Searcharxiv⌕ Search

arXiv subjects

Zihao Gao

Publications and source records attributed to Zihao Gao.

5 recordsLinked to original sources

Confidence-Based Materials Data Quality Assessment via Domain Knowledge and Graph-Structured Learning

The rapid development of materials machine learning relies on large-scale, heterogeneous datasets, yet model reliability is often compromised by imperfect data from diverse experiments, simulations, and database integration. Real-world materials databases may contain measurement errors, failed calculations, missing or duplicate records, and hidden violations of physicochemical constraints. Such problems are difficult to identify because corrupted samples may remain statistically plausible rather than appearing as explicit outliers. In this paper, we propose a confidence-based materials data quality assessment framework that integrates domain knowledge with graph-structured learning. Data quality is formulated as a continuous confidence estimation problem rather than a binary anomaly identification task, enabling sample-level assessment and historical-reference-based batch comparison. A nine-dimensional feature-quality prior and duplicate-aware sample initialization are calibrated using materials domain rules and statistical evidence. Confidence is further refined through sample-feature bipartite graph propagation and weakly supervised graph attention, while a Memory-GAT module provides historical reference scoring. Experiments on 23 ODDS benchmark datasets show that the generic module achieves an average AUC-ROC of 0.866. In five repeated mixed-contamination experiments on mp\_20, rule calibration improves HistGradientBoostingRegressor test $R^2$ from $0.538\pm0.008$ to $0.616\pm0.014$. Graph refinement increases average precision from $0.943\pm0.004$ to 0.953\pm0.004$, while the rule-calibrated configuration has the highest mean test $R^2$ among the evaluated screening configurations. Five repeated MACE subset experiments further show an increase in $R^2$ from $0.590\pm0.053$ to $0.880\pm0.027$ after rule-based screening.

cs.LG↗

TabXLLM: Unlocking Zero- and Few-Shot Tabular Prediction in LLMs via Structural-Prior Reinforcement Learning

Tabular prediction has long been dominated by gradient-boosted decision trees and specialized deep tabular models, while large language models (LLMs) remain difficult to make competitive despite their cross-task adaptability and transparent reasoning traces. We address this gap by incorporating tabular structural priors into LLM post-training. Specifically, we propose Permutation Relative Policy Optimization (PRPO), which operationalizes column-permutation invariance through label-preserving column permutations and two-level advantage estimation. This design converts sparse outcome rewards into denser and more stable optimization signals. Using PRPO, we develop TabXLLM, an 8B LLM for tabular prediction. Extensive experiments on 139 OpenML datasets show that TabXLLM reaches a genuinely competitive regime against strong specialized tabular baselines. It achieves strong fully supervised performance, dominates zero-shot settings, and performs on par with 32-shot strong baselines. Moreover, it substantially outperforms much larger general-purpose and reasoning LLMs, including up to a 53.17% improvement over DeepSeek-R1 (685B). These results show that structural-prior RL post-training is an effective route for making LLMs competitive in tabular prediction.

cs.LG↗

Improved theoretical prediction of nanoparticle sizes with the resistive-pulse technique

With the resistive-pulse technique (RPT), nanopores serve as the nanofluidic sensors of various analytes for their many physical and chemical properties. Here, we focus on the size measurement and its theoretical prediction for sub-200 nm nanoparticles with RPT. Through systematical investigation of the current blockade of nanoparticles across cylindrical nanopores with simulations, Maxwell method considering the shape coefficient and access resistances agrees well with simulation results. However, the widely used integration method of the resistance has distinct deviations in various cases. With the introduction of a correction factor \b{eta} to the integration method, our revised equations can provide good predictions for simulation results. \b{eta} shows a strong dependence on the diameter ratio (d over D) of the nanoparticle and nanopore. Following the same strategy, modified equations are provided for the accurate size prediction for nanoparticles across conical nanopores, where the integration method is the default convenient way. The correction factor \b{eta}' relates to \b{eta} in cylindrical nanopores. \b{eta}' exhibits independence on the pore geometry parameters and diameters of nanoparticles, but dependence on the surface charge density of conical nanopores. Our improved equations can provide theoretical predictions for the accurate size detection of 100-200 nm diameter nanoparticles across cylindrical and conical nanopores.

physics.chem-ph↗

Theoretical prediction of diffusive ionic current through nanopores under salt gradients

In charged nanopores, ionic diffusion current reflects the ionic selectivity and ionic permeability of nanopores which determines the performance of osmotic energy conversion, i.e. the output power and efficiency. Here, theoretical predictions of the diffusive currents through cation-selective nanopores have been developed based on the investigation of diffusive ionic transport under salt gradients with simulations. The ionic diffusion current I satisfies a reciprocal relationship with the pore length I correlates with a/L (a is a constant) in long nanopores. a is determined by the cross-sectional areas of diffusion paths for anions and cations inside nanopores which can be described with a quadratic power of the diameter, and the superposition of a quadratic power and a first power of the diameter, respectively. By using effective concentration gradients instead of nominal ones, the deviation caused by the concentration polarization can be effectively avoided in the prediction of ionic diffusion current. With developed equations of effective concentration difference and ionic diffusion current, the diffusion current across nanopores can be well predicted in cases of nanopores longer than 100 nm and without overlapping of electric double layers. Our results can provide a convenient way for the quantitative prediction of ionic diffusion currents under salt gradients.

physics.chem-ph↗

Classification of stable Dirac and Weyl semimetals with reflection and rotational symmetry

Three dimensional (3D) Dirac semimetal is a novel state of quantum matter, characterized by the gapless bulk four-fold degeneracy near Fermi energy. Soon after its discovery, the classification of stable 3D Dirac semimetals with inversion and rotational symmetry have been studied. However, only ten out of thirty-two point groups have both inversion and rotational symmetry, and we need a more complete classification of stable 3D Dirac semimetals. Here we classify stable 3D Dirac semimetals with reflection symmetry and rotational symmetry in the presence of time reversal symmetry, which belong to seventeen different point groups. These systems include the systems preserving inversion symmetry except $\mathrm{C_{3i}}$. They have two classes of reflection symmetry, with the mirror plane parallel to rotation axis and the mirror plane perpendicular to rotation axis. In both cases two types of Dirac semimetals are determined by four different reflection symmetries. The first type of Dirac semimetals will appear through accidental band crossing (ABC). The second type of Dirac semimetals have a Dirac point at a time reversal invariant momentum (TBC). We show that in both mirror parallel and perpendicular cases, $C_{2,3}$ symmetry can only protect stable Dirac points via TBC, while $C_{4,6}$ symmetry can have stable Dirac points as ABC or TBC. We further discuss that Weyl line nodes and Dirac semimetal can exist in Brillouin zone at the same time using $\mathrm{C_{4v}}$ symmetry as an example. Finally we classify Dirac line nodes and Weyl line nodes to show in which types of mirror plane they can exist.

cond-mat.mes-hall↗