arXiv subjects

Xinzijian Liu

Publications and source records attributed to Xinzijian Liu.

12 recordsLinked to original sources

arXiv2026

Inverse Knowledge Search over Verifiable Reasoning: Synthesizing a Scientific Encyclopedia from a Long Chains-of-Thought Knowledge Base

Most scientific materials compress reasoning, presenting conclusions while omitting the derivational chains that justify them. This compression hinders verification by lacking explicit, step-wise justifications and inhibits cross-domain links by collapsing the very pathways that establish the logical and causal connections between concepts. We introduce a scalable framework that decompresses scientific reasoning, constructing a verifiable Long Chain-of-Thought (LCoT) knowledge base and projecting it into an emergent encyclopedia, SciencePedia. Our pipeline operationalizes an endpoint-driven, reductionist strategy: a Socratic agent, guided by a curriculum of around 200 courses, generates approximately 3 million first-principles questions. To ensure high fidelity, multiple independent solver models generate LCoTs, which are then rigorously filtered by prompt sanitization and cross-model answer consensus, retaining only those with verifiable endpoints. This verified corpus powers the Brainstorm Search Engine, which performs inverse knowledge search -- retrieving diverse, first-principles derivations that culminate in a target concept. This engine, in turn, feeds the Plato synthesizer, which narrates these verified chains into coherent articles. The initial SciencePedia comprises approximately 200,000 fine-grained entries spanning mathematics, physics, chemistry, biology, engineering, and computation. In evaluations across six disciplines, Plato-synthesized articles (conditioned on retrieved LCoTs) exhibit substantially higher knowledge-point density and significantly lower factual error rates than an equally-prompted baseline without retrieval (as judged by an external LLM). Built on this verifiable LCoT knowledge base, this reasoning-centric approach enables trustworthy, cross-domain scientific synthesis at scale and establishes the foundation for an ever-expanding encyclopedia.

cs.AI↗

arXiv2026

Deploy-Master: Automating the Deployment of 50,000+ Agent-Ready Scientific Tools in One Day

Open-source scientific software is abundant, yet most tools remain difficult to compile, configure, and reuse, sustaining a small-workshop mode of scientific computing. This deployment bottleneck limits reproducibility, large-scale evaluation, and the practical integration of scientific tools into modern AI-for-Science (AI4S) and agentic workflows. We present Deploy-Master, a one-stop agentic workflow for large-scale tool discovery, build specification inference, execution-based validation, and publication. Guided by a taxonomy spanning 90+ scientific and engineering domains, our discovery stage starts from a recall-oriented pool of over 500,000 public repositories and progressively filters it to 52,550 executable tool candidates under license- and quality-aware criteria. Deploy-Master transforms heterogeneous open-source repositories into runnable, containerized capabilities grounded in execution rather than documentation claims. In a single day, we performed 52,550 build attempts and constructed reproducible runtime environments for 50,112 scientific tools. Each successful tool is validated by a minimal executable command and registered in SciencePedia for search and reuse, enabling direct human use and optional agent-based invocation. Beyond delivering runnable tools, we report a deployment trace at the scale of 50,000 tools, characterizing throughput, cost profiles, failure surfaces, and specification uncertainty that become visible only at scale. These results explain why scientific software remains difficult to operationalize and motivate shared, observable execution substrates as a foundation for scalable AI4S and agentic science.

cs.SE↗

arXiv2025

Bohrium + SciMaster: Building the Infrastructure and Ecosystem for Agentic Science at Scale

AI agents are emerging as a practical way to run multi-step scientific workflows that interleave reasoning with tool use and verification, pointing to a shift from isolated AI-assisted steps toward \emph{agentic science at scale}. This shift is increasingly feasible, as scientific tools and models can be invoked through stable interfaces and verified with recorded execution traces, and increasingly necessary, as AI accelerates scientific output and stresses the peer-review and publication pipeline, raising the bar for traceability and credible evaluation. However, scaling agentic science remains difficult: workflows are hard to observe and reproduce; many tools and laboratory systems are not agent-ready; execution is hard to trace and govern; and prototype AI Scientist systems are often bespoke, limiting reuse and systematic improvement from real workflow signals. We argue that scaling agentic science requires an infrastructure-and-ecosystem approach, instantiated in Bohrium+SciMaster. Bohrium acts as a managed, traceable hub for AI4S assets -- akin to a HuggingFace of AI for Science -- that turns diverse scientific data, software, compute, and laboratory systems into agent-ready capabilities. SciMaster orchestrates these capabilities into long-horizon scientific workflows, on which scientific agents can be composed and executed. Between infrastructure and orchestration, a \emph{scientific intelligence substrate} organizes reusable models, knowledge, and components into executable building blocks for workflow reasoning and action, enabling composition, auditability, and improvement through use. We demonstrate this stack with eleven representative master agents in real workflows, achieving orders-of-magnitude reductions in end-to-end scientific cycle time and generating execution-grounded signals from real workloads at multi-million scale.

cs.AI↗

arXiv2025

DeePMD-kit v3: A Multiple-Backend Framework for Machine Learning Potentials

In recent years, machine learning potentials (MLPs) have become indispensable tools in physics, chemistry, and materials science, driving the development of software packages for molecular dynamics (MD) simulations and related applications. These packages, typically built on specific machine learning frameworks such as TensorFlow, PyTorch, or JAX, face integration challenges when advanced applications demand communication across different frameworks. The previous TensorFlow-based implementation of DeePMD-kit exemplified these limitations. In this work, we introduce DeePMD-kit version 3, a significant update featuring a multi-backend framework that supports TensorFlow, PyTorch, JAX, and PaddlePaddle backends, and demonstrate the versatility of this architecture through the integration of other MLPs packages and of Differentiable Molecular Force Field. This architecture allows seamless backend switching with minimal modifications, enabling users and developers to integrate DeePMD-kit with other packages using different machine learning frameworks. This innovation facilitates the development of more complex and interoperable workflows, paving the way for broader applications of MLPs in scientific research.

physics.chem-ph↗

arXiv2025

The OpenLAM Challenges

Inspired by the success of Large Language Models (LLMs), the development of Large Atom Models (LAMs) has gained significant momentum in scientific computation. Since 2022, the Deep Potential team has been actively pretraining LAMs and launched the OpenLAM Initiative to develop an open-source foundation model spanning the periodic table. A core objective is establishing comprehensive benchmarks for reliable LAM evaluation, addressing limitations in existing datasets. As a first step, the LAM Crystal Philately competition has collected over 19.8 million valid structures, including 1 million on the OpenLAM convex hull, driving advancements in generative modeling and materials science applications.

cs.LG↗

arXiv2024

DPA-2: a large atomic model as a multi-task learner

The rapid advancements in artificial intelligence (AI) are catalyzing transformative changes in atomic modeling, simulation, and design. AI-driven potential energy models have demonstrated the capability to conduct large-scale, long-duration simulations with the accuracy of ab initio electronic structure methods. However, the model generation process remains a bottleneck for large-scale applications. We propose a shift towards a model-centric ecosystem, wherein a large atomic model (LAM), pre-trained across multiple disciplines, can be efficiently fine-tuned and distilled for various downstream tasks, thereby establishing a new framework for molecular modeling. In this study, we introduce the DPA-2 architecture as a prototype for LAMs. Pre-trained on a diverse array of chemical and materials systems using a multi-task approach, DPA-2 demonstrates superior generalization capabilities across multiple downstream tasks compared to the traditional single-task pre-training and fine-tuning methodologies. Our approach sets the stage for the development and broad application of LAMs in molecular and materials simulation research.

physics.chem-ph↗

arXiv2024

Dflow, a Python framework for constructing cloud-native AI-for-Science workflows

In the AI-for-science era, scientific computing scenarios such as concurrent learning and high-throughput computing demand a new generation of infrastructure that supports scalable computing resources and automated workflow management on both cloud and high-performance supercomputers. Here we introduce Dflow, an open-source Python toolkit designed for scientists to construct workflows with simple programming interfaces. It enables complex process control and task scheduling across a distributed, heterogeneous infrastructure, leveraging containers and Kubernetes for flexibility. Dflow is highly observable and can scale to thousands of concurrent nodes per workflow, enhancing the efficiency of complex scientific computing tasks. The basic unit in Dflow, known as an Operation (OP), is reusable and independent of the underlying infrastructure or context. Dozens of workflow projects have been developed based on Dflow, spanning a wide range of projects. We anticipate that the reusability of Dflow and its components will encourage more scientists to publish their workflows and OP components. These components, in turn, can be adapted and reused in various contexts, fostering greater collaboration and innovation in the scientific community.

cs.DC↗

arXiv2024

An Extendable Cloud-Native Alloy Property Explorer

The ability to rapidly evaluate materials properties through atomistic simulation approaches is the foundation of many new artificial intelligence-based approaches to materials identification and design. This depends on the availability of accurate descriptions of atomic bonding through various forms of interatomic potentials. We present an efficient, robust platform for calculating materials properties, i.e., APEX, the Alloy Property Explorer. APEX enables the rapid evolution of interatomic potential development and optimization, which is of particular importance in fine-tuning new classes of general AI-based foundation models to forms that are readily applicable to impacting materials development. APEX is an open-source, extendable, and cloud-native platform for material property calculations using a range of atomistic simulation methodologies that effectively manages diverse computational resources and is built upon user-friendly features including automatic results visualization, web-based platforms and NoSQL database client. It is designed for expert and non-specialist users, lowers the barrier to entry for interdisciplinary research within the "AI for Materials" framework. We describe the foundation and use of APEX, as well as provide an example of its application to properties of titanium for a wide-range of bonding descriptions.

cond-mat.mtrl-sci↗

arXiv2018

Critical role of quantum dynamical effects in the Raman spectroscopy of liquid water

Understanding the Raman spectroscopy at the atomistic level is important for the elucidation of dynamical processes in liquid water. Because the polarizability (or its time derivative) is often a highly nonlinear function of coordinates or/and momenta, we employ the linearized semiclassical initial value representation for quantum dynamical simulations of liquid water (and heavy water) under ambient conditions based on an ab initio based, flexible, polarizable model (the POLI2VS force field). It is shown that quantum dynamical effects play a critical role in reproducing the peaks in the intermediate region between the librational and bending bands, those between the bending and stretching bands, and the double-peak in the stretching band in the experimental isotropic Raman spectrum. In contrast, quantum dynamical effects are important but less decisive in the anisotropic Raman spectrum. By selectively freezing either the intramolecular O-H stretching or H-O-H bending mode, we demonstrate that the peak in the intermediate region (2000-2400 cm-1) of the isotropic Raman spectrum arises from the interplay of the stretching and bending motions while a substantial part of the peak in the same intermediate region of the anisotropic Raman spectrum may be attributed to the combined motion of the bending and librational modes.

physics.chem-ph↗

arXiv2017

Path Integral Molecular Dynamics for Exact Quantum Statistics of Multi-Electronic-State Systems

An exact approach to compute physical properties for general multi-electronic-state (MES) systems in thermal equilibrium is presented. The approach is extended from our recent progress on path integral molecular dynamics (PIMD) [J. Chem. Phys. 145, 024103 (2016); 147, 034109 (2017)] for quantum statistical mechanics when a single potential energy surface is involved. We first define an effective potential function that is numerically favorable for MES-PIMD, and then derive corresponding estimators in MES-PIMD for evaluating various physical properties. Its application to several representative one-dimensional and multi-dimensional models demonstrates that MES-PIMD in principle offers a practical tool in either of the diabatic and adiabatic representations for studying exact quantum statistics of complex/large MES systems when the Born-Oppenheimer approximation, Condon approximation, and harmonic bath approximation are broken.

physics.chem-ph↗

arXiv2017

A unified thermostat scheme for efficient configurational sampling for classical/quantum canonical ensembles via molecular dynamics

We show a unified second-order scheme for constructing simple, robust and accurate algorithms for typical thermostats for configurational sampling for the canonical ensemble. When Langevin dynamics is used, the scheme leads to the BAOAB algorithm that has been recently investigated. We show that the scheme is also useful for other types of thermostat, such as the Andersen thermostat and Nosé-Hoover chain. Two 1-dimensional models and three typical realistic molecular systems that range from the gas phase, clusters, to the condensed phase are used in numerical examples for demonstration. Accuracy may be increased by an order of magnitude for estimating coordinate-dependent properties in molecular dynamics (when the same time interval is used), irrespective of which type of thermostat is applied. The scheme is especially useful for path integral molecular dynamics, because it consistently improves the efficiency for evaluating all thermodynamic properties for any type of thermostat.

physics.chem-ph↗

arXiv2016

A simple and accurate algorithm for path integral molecular dynamics with the Langevin thermostat

We introduce a novel simple algorithm for thermostatting path integral molecular dynamics (PIMD) with the Langevin equation. The staging transformation of path integral beads is employed for demonstration. The optimum friction coefficients for the staging modes in the free particle limit are used for all systems. In comparison to the path integral Langevin equation (PILE) thermostat, the new algorithm exploits a different order of splitting for the phase space propagator associated to the Langevin equation. While the error analysis is made for both algorithms, they are also employed in the PIMD simulations of three realistic systems (the H2O molecule, liquid para-hydrogen, and liquid water) for comparison. It is shown that the new thermostat increases the time interval of PIMD by a factor of 4~6 or more for achieving the same accuracy. In addition, supplemental material shows the error analysis made for the algorithms when the normal-mode transformation of path integral beads is used.

cond-mat.stat-mech↗