arXiv subjects

Weile Jia

Publications and source records attributed to Weile Jia.

At least 19 recordsLinked to original sources

arXiv2026

Scalable High-Fidelity Macromolecular Docking for GPU-Accelerated Supercomputers

Flexible macromolecular docking offers high-fidelity predictions of biomolecular interactions, but remains prohibitively expensive at scale. Among existing approaches, LightDock leverages Glowworm Swarm Optimization (GSO) for accuracy, yet suffers from limited parallelism, irregular computation, and severe load imbalance, preventing efficient execution on GPU supercomputers. We present SparkleDock, a scalable GSO-based docking framework enabling near-real-time flexible docking. We redesign GSO to expose massive fine-grained parallelism at the glowworm-agent level, and restructure the dominant energy scoring computation into a Tensor Core-compatible formulation, enabling efficient execution of irregular pairwise interactions through structured matrix operations. We further introduce a performance-model-driven scheduling for load balancing and out-of-core scaling across GPUs. SparkleDock achieves 9.7 $\times$ and 18.9 $\times$ speedups over LightDock on single A100 and H100 GPU, and delivers over two orders of magnitude acceleration at scale. On 512 GPUs, it reduces docking time from hours to seconds, enabling large-scale, high-fidelity virtual screening previously impractical with flexible docking.

cs.DC↗

arXiv2026

Libra: Taming Attention Workload Skew in Long-Context LLM Training with Bounded Sequence Pool

Long-context LLM training suffers from a load-balancing problem that sequence packing does not solve. Packing samples into fixed-token sequences balances memory and linear-cost operators, but the dominant attention cost scales with the sum of squared sequence lengths. Thus, equally sized packed sequences drawn from a long-tailed corpus can carry substantially different attention workloads, creating data-parallel stragglers and pipeline bubbles. Existing approaches either balance at the granularity of sequences or microbatches, where an outlier can dominate an assignment, or disaggregate attention over a global worker pool whose communication domain grows with the data-parallel (DP) degree. We present Libra, which operationalizes the law of large numbers (LLN) as a scaling principle for load balancing: the attention-balancing pool need not grow with the DP degree. Libra groups packed sequences and their CP groups into fixed-size sequence pools. As DP scales out, Libra adds pools rather than enlarging each one, bounding every attention exchange. Variance-Reduced Sequence Placement makes this effective for finite, long-tailed workloads by co-locating sequences with complementary attention workloads to reduce residual inter-pool skew. Within each pool, Tiled Attention Pooling dispatches sequence-head SH-Tiles across GPUs, while a pipelined runtime overlaps tile exchange with attention. Libra exposes a drop-in context-parallel attention operator and a pluggable data sampler, requiring no changes to model layers, optimizers, or pipeline schedules. On Qwen3-Turbo training with 256K- and 1M-token workloads, Libra improves end-to-end throughput by up to 2.54x over Ulysses, with up to 3.14x worst-step straggler-attention speedup in microbenchmarks. Libra has run for hundreds of thousands of GPU-hours in production on jobs spanning 32K to 1M tokens while preserving training semantics.

cs.DC↗

arXiv2026

Direct Model State Migration for Elastic Training of Large Language Models

Large language model (LLM) training shall adapt to dynamic resources in shared clusters to tackle the elasticity, including passive preemption and optimistic scaling. State migration across device sets is required when altering the hybrid-parallel configuration due to dynamic resources. Existing solutions rely on checkpoint-based mechanisms, which persist complete states to storage for resuming with re-assigned resources, forcing all GPUs to stall when transferring model states. Despite optimization efforts, checkpoint-based solutions incur prohibitive latency due to data movement across memory hierarchies. We propose ETC, a checkpoint-free state migration framework for elastic hybrid-parallel LLM training. We exploits the state locality to minimize inter-GPU data movement, replacing storage persistence with direct peer-to-peer communication. Besides, we eliminate node fragmentation through communication coalescing. Integrated with Megatron-LM, ETC reduces migration overhead by 2.33$\times$ to 6.37$\times$ compared to checkpoint-based solutions across diverse parallel configurations. By enabling efficient migration, ETC unlocks practical elastic training in production environments.

cs.DC↗

arXiv2026

JanusPipe: Efficient Pipeline Parallel Training for Machine Learning Interatomic Potentials

Discovering atom-level phenomena requires molecular dynamics (MD) simulations with ab initio accuracy. Machine learning interatomic potentials (MLIPs) enable stable, high-accuracy MD simulations, and their models exhibit scaling-law trends similar to large language models. However, the lack of scalable and efficient distributed training systems for conservative MLIPs makes them difficult to scale. This is because conservative MLIPs inherently follow a double-backward execution pattern, which involves computing gradients during the forward pass. This pattern creates a mismatch with existing distributed training systems, especially for pipeline parallelism. Therefore, we present JanusPipe, an efficient 3D-parallel (PP/DP/GP) training system tailored for conservative MLIPs. It integrates SymFold to enable memory-efficient pipeline parallelism for conservative MLIPs, and WaveK to reduce pipeline bubbles by balancing the four-phase compute time. Experimental results on 32 GPUs show that JanusPipe improves throughput by $1.51\times$ and $1.45\times$ on average over 1F1B and Hanayo, respectively.

cs.DC↗

arXiv2026

Compact SO(3) Equivariant Atomistic Foundation Models via Structural Pruning

SO(3) equivariant graph neural networks have become the dominant paradigm for atomistic foundation models, achieving high accuracy and data efficiency by building rotational symmetry directly into the architecture. Yet the computational cost of their higher-order tensor operations creates a tough trade-off between model accuracy and inference efficiency. In this paper, we propose a structural pruning method for SO(3) equivariant atomistic foundation models to bridge this accuracy-efficiency gap. The pruning is applied along the channel and order dimensions, with each irreducible representation kept or removed as a complete block, thereby retaining SO(3) equivariance. Starting from a large checkpoint, the pruned model substantially reduces the inference cost while retaining higher accuracy than an independently trained small model. The pruned MACE-MP model outperforms the official from-scratch trained small model on 7 of 9 metrics on the Matbench Discovery leaderboard. In terms of efficiency, compressed MACE-MP and MACE-OFF models contain 1.5$\times$ to 4$\times$ fewer parameters and require 2.5$\times$ to 4$\times$ less pre-training compute than training a small model from scratch. For downstream applications, fine-tuning the pruned model reduces energy and force errors by 70.1% and 34.4% compared to training task-specific models from scratch across eight representative downstream datasets. We demonstrate that the method generalizes to other SO(3) equivariant architectures (SevenNet, eSCN) and can be combined with quantization and knowledge distillation for further gains.

cs.LG↗

arXiv2026

Breaking the Training Barrier of Billion-Parameter Universal Machine Learning Interatomic Potentials

Universal Machine Learning Interatomic Potentials (uMLIPs), pre-trained on massively diverse datasets encompassing inorganic materials and organic molecules across the entire periodic table, serve as foundational models for quantum-accurate physical simulations. However, uMLIP training requires second-order derivatives, which lack corresponding parallel training frameworks; moreover, scaling to the billion-parameter regime causes explosive growth in computation and communication overhead, making its training a tremendous challenge. We introduce MatRIS-MoE, a billion-parameter Mixture-of-Experts model built upon invariant architecture, and {Janus}, a pioneering high-dimensional distributed training framework for uMLIPs with hardware-aware optimizations. Deployed across two Exascale supercomputers, our code attains a peak performance of 1.2/1.0 EFLOPS (24\%/{35.5\%} of theoretical peak) in single precision at over 90\% parallel efficiency, compressing the training of billion-parameter uMLIPs from weeks to hours. This work establishes a new high-water mark for AI-for-Science (AI4S) foundation models at Exascale and provides essential infrastructure for rapid scientific discovery.

cs.DC↗

arXiv2026

SparseBalance: Load-Balanced Long Context Training with Dynamic Sparse Attention

While sparse attention mitigates the computational bottleneck of long-context LLM training, its distributed training process exhibits extreme heterogeneity in both \textit{1)} sequence length and \textit{2)} sparsity sensitivity, leading to a severe imbalance problem and sub-optimal model accuracy. Existing algorithms and training frameworks typically focus on single issue, failing to systematically co-optimize these two problems. Therefore, we propose SparseBalance, a novel algorithm-system co-design framework, which exploits the sparsity and sequence heterogeneity to optimize model accuracy and system efficiency jointly. First, we propose workload-aware dynamic sparsity tuning, which employs a bidirectional sparsity adjustment to eliminate stragglers and exploit inherent bubbles for free accuracy. Second, we propose a sparsity-aware batching strategy to achieve coarse-grained balance, which complements dynamic sparsity tuning. Experimental results demonstrate that SparseBalance achieves up to a 1.33$\times$ end-to-end speedup while still improving the long-context capability by 0.46\% on the LongBench benchmark.

cs.LG↗

arXiv2026

MatRIS: Toward Reliable and Efficient Pretrained Machine Learning Interatomic Potentials

Foundation MLIPs demonstrate broad applicability across diverse material systems and have emerged as a powerful and transformative paradigm in chemical and computational materials science. Equivariant MLIPs achieve state-of-the-art accuracy in a wide range of benchmarks by incorporating equivariant inductive bias. However, the reliance on tensor products and high-degree representations makes them computationally costly. This raises a fundamental question: as quantum mechanical-based datasets continue to expand, can we develop a more compact model to thoroughly exploit high-dimensional atomic interactions? In this work, we present MatRIS (\textbf{Mat}erials \textbf{R}epresentation and \textbf{I}nteraction \textbf{S}imulation), an invariant MLIP that introduces attention-based modeling of three-body interactions. MatRIS leverages a novel separable attention mechanism with linear complexity $O(N)$, enabling both scalability and expressiveness. MatRIS delivers accuracy comparable to that of leading equivariant models on a wide range of popular benchmarks (Matbench-Discovery, MatPES, MDR phonon, Molecular dataset, etc). Taking Matbench-Discovery as an example, MatRIS achieves an F1 score of up to 0.847 and attains comparable accuracy at a lower training cost. The work indicates that our carefully designed invariant models can match or exceed the accuracy of equivariant models at a fraction of the cost, shedding light on the development of accurate and efficient MLIPs.

cs.LG↗

arXiv2026

fix pimd/langevin: An Efficient Implementation of Path Integral Molecular Dynamics in LAMMPS

Path integral molecular dynamics (PIMD), which maps a quantum particle onto a fictitious classical system of ring polymers and propagates the "beads" of this extended classical system using molecular dynamics, is widely used to capture nuclear quantum effects (NQEs) in molecular simulations. Accurate PIMD calculations typically require a large number of beads and are therefore computationally demanding. While software packages such as i-PI offer comprehensive PIMD functionality, the high efficiency of simulations driven by machine learning interatomic potentials, such as Deep Potential (DP), calls for more efficient PIMD implementations that fully exploit modern massively parallel supercomputers. Here we present fix pimd/langevin, an efficient PIMD implementation in LAMMPS that supports commonly used features and leverages the Message Passing Interface architecture of LAMMPS to achieve high computational efficiency. We demonstrate the usage and validate the correctness of our code using liquid water as a representative example, and provide a comprehensive overview of the supported features. Then we discuss several important technical aspects of the implementation. Using DP simulations of water as a benchmark, we show that our implementation achieves several-fold acceleration compared to i-PI. Finally, we report strong and weak scaling results that demonstrate the favorable parallel performance of our code.

physics.chem-ph↗

arXiv2025

Exploring Landscapes for Better Minima along Valleys

Finding lower and better-generalizing minima is crucial for deep learning. However, most existing optimizers stop searching the parameter space once they reach a local minimum. Given the complex geometric properties of the loss landscape, it is difficult to guarantee that such a point is the lowest or provides the best generalization. To address this, we propose an adaptor "E" for gradient-based optimizers. The adapted optimizer tends to continue exploring along landscape valleys (areas with low and nearly identical losses) in order to search for potentially better local minima even after reaching a local minimum. This approach increases the likelihood of finding a lower and flatter local minimum, which is often associated with better generalization. We also provide a proof of convergence for the adapted optimizers in both convex and non-convex scenarios for completeness. Finally, we demonstrate their effectiveness in an important but notoriously difficult training scenario, large-batch training, where Lamb is the benchmark optimizer. Our testing results show that the adapted Lamb, ALTO, increases the test accuracy (generalization) of the current state-of-the-art optimizer by an average of 2.5% across a variety of large-batch training tasks. This work potentially opens a new research direction in the design of optimization algorithms.

cs.LG↗

arXiv2025

Deep Learning-Enabled Supercritical Flame Simulation at Detailed Chemistry and Real-Fluid Accuracy Towards Trillion-Cell Scale

For decades, supercritical flame simulations incorporating detailed chemistry and real-fluid transport have been limited to millions of cells, constraining the resolved spatial and temporal scales of the physical system. We optimize the supercritical flame simulation software DeepFlame -- which incorporates deep neural networks while retaining the real-fluid mechanical and chemical accuracy -- from three perspectives: parallel computing, computational efficiency, and I/O performance. Our highly optimized DeepFlame achieves supercritical liquid oxygen/methane (LOX/\ce{CH4}) turbulent combustion simulation of up to 618 and 154 billion cells with unprecedented time-to-solution, attaining 439/1186 and 187/316 PFlop/s (32.3\%/21.8\% and 37.4\%/31.8\% of the peak) in FP32/mixed-FP16 precision on Sunway (98,304 nodes) and Fugaku (73,728 nodes) supercomputers, respectively. This computational capability surpasses existing capacities by three orders of magnitude, enabling the first practical simulation of rocket engine combustion with >100 LOX/\ce{CH4} injectors. This breakthrough establishes high-fidelity supercritical flame modeling as a critical design tool for next-generation rocket propulsion and ultra-high energy density systems.

cs.DC↗

arXiv2025

WISCA: A Lightweight Model Transition Method to Improve LLM Training via Weight Scaling

Transformer architecture gradually dominates the LLM field. Recent advances in training optimization for Transformer-based large language models (LLMs) primarily focus on architectural modifications or optimizer adjustments. However, these approaches lack systematic optimization of weight patterns during training. Weight pattern refers to the distribution and relative magnitudes of weight parameters in a neural network. To address this issue, we propose a Weight Scaling method called WISCA to enhance training efficiency and model quality by strategically improving neural network weight patterns without changing network structures. By rescaling weights while preserving model outputs, WISCA indirectly optimizes the model's training trajectory. Experiments demonstrate that WISCA significantly improves convergence quality (measured by generalization capability and loss reduction), particularly in LLMs with Grouped Query Attention (GQA) architectures and LoRA fine-tuning tasks. Empirical results show 5.6% average improvement on zero-shot validation tasks and 2.12% average reduction in training perplexity across multiple architectures.

cs.LG↗

arXiv2025

Large-scale Neural Network Quantum States for ab initio Quantum Chemistry Simulations on Fugaku

Solving quantum many-body problems is one of the fundamental challenges in quantum chemistry. While neural network quantum states (NQS) have emerged as a promising computational tool, its training process incurs exponentially growing computational demands, becoming prohibitively expensive for large-scale molecular systems and creating fundamental scalability barriers for real-world applications. To address above challenges, we present \ours, a high-performance NQS training framework for \textit{ab initio} electronic structure calculations. First, we propose a scalable sampling parallelism strategy with multi-layers workload division and hybrid sampling scheme, which break the scalability barriers for large-scale NQS training. Then, we introduce multi-level parallelism local energy parallelism, enabling more efficient local energy computation. Last, we employ cache-centric optimization for transformer-based \textit{ansatz} and incorporate it with sampling parallelism strategy, which further speedup up the NQS training and achieve stable memory footprint at scale. Experiments demonstrate that \ours accelerate NQS training with up to 8.41x speedup and attains a parallel efficiency up to 95.8\% when scaling to 1,536 nodes.

cs.DC↗

arXiv2025

A Graph Neural Network for the Era of Large Atomistic Models

Foundation models, or large atomistic models (LAMs), aim to universally represent the ground-state potential energy surface (PES) of atomistic systems as defined by density functional theory (DFT). The scaling law is pivotal in the development of large models, suggesting that their generalizability in downstream tasks consistently improves with increased model size, expanded training datasets, and larger computational budgets. In this study, we present DPA3, a multi-layer graph neural network founded on line graph series (LiGS), designed explicitly for the era of LAMs. We demonstrate that the generalization error of the DPA3 model adheres to the scaling law. The scalability in the number of model parameters is attained by stacking additional layers within DPA3. Additionally, the model employs a dataset encoding mechanism that decouples the scaling of training data size from the model size within its multi-task training framework. When trained as problem-oriented potential energy models, the DPA3 model exhibits superior accuracy in the majority of benchmark cases, encompassing systems with diverse features, including molecules, bulk materials, surface and cluster catalysts, two-dimensional materials, and battery materials. When trained as a LAM on the OpenLAM-v1 dataset, the DPA-3.1-3M model exhibits lowest overall zero-shot generalization error across 12 downstream tasks spanning a diverse array of research domains. This performance suggests superior accuracy as an out-of-the-box potential model, requiring minimal fine-tuning data for downstream scientific applications.

physics.comp-ph↗

arXiv2025

Skrull: Towards Efficient Long Context Fine-tuning through Dynamic Data Scheduling

Long-context supervised fine-tuning (Long-SFT) plays a vital role in enhancing the performance of large language models (LLMs) on long-context tasks. To smoothly adapt LLMs to long-context scenarios, this process typically entails training on mixed datasets containing both long and short sequences. However, this heterogeneous sequence length distribution poses significant challenges for existing training systems, as they fail to simultaneously achieve high training efficiency for both long and short sequences, resulting in sub-optimal end-to-end system performance in Long-SFT. In this paper, we present a novel perspective on data scheduling to address the challenges posed by the heterogeneous data distributions in Long-SFT. We propose Skrull, a dynamic data scheduler specifically designed for efficient long-SFT. Through dynamic data scheduling, Skrull balances the computation requirements of long and short sequences, improving overall training efficiency. Furthermore, we formulate the scheduling process as a joint optimization problem and thoroughly analyze the trade-offs involved. Based on those analysis, Skrull employs a lightweight scheduling algorithm to achieve near-zero cost online scheduling in Long-SFT. Finally, we implement Skrull upon DeepSpeed, a state-of-the-art distributed training system for LLMs. Experimental results demonstrate that Skrull outperforms DeepSpeed by 3.76x on average (up to 7.54x) in real-world long-SFT scenarios.

cs.LG↗

arXiv2025

Scaling Neural-Network-Based Molecular Dynamics with Long-Range Electrostatic Interactions to 51 Nanoseconds per Day

Neural network-based molecular dynamics (NNMD) simulations incorporating long-range electrostatic interactions have significantly extended the applicability to heterogeneous and ionic systems, enabling effective modeling critical physical phenomena such as protein folding and dipolar surface and maintaining ab initio accuracy. However, neural network inference and long-range force computation remain the major bottlenecks, severely limiting simulation speed. In this paper, we target DPLR, a state-of-the-art NNMD package that supports long-range electrostatics, and propose a set of comprehensive optimizations to enhance computational efficiency. We introduce (1) a hardware-offloaded FFT method to reduce the communication overhead; (2) an overlapping strategy that hides long-range force computations using a single core per node, and (3) a ring-based load balancing method that enables atom-level task evenly redistribution with minimal communication overhead. Experimental results on the Fugaku supercomputer show that our work achieves a 37x performance improvement, reaching a maximum simulation speed of 51 ns/day.

cs.DC↗

arXiv2025

Efficient Long Context Fine-tuning with Chunk Flow

Long context fine-tuning of large language models(LLMs) involves training on datasets that are predominantly composed of short sequences and a small proportion of longer sequences. However, existing approaches overlook this long-tail distribution and employ training strategies designed specifically for long sequences. Moreover, these approaches also fail to address the challenges posed by variable sequence lengths during distributed training, such as load imbalance in data parallelism and severe pipeline bubbles in pipeline parallelism. These issues lead to suboptimal training performance and poor GPU resource utilization. To tackle these problems, we propose a chunk-centric training method named ChunkFlow. ChunkFlow reorganizes input sequences into uniformly sized chunks by consolidating short sequences and splitting longer ones. This approach achieves optimal computational efficiency and balance among training inputs. Additionally, ChunkFlow incorporates a state-aware chunk scheduling mechanism to ensure that the peak memory usage during training is primarily determined by the chunk size rather than the maximum sequence length in the dataset. Integrating this scheduling mechanism with existing pipeline scheduling algorithms further enhances the performance of distributed training. Experimental results demonstrate that, compared with Megatron-LM, ChunkFlow can be up to 4.53x faster in the long context fine-tuning of LLMs. Furthermore, we believe that ChunkFlow serves as an effective solution for a broader range of scenarios, such as long context continual pre-training, where datasets contain variable-length sequences.

cs.DC↗

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↗