, most deep learning-based methods primarily focus on learning unimodal visual information, while neglecting the rich semantic information provided by multimodal data such as text. To address this limitation, we propose a novel Language-Guided Change Detection model (LG-CD). This model leverages natural language prompts to direct the network's attention to regions of interest, significantly improving the accuracy and robustness of change detection. Specifically, LG-CD utilizes a visual foundational model (SAM2) as a feature extractor to capture multi-scale pyramid features from high-resolution to low-resolution across bi-temporal remote sensing images. Subsequently, multi-layer adapters are employed to fine-tune the model for downstream tasks, ensuring its effectiveness in remote sensing change detection. Additionally, we design a Text Fusion Attention Module (TFAM) to align visual and textual information, enabling the model to focus on target change regions using text prompts. Finally, a Vision-Semantic Fusion Decoder (V-SFD) is implemented, which deeply integrates visual and semantic information through a cross-attention mechanism to produce highly accurate change detection masks. Our experiments on three datasets (LEVIR-CD, WHU-CD, and SYSU-CD) demonstrate that LG-CD consistently outperforms state-of-the-art change detection methods. Furthermore, our approach provides new insights into achieving generalized change detection by leveraging multimodal information.

cs.CV↗

arXiv2025

Natural Language-Driven Viewpoint Navigation for Volume Exploration via Semantic Block Representation

Exploring volumetric data is crucial for interpreting scientific datasets. However, selecting optimal viewpoints for effective navigation can be challenging, particularly for users without extensive domain expertise or familiarity with 3D navigation. In this paper, we propose a novel framework that leverages natural language interaction to enhance volumetric data exploration. Our approach encodes volumetric blocks to capture and differentiate underlying structures. It further incorporates a CLIP Score mechanism, which provides semantic information to the blocks to guide navigation. The navigation is empowered by a reinforcement learning framework that leverage these semantic cues to efficiently search for and identify desired viewpoints that align with the user's intent. The selected viewpoints are evaluated using CLIP Score to ensure that they best reflect the user queries. By automating viewpoint selection, our method improves the efficiency of volumetric data navigation and enhances the interpretability of complex scientific phenomena.

cs.AI↗

arXiv2025

Automatic Semantic Alignment of Flow Pattern Representations for Exploration with Large Language Models

Explorative flow visualization allows domain experts to analyze complex flow structures by interactively investigating flow patterns. However, traditional visual interfaces often rely on specialized graphical representations and interactions, which require additional effort to learn and use. Natural language interaction offers a more intuitive alternative, but teaching machines to recognize diverse scientific concepts and extract corresponding structures from flow data poses a significant challenge. In this paper, we introduce an automated framework that aligns flow pattern representations with the semantic space of large language models (LLMs), eliminating the need for manual labeling. Our approach encodes streamline segments using a denoising autoencoder and maps the generated flow pattern representations to LLM embeddings via a projector layer. This alignment empowers semantic matching between textual embeddings and flow representations through an attention mechanism, enabling the extraction of corresponding flow patterns based on textual descriptions. To enhance accessibility, we develop an interactive interface that allows users to query and visualize flow structures using natural language. Through case studies, we demonstrate the effectiveness of our framework in enabling intuitive and intelligent flow exploration.

cs.HC↗