arXiv subjects

Xiaojun Hu

Publications and source records attributed to Xiaojun Hu.

12 recordsLinked to original sources

arXiv2026

Structural Compression for Phylogenetic Inference under Alignment Instability and Indel-Rich Evolution

Phylogenetic inference traditionally relies on aligned characters under substitution models, but this framework becomes less reliable when alignments are unstable or when evolution is dominated by insertions, deletions, repeats, and other structural changes. We adapt Ladderpath as an alignment-free distance approach for phylogenetic inference. Motivated by algorithmic information theory, Ladderpath decomposes sequences into derived, reusable units (``ladderons'', rather than fixed-length $k$-mers) organized hierarchically, from which pairwise distances are computed. The premise is that shared derived sequence structure, including repeated or reused segments that are poorly represented by column-wise substitutions, can retain phylogenetic information. The bacteriophage T7 known lineage, the cpSSR repeat-rich marker, and a cytochrome~$c$ protein dataset confirm that Ladderpath recovers topologies consistent with the known experimental history or with established alignment-based methods. Its advantage emerges under stress: in block-translocation and indel-dominated simulations Ladderpath remains stable while alignment-dependent pipelines deteriorate; on banana mitochondrial and plastome genomes it scales to genome length and captures the expected contrast between organellar histories, all from unaligned input. These results support Ladderpath as an alignment-free, structurally informed method that could complement standard pipelines in cases where higher-order sequence structure carries phylogenetic signal.

q-bio.PE↗

arXiv2026

Text Distance from Nested and Hierarchical Repetitions: A Compression-Based Perspective

We present a new method for structural sequence analysis grounded in Algorithmic Information Theory (AIT). At its core is the Ladderpath approach, which extracts nested and hierarchical relationships among repeated substructures in linguistic sequences -- an instantiation of AIT's principle of describing data through minimal generative programs. These structures are then used to define three distance measures: a normalized compression distance (NCD), and two alternative distances derived directly from the Ladderpath representation. Integrated with a $k$-nearest neighbor classifier, these distances achieve strong and consistent performance across in-distribution, out-of-distribution (OOD), and few-shot text classification tasks. In particular, all three methods outperform both gzip-based NCD and BERT under OOD and low-resource settings. These results demonstrate that the structured representations captured by Ladderpath preserve intrinsic properties of sequences and provide a lightweight, interpretable, and training-free alternative for text modeling. This work highlights the potential of AIT-based approaches for structural and domain-agnostic sequence understanding.

cs.CL↗

arXiv2026

VideoMemory: Toward Consistent Video Generation via Memory Integration

Maintaining consistent characters, props, and environments across multiple shots is a central challenge in narrative video generation. Existing models can produce high-quality short clips but often fail to preserve entity identity and appearance when scenes change or when entities reappear after long temporal gaps. We present VideoMemory, an entity-centric framework that integrates narrative planning with visual generation through a Dynamic Memory Bank. Given a structured script, a multi-agent system decomposes the narrative into shots, retrieves entity representations from memory, and synthesizes keyframes and videos conditioned on these retrieved states. The Dynamic Memory Bank stores explicit visual and semantic descriptors for characters, props, and backgrounds, and is updated after each shot to reflect story-driven changes while preserving identity. This retrieval-update mechanism enables consistent portrayal of entities across distant shots and supports coherent long-form generation. To evaluate this setting, we construct a 54-case multi-shot consistency benchmark covering character-, prop-, and background-persistent scenarios. Extensive experiments show that VideoMemory achieves strong entity-level coherence and high perceptual quality across diverse narrative sequences.

cs.CV↗

arXiv2024

A simple model of decision-making in the application process

In decision-making, individuals often rely on intuition, which can occasionally yield suboptimal outcomes. This study examines the impact of intuitive decision-making on individuals who are confronted with limited position information in the job application process. We propose a measure, the mismatch index, that gauges allocation efficiency by comparing the final application rate to the preset admission rate. By simulation and analytical results, we counter-intuitively find that under the intuitive strategy, acquiring more information does not always lead to more efficient allocation. Additionally, a shift from despondency to a bandwagon effect occurs when the initial application rate surpasses the admission rate, which can be observed in our field experiments. Meanwhile, experimental data also unveil variations in individuals' reliance on intuition, indicating the presence of inherent adventurous and conservative inclinations. To account for these effects, we introduce an enhancement factor into our model. The improved results align well with these real data, showing that compared to mediate competitive scenarios, individuals exhibit a stronger conservative tendency in fierce or less competitive scenarios. These findings offer significant insights into resource allocation, especially in the competitive job market context.

physics.soc-ph↗

arXiv2023

Dual-Stream Pyramid Registration Network

We propose a Dual-Stream Pyramid Registration Network (referred as Dual-PRNet) for unsupervised 3D medical image registration. Unlike recent CNN-based registration approaches, such as VoxelMorph, which explores a single-stream encoder-decoder network to compute a registration fields from a pair of 3D volumes, we design a two-stream architecture able to compute multi-scale registration fields from convolutional feature pyramids. Our contributions are two-fold: (i) we design a two-stream 3D encoder-decoder network which computes two convolutional feature pyramids separately for a pair of input volumes, resulting in strong deep representations that are meaningful for deformation estimation; (ii) we propose a pyramid registration module able to predict multi-scale registration fields directly from the decoding feature pyramids. This allows it to refine the registration fields gradually in a coarse-to-fine manner via sequential warping, and enable the model with the capability for handling significant deformations between two volumes, such as large displacements in spatial domain or slice space. The proposed Dual-PRNet is evaluated on two standard benchmarks for brain MRI registration, where it outperforms the state-of-the-art approaches by a large margin, e.g., having improvements over recent VoxelMorph [2] with 0.683->0.778 on the LPBA40, and 0.511->0.631 on the Mindboggle101, in term of average Dice score. Code is available at: https://github.com/kangmiao15/Dual-Stream-PRNet-Plus.

cs.CV↗

arXiv2022

Chemically induced graphene to diamond transition: a DFT study

The conversion of graphene into diamond is a new way for preparing ultrathin diamond film without pressure. Herein, we investigated the transformation mechanism of surface-hydrogenated bilayer graphene (SHBG) into surface-hydrogenated single-layer diamond (SHSLD) crystal, inserting fifteen kinds of single metal atoms without any pressure, by using the systematical first-principles calculations. Compared with the configuration without metal atom, SHBG can be transformed into SHSLD spontaneously in thermodynamics under the action of single metal atom, and its formation energy can even decrease from 0.82 eV to -5.79 eV under the action of Hf atom. According to our results, the outer electron orbits and atomic radius of metal atom are two important factors that affect the conversion. For the phase transition to occur, the metal atom needs to have enough empty d orbitals, and the radius of the metal atom is in the range of 0.136-0.159 nm. Through further analysis, we find that the p orbitals of carbon atoms and d orbital of metal atom in SHBG will be strongly hybridized, thereby promoting the conversion. The results supply important significance to experimentally prepare diamond without pressure through hydrogenated graphene.

cond-mat.mtrl-sci↗

arXiv2019

Does Environmental Economics lead to patentable research?

In this feasibility study, the impact of academic research from social sciences and humanities on technological innovation is explored through a study of citations patterns of journal articles in patents. Specifically we focus on citations of journals from the field of environmental economics in patents included in an American patent database (USPTO). Three decades of patents have led to a small set of journal articles (85) that are being cited from the field of environmental economics. While this route of measuring how academic research is validated through its role in stimulating technological progress may be rather limited (based on this first exploration), it may still point to a valuable and interesting topic for further research.

cs.DL↗

arXiv2018

Two-dimensional Boron Monosulfides: Semiconducting and Metallic Polymorphs

The typical two-dimensional semiconductors, group \uppercase\expandafter{\romannumeral3A} chalcogenides, have garnered tremendous interest for their outstanding electronic, mechanical, and chemical properties. However, so far, there have been almost no reports on boron monosulfides (BS) binary material. Here, four two-dimensional BS sheets, namely the $α$-, $β$-, $γ$-, and $δ$-BS sheets, are proposed and discussed from $\emph{ab initio}$ calculations. State-of-the-art first-principles calculations reveal all these structures are thermally and dynamically stable, indicating the potential for successful experimental synthesis. Especially, for $α$-BS, it has a calculated exfoliation energy of 0.96 J m$^{-2}$, suggesting the preparation of $α$-BS is feasible by the exfoliation of bulk rhombohedral-BS. Our results show that $α$-, $β$-, $γ$-BS are semiconductors, whereas $δ$-BS is a metallic system. Remarkably, our calculations indicate that $δ$-BS is a superconductor with a large electron-phonon coupling ($λ$ = 1.51) leading a high superconducting critical temperature ($T_c$ $\approx $ 21.56 K), which is the first report of intrinsic superconducting property among all two-dimensional group \uppercase\expandafter{\romannumeral3A} chalcogenides. The desired mechanical and electronic properties render the BS sheets as the promising two-dimensional materials for future applications in nanoelectronics.

cond-mat.mtrl-sci↗

arXiv2018

Two-dimensional Iron Monocarbide with Planar Hypercoordinate Iron and Carbon

We report on the theoretical discovery of Iron monocarbide binary sheets stabilized at two-dimensional confined space, which we call tetragonal-FeC (t-FeC) and orthorhombic-FeC (o-FeC), respectively. From the energy viewpoint, the proposed t-FeC is the global minimum configuration in the 2D space, and each carbon atom is four-coordinated with ambient four Iron atoms. Strikingly, the o-FeC monolayer is an orthorhombic phase with planar pentacoordinate carbon moiety and planar seven-coordinate Fe moiety. To our knowledge, this monolayer is the first example of a simultaneously pentacoordinate carbon and planar seven-coordinate Fe-containing material. State-of-the-art theoretical calculations confirm that all these monolayers have significantly dynamic, mechanical, and thermal stabilities. Among these two monolayers, t-FeC monolayer shows a higher theoretical capacity (395 mAh g-1 ), and can stably adsorb Li up to t-FeCLi3 . Low migration energy barrier is predicted as small as 0.26 eV for Li, which result in the fast diffusion of Li atom on this monolayer. Moreover, electron-phonon calculations coupled with Bardeen-Cooper-Schrieffer arguments suggest t-FeC can be potential two-dimensional superconductors with 6.77 K superconducting transition temperature.

cond-mat.mtrl-sci↗

arXiv2017

D-carbon: A New sp3 Carbon Allotrope

We have investigated the structural, mechanical and electronic properties of D-carbon by ab initio calculations, a new phase of crystalline sp3 carbon (pace group D2h5, Pmma-orthorhombic). Total-energy calculations demonstrate that D-carbon is energetically more favorable than previously proposed T6 structure. State-of-the-art theoretical calculations show that this new phase is dynamic, mechanical, and thermal stable at zero pressure, and more stable than graphite beyond 63.7 GPa. Importantly, the calculations reveal that D-carbon possesses high Vickers hardness (86.58 GPa) and bulk modulus (369 GPa), which are comparable to diamond. D-carbon is a semiconductor with a band gap of 4.33 eV, lower than diamond's gap (5.47 eV). The simulated X-ray diffraction pattern is in satisfactory agreement with the previously experimental data in chimney or detonation soot, suggesting its possible presence in the specimen. Equally important, the possible transition path from diamond to D-carbon has been investigated, indicating a possible approach to synthesize this new phase.

cond-mat.mtrl-sci↗

arXiv2016

Novel bonding patterns and optoelectronic properties of the two-dimensional Si$_x$C$_y$ monolayers

The search of new two-dimensional (2D) materials with novel optical and electronic properties is always desirable for material development. Here, we report a comprehensive theoretical prediction of 2D SiC compounds with different stoichiometries from C-rich to Si-rich. Besides the previously known hexagonal SiC sheet, we identified two types of hitherto-unknown structural motifs with distinctive bonding features. The first type of 2D SiC monolayer, including t-SiC and t-Si$_2$C sheet, can be described by tetragonal lattice. Among them,t-SiC monolayer sheet is featured by each carbon atom binds with four neighboring silicon atoms in almost the same plane, constituting a quasi-planar four-coordinated rectangular moiety. More interestingly, our calculations demonstrate that this structure exhibits a strain-dependent insulator-semimetal transition, suggesting promising applications in strain-dependent optoelectronic sensors. The second type of 2D SiC sheet is featured by silagraphyne with acetylenic linkages(-C$\equiv$C-). Silagraphyne shows both high pore sizes and Poisson's ratio. These properties make them a potentially important material for applications in separation membranes and catalysis. Moreover, one of the proposed structures, $γ$-silagraphyne, is a direct-band-gap semiconductor with a bandgap of 0.89 eV, which has a strong absorption peak in the visible-light region, giving a promising application in ultra-thin transistors, optical sensor devices and solar cell devices.

cond-mat.mtrl-sci↗

arXiv2014

Synthetic biology: From a word to a world

Synthetic biology is one of the battlefields where the main countries fight for the supremacy in science. The word synthetic biology hides a big world, ready to be explored by interdisciplinary research collaborations. The purpose of this investigation is to reveal the what, where, when of the current situation in this emerging field. A keyword search string for the field was constructed and applied in the Web of Science and in the Derwent Innovations Index. In particular, we calculated year based h-type indices for high-frequent keywords.

q-bio.OT↗