arXiv subjects

Chen Zhang

Publications and source records attributed to Chen Zhang.

At least 199 records · Page 11Linked to original sources

arXiv2025

NeuroBridge: Bio-Inspired Self-Supervised EEG-to-Image Decoding via Cognitive Priors and Bidirectional Semantic Alignment

Visual neural decoding seeks to reconstruct or infer perceived visual stimuli from brain activity patterns, providing critical insights into human cognition and enabling transformative applications in brain-computer interfaces and artificial intelligence. Current approaches, however, remain constrained by the scarcity of high-quality stimulus-brain response pairs and the inherent semantic mismatch between neural representations and visual content. Inspired by perceptual variability and co-adaptive strategy of the biological systems, we propose a novel self-supervised architecture, named NeuroBridge, which integrates Cognitive Prior Augmentation (CPA) with Shared Semantic Projector (SSP) to promote effective cross-modality alignment. Specifically, CPA simulates perceptual variability by applying asymmetric, modality-specific transformations to both EEG signals and images, enhancing semantic diversity. Unlike previous approaches, SSP establishes a bidirectional alignment process through a co-adaptive strategy, which mutually aligns features from two modalities into a shared semantic space for effective cross-modal learning. NeuroBridge surpasses previous state-of-the-art methods under both intra-subject and inter-subject settings. In the intra-subject scenario, it achieves the improvements of 12.3% in top-1 accuracy and 10.2% in top-5 accuracy, reaching 63.2% and 89.9% respectively on a 200-way zero-shot retrieval task. Extensive experiments demonstrate the effectiveness, robustness, and scalability of the proposed framework for neural visual decoding.

cs.CV↗

arXiv2025

Analytical solutions of bound timelike geodesic orbits in effective-one-body frame

We derive the approximate analytical solutions of the bound timelike geodesic orbits in the effective-one-body (EOB) frame with extreme-mass ratio limit. The analytical solutions are expressed in terms of the elliptic integrals using Mino time $λ$ as the independent variable. Since Mino time decouples the $r$ and $θ$-motion, we also give explicit expressions for three orbital frequencies $Ω_r, ~Ω_θ, ~Ω_ϕ$ using the Fourier series expansion. With these analytical expressions at hand, we can perform Fourier expansions in Mino time $λ$ for any function expressed in terms of the coordinates $(r,θ,ϕ)$. In particular, the observer's time t is decomposed into Mino time $λ$, and the frequency-domain description is constructed from the $λ$-Fourier expansion and the expansion of t. These analytical expressions are quite simple to implement, and can be applicable for calculating gravitational waves (GWs) from extreme mass-ratio inspirals (EMRIs) with the frequency-domain Teukolsky equation.

gr-qc↗

arXiv2025

Reinforcement Learning Foundations for Deep Research Systems: A Survey

Deep research systems, agentic AI that solve complex, multi-step tasks by coordinating reasoning, search across the open web and user files, and tool use, are moving toward hierarchical deployments with a Planner, Coordinator, and Executors. In practice, training entire stacks end-to-end remains impractical, so most work trains a single planner connected to core tools such as search, browsing, and code. While SFT imparts protocol fidelity, it suffers from imitation and exposure biases and underuses environment feedback. Preference alignment methods such as DPO are schema and proxy-dependent, off-policy, and weak for long-horizon credit assignment and multi-objective trade-offs. A further limitation of SFT and DPO is their reliance on human defined decision points and subskills through schema design and labeled comparisons. Reinforcement learning aligns with closed-loop, tool-interaction research by optimizing trajectory-level policies, enabling exploration, recovery behaviors, and principled credit assignment, and it reduces dependence on such human priors and rater biases. This survey is, to our knowledge, the first dedicated to the RL foundations of deep research systems. It systematizes recent work along three axes: (i) data synthesis and curation; (ii) RL methods for agentic research covering stability, sample efficiency, long context handling, reward and credit design, multi-objective optimization, and multimodal integration; and (iii) agentic RL training systems and frameworks. We also cover agent architecture and coordination, as well as evaluation and benchmarks, including recent QA, VQA, long-form synthesis, and domain-grounded, tool-interaction tasks. We distill recurring patterns, surface infrastructure bottlenecks, and offer practical guidance for training robust, transparent deep research agents with RL.

cs.AI↗

arXiv2025

AnchorTP: Resilient LLM Inference with State-Preserving Elastic Tensor Parallelism

Large Language Model (LLM) inference services demand exceptionally high availability and low latency, yet multi-GPU Tensor Parallelism (TP) makes them vulnerable to single-GPU failures. We present AnchorTP, a state-preserving elastic TP framework for fast recovery. It (i) enables Elastic Tensor Parallelism (ETP) with unequal-width partitioning over any number of GPUs and compatibility with Mixture-of-Experts (MoE), and (ii) preserves model parameters and KV caches in GPU memory via a daemon decoupled from the inference process. To minimize downtime, we propose a bandwidth-aware planner based on a Continuous Minimal Migration (CMM) algorithm that minimizes reload bytes under a byte-cost dominance assumption, and an execution scheduler that pipelines P2P transfers with reloads. These components jointly restore service quickly with minimal data movement and without changing service interfaces. In typical failure scenarios, AnchorTP reduces Time to First Success (TFS) by up to 11x and Time to Peak (TTP) by up to 59% versus restart-and-reload.

cs.DC↗

arXiv2025

A Survey on LLM Mid-Training

Recent advances in foundation models have highlighted the significant benefits of multi-stage training, with a particular emphasis on the emergence of mid-training as a vital stage that bridges pre-training and post-training. Mid-training is distinguished by its use of intermediate data and computational resources, systematically enhancing specified capabilities such as mathematics, coding, reasoning, and long-context extension, while maintaining foundational competencies. This survey provides a formal definition of mid-training for large language models (LLMs) and investigates optimization frameworks that encompass data curation, training strategies, and model architecture optimization. We analyze mainstream model implementations in the context of objective-driven interventions, illustrating how mid-training serves as a distinct and critical stage in the progressive development of LLM capabilities. By clarifying the unique contributions of mid-training, this survey offers a comprehensive taxonomy and actionable insights, supporting future research and innovation in the advancement of LLMs.

cs.CL↗

arXiv2025

Exploringand Unleashing the Power of Large Language Models in CI/CD Configuration Translation

Continuous Integration (CI) is a cornerstone of modern collaborative software development, and numerous CI platforms are available. Differences in maintenance overhead, reliability, and integration depth with code-hosting platforms make migration between CI platforms a common practice. A central step in migration is translating CI configurations, which is challenging due to the intrinsic complexity of CI configurations and the need to understand semantic differences and relationships across CI platforms. With the advent of large language models (LLMs), recent advances in software engineering highlight their potential for CI configuration translation. In this paper, we present a study on LLM-based CI configuration translation, focusing on the migration from Travis CI to GitHub Actions. First, using 811 migration records, we quantify the effort involved and find that developers read an average of 38 lines of Travis configuration and write 58 lines of GitHub Actions configuration, with nearly half of the migrations requiring multiple commits. We further analyze translations produced by each of the four LLMs and identify 1,121 issues grouped into four categories: logic inconsistencies (38%), platform discrepancies (32%), environment errors (25%), and syntax errors (5%). Finally, we evaluate three enhancement strategies and show that combining guideline-based prompting with iterative refinement achieves the best performance, reaching a Build Success Rate of 75.5%-nearly a threefold improvement over GPT-4o with a basic prompt.

cs.SE↗

arXiv2025

Hybrid Quark Stars with Quark-Quark Phase Transitions

We explore the possibility of phase transitions between different quark matter phases occurring within quark stars, giving rise to the hybrid quark stars (HybQSs). First, we obtain the generic phase diagram of possible mass-radius relation forms for the HybQSs. Then, utilizing a well-established general parameterization of interacting quark matter, we construct quark star models featuring sharp first-order quark-quark phase transitions of various types, in contrast to the hadron-quark transition in conventional hybrid stars. We systematically investigate how recent observations, such as the pulsar mass measurement $M_{\rm TOV}\gtrsim2M_{\odot}$ and the GW170817's tidal deformability bound $Λ_{1.4M_{\odot}}<800$, constrain the viable parameter space. We also identified twin stars in some of the HybQS parameter space. Furthermore, we found that the quark-quark phase transitions in hybrid quark stars may also cause the deviation from the approximate universal relation between the dominant postmerger frequency $f_{\rm peak}$ and tidal deformability $Λ_{1.35M_{\odot}}$, which was previously believed to only be caused by the hadron-quark phase transitions in hybrid neutron stars. This work unveils new possibilities of phase transitions and the resulting new types of compact stars in realistic astrophysical scenarios.

astro-ph.HE↗

arXiv2025

Convergence analysis of Sobolev Gradient flows for the rotating Gross-Pitaevskii energy functional

This paper studies the numerical approximation of the ground state of rotating Bose--Einstein condensates, formulated as the minimization of the Gross--Pitaevskii energy functional under a mass conservation constraint. To solve this problem, we consider three Sobolev gradient flow schemes: the $H_0^1$ scheme, the $a_0$ scheme, and the $a_u$ scheme. Convergence of these schemes in the non-rotating case was established by Chen et al., and the rotating $a_u$ scheme was analyzed in Henning et al. In this work, we prove the global convergence of the $H_0^1$ and $a_0$ schemes in the rotating case, and establish local linear convergence for all three schemes near the ground state. Numerical experiments confirm our theoretical findings.

math.NA↗

arXiv2025

Doc-Researcher: A Unified System for Multimodal Document Parsing and Deep Research

Deep Research systems have revolutionized how LLMs solve complex questions through iterative reasoning and evidence gathering. However, current systems remain fundamentally constrained to textual web data, overlooking the vast knowledge embedded in multimodal documents Processing such documents demands sophisticated parsing to preserve visual semantics (figures, tables, charts, and equations), intelligent chunking to maintain structural coherence, and adaptive retrieval across modalities, which are capabilities absent in existing systems. In response, we present Doc-Researcher, a unified system that bridges this gap through three integrated components: (i) deep multimodal parsing that preserves layout structure and visual semantics while creating multi-granular representations from chunk to document level, (ii) systematic retrieval architecture supporting text-only, vision-only, and hybrid paradigms with dynamic granularity selection, and (iii) iterative multi-agent workflows that decompose complex queries, progressively accumulate evidence, and synthesize comprehensive answers across documents and modalities. To enable rigorous evaluation, we introduce M4DocBench, the first benchmark for Multi-modal, Multi-hop, Multi-document, and Multi-turn deep research. Featuring 158 expert-annotated questions with complete evidence chains across 304 documents, M4DocBench tests capabilities that existing benchmarks cannot assess. Experiments demonstrate that Doc-Researcher achieves 50.6% accuracy, 3.4xbetter than state-of-the-art baselines, validating that effective document research requires not just better retrieval, but fundamentally deep parsing that preserve multimodal integrity and support iterative research. Our work establishes a new paradigm for conducting deep research on multimodal document collections.

cs.IR↗

arXiv2025

Fighter: Unveiling the Graph Convolutional Nature of Transformers in Time Series Modeling

Transformers have achieved remarkable success in time series modeling, yet their internal mechanisms remain opaque. This work demystifies the Transformer encoder by establishing its fundamental equivalence to a Graph Convolutional Network (GCN). We show that in the forward pass, the attention distribution matrix serves as a dynamic adjacency matrix, and its composition with subsequent transformations performs computations analogous to graph convolution. Moreover, we demonstrate that in the backward pass, the update dynamics of value and feed-forward projections mirror those of GCN parameters. Building on this unified theoretical reinterpretation, we propose \textbf{Fighter} (Flexible Graph Convolutional Transformer), a streamlined architecture that removes redundant linear projections and incorporates multi-hop graph aggregation. This perspective yields an explicit and interpretable representation of temporal dependencies across different scales, naturally expressed as graph edges. Experiments on standard forecasting benchmarks confirm that Fighter achieves competitive performance while providing clearer mechanistic interpretability of its predictions.

cs.LG↗

arXiv2025

FSRF: Factorization-guided Semantic Recovery for Incomplete Multimodal Sentiment Analysis

In recent years, Multimodal Sentiment Analysis (MSA) has become a research hotspot that aims to utilize multimodal data for human sentiment understanding. Previous MSA studies have mainly focused on performing interaction and fusion on complete multimodal data, ignoring the problem of missing modalities in real-world applications due to occlusion, personal privacy constraints, and device malfunctions, resulting in low generalizability. To this end, we propose a Factorization-guided Semantic Recovery Framework (FSRF) to mitigate the modality missing problem in the MSA task. Specifically, we propose a de-redundant homo-heterogeneous factorization module that factorizes modality into modality-homogeneous, modality-heterogeneous, and noisy representations and design elaborate constraint paradigms for representation learning. Furthermore, we design a distribution-aligned self-distillation module that fully recovers the missing semantics by utilizing bidirectional knowledge transfer. Comprehensive experiments on two datasets indicate that FSRF has a significant performance advantage over previous methods with uncertain missing modalities.

cs.LG↗

arXiv2025

Merge and Guide: Unifying Model Merging and Guided Decoding for Controllable Multi-Objective Generation

Adapting to diverse user needs at test time is a key challenge in controllable multi-objective generation. Existing methods are insufficient: merging-based approaches provide indirect, suboptimal control at the parameter level, often disregarding the impacts of multiple objectives. While decoding-based guidance is more direct, it typically requires aggregating logits from multiple expert models, incurring significant space overhead and relying heavily on individual model capacity. To address these issues, we introduce Merge-And-GuidE (MAGE), a two-stage framework that leverages model merging for guided decoding. We first identify a critical compatibility problem between the guidance and base models. In Stage 1, MAGE resolves this by dynamically constructing a more robust base model, merging a series of backbone models that account for multiple objectives. In Stage 2, we merge explicit and implicit value models into a unified guidance proxy, which then steers the decoding of the base model from Stage 1. Our analysis empirically validates Linear Mode Connectivity (LMC) in value models, explores the relationship between model merging and prediction ensembling, and demonstrates the enhanced controllability afforded by our approach. Extensive experiments show that our method outperforms existing approaches, achieving superior controllability, Pareto-optimal performance, and enhanced adaptability.

cs.LG↗

arXiv2025

3C Resources Joint Allocation for Time-Deterministic Remote Sensing Image Backhaul in the Space-Ground Integrated Network

Low-Earth-orbit (LEO) satellites assist observation satellites (OSs) to compress and backhaul more time-determined images (TDI) has become a new paradigm, which is used to enhance the timeout caused by the limited computing resources of OSs. However, how to capture the time-varying and dynamic characteristics of multi-dimensional resources is challenging for efficient collaborative scheduling. Motivated by this factor, we design a highly succinct multi-dimensional resource time-expanded graph (MDR-TEG) modell. Specifically, by employing a slots division mechanism and introducing an external virtual node, the time-varying communication, caching, and computing (3C) resources are depicted in low complexity by the link weights within, between, and outside the slots. Based on the MDR-TEG, the maximizing successful transmission ratio of TDI (MSTR-TDI) is modeled as a mixed integer linear programming (MILP) problem. Which further relaxed decomposed into two tractable sub-problems: maximizing the successful transmission rate of images (MSTRI) and ensuring the timeliness problem (ETP). Subsequently, an efficient subgradient of relaxation computing constraint (SRCC) algorithm is proposed. The upper and lower bounds of MSTR-TDI are obtained by solving the two subproblems and the dual problem (DP), and the direction of the next iteration is obtained by feedback. Furthermore, arranging the sending sequences of images to improve the quality of the solution. The approximate optimal solution of MSTR-TDI is eventually obtained through repeated iterations. The simulation results verify the superiority of the proposed MDR-TEG model and the effectiveness of the SRCC.

eess.SY↗

arXiv2025

Relative higher index theory on quotients of Roe algebras and positive scalar curvature at infinity

In this paper, we employ quotients of Roe algebras as index containers for elliptic differential operators to study the existence problem of Riemannian metrics with positive scalar curvature on non-compact complete Riemannian manifolds. The non-vanishing of such an index locates the precise direction at infinity of the obstructions to positive scalar curvature, and may be viewed as a refinement of the positive scalar curvature problem. To achieve this, we formulate the relative coarse Baum-Connes conjecture and the relative coarse Novikov conjecture, together with their maximal versions, for general metric spaces as a program to compute the $K$-theory of the quotients of the Roe algebras relative to specific ideals. We show that if the metric space admits a relative fibred coarse embedding into Hilbert space or an $\ell^p$-space, certain cases of these conjectures can be verified, which yield obstructions to the existence of uniformly positive scalar curvature metrics in specified directions at infinity. As an application, we prove that the maximal coarse Baum-Connes conjecture holds for finite products of certain expander graphs that fail to admit fibred coarse embeddings into Hilbert space.

math.KT↗

arXiv2025

FAME: Adaptive Functional Attention with Expert Routing for Function-on-Function Regression

Functional data play a pivotal role across science and engineering, yet their infinite-dimensional nature makes representation learning challenging. Conventional statistical models depend on pre-chosen basis expansions or kernels, limiting the flexibility of data-driven discovery, while many deep-learning pipelines treat functions as fixed-grid vectors, ignoring inherent continuity. In this paper, we introduce Functional Attention with a Mixture-of-Experts (FAME), an end-to-end, fully data-driven framework for function-on-function regression. FAME forms continuous attention by coupling a bidirectional neural controlled differential equation with MoE-driven vector fields to capture intra-functional continuity, and further fuses change to inter-functional dependencies via multi-head cross attention. Extensive experiments on synthetic and real-world functional-regression benchmarks show that FAME achieves state-of-the-art accuracy, strong robustness to arbitrarily sampled discrete observations of functions.

cs.LG↗

arXiv2025

Superradiant dark matter production from primordial black holes: Impact of multiple modes and gravitational wave emission

Rotating primordial black holes (PBHs) in the early universe can emit particles through superradiance, a process particularly efficient when the particle's Compton wavelength is comparable to the PBH's gravitational radius. Superradiance leads to an exponential growth of particle occupation numbers in gravitationally bound states. We present an analysis of heavy bosonic dark matter (DM) production through three gravitational mechanisms: Hawking radiation, superradiant instabilities, and ultraviolet (UV) freeze-in. We consider PBHs that evaporate before Big Bang Nucleosynthesis (BBN). For both scalar and vector DM, our analysis incorporates the evolution of a second superradiant mode. We demonstrate that the growth of a second superradiant mode causes the decay of the first mode, and thus the second mode cannot further enhance the DM abundance beyond that already achieved by the first mode. Our study also reveals that while superradiance generally enhances DM production, gravitational wave (GW) emission from the superradiant cloud may significantly modify this picture. For scalar DM, GW emission reduces the parameter space where superradiance effectively augments relic abundance. For vector DM, rapid GW emission from the superradiant cloud may yield relic abundances below those achieved through Hawking radiation alone. These findings demonstrate that multiple-mode effect and GW emission play critical roles in modeling DM production from PBHs in the early universe.

astro-ph.CO↗

arXiv2025

Tensor State Space-based Dynamic Multilayer Network Modeling

Understanding the complex interactions within dynamic multilayer networks is critical for advancements in various scientific domains. Existing models often fail to capture such networks' temporal and cross-layer dynamics. This paper introduces a novel Tensor State Space Model for Dynamic Multilayer Networks (TSSDMN), utilizing a latent space model framework. TSSDMN employs a symmetric Tucker decomposition to represent latent node features, their interaction patterns, and layer transitions. Then by fixing the latent features and allowing the interaction patterns to evolve over time, TSSDMN uniquely captures both the temporal dynamics within layers and across different layers. The model identifiability conditions are discussed. By treating latent features as variables whose posterior distributions are approximated using a mean-field variational inference approach, a variational Expectation Maximization algorithm is developed for efficient model inference. Numerical simulations and case studies demonstrate the efficacy of TSSDMN for understanding dynamic multilayer networks.

stat.ML↗

arXiv2025

Projective crystal symmetry and topological phases

Quantum states naturally represent symmetry groups, though often in a projective sense. Intriguingly, the projective nature of crystalline symmetries has remained underexplored until very recently. A series of groundbreaking theoretical and experimental studies have now brought this to light, demonstrating that projective representations of crystal symmetries lead to remarkable consequences in condensed matter physics and various artificial crystals, particularly in their connection to topological phenomena. In this article, we explain the basic ideas and notions underpinning these recent developments and share our perspective on this emerging research area. We specifically highlight that the appearance of momentum-space nonsymmorphic symmetry is a unique feature of projective crystal symmetry representations. This, in turn, has the profound consequence of reducing the fundamental domain of momentum space to all possible flat compact manifolds, which include torus and Klein bottle in 2D and the ten platycosms in 3D, presenting a significantly richer landscape for topological structures than conventional settings. Finally, the ongoing efforts and promising future research directions are discussed.

cond-mat.mes-hall↗