Sampling physically valid ligand-binding poses remains a major challenge in molecular docking, particularly for unseen or structurally diverse targets. We introduce PocketVina, a fast and memory-efficient, search-based docking framework that combines pocket prediction with systematic multi-pocket exploration. We evaluate PocketVina across four established benchmarks--PDBbind2020 (timesplit and unseen), DockGen, Astex, and PoseBusters--and observe consistently strong performance in sampling physically valid docking poses. PocketVina achieves state-of-the-art performance when jointly considering ligand RMSD and physical validity (PB-valid), while remaining competitive with deep learning-based approaches in terms of RMSD alone, particularly on structurally diverse and previously unseen targets. PocketVina also maintains state-of-the-art physically valid docking accuracy across ligands with varying degrees of flexibility. We further introduce TargetDock-AI, a benchmarking dataset we curated, consisting of over 500000 protein-ligand pairs, and a partition of the dataset labeled with PubChem activity annotations. On this large-scale dataset, PocketVina successfully discriminates active from inactive targets, outperforming a deep learning baseline while requiring significantly less GPU memory and runtime. PocketVina offers a robust and scalable docking strategy that requires no task-specific training and runs efficiently on standard GPUs, making it well-suited for high-throughput virtual screening and structure-based drug discovery.
Large language models (LLMs) deliver strong performance but are difficult to deploy due to high memory and compute costs. While pruning reduces these demands, most methods ignore activation sparsity observed at runtime. We reinterpret activation sparsity as dynamic structured weight sparsity and propose DuoGPT, a unified framework that constructs dual-sparse (spMspV) workloads by combining unstructured weight pruning with activation sparsity. To preserve accuracy, we extend the Optimal Brain Compression (OBC) framework with activation-aware calibration and introduce output residuals from the dense model as correction terms. We further optimize the solution for efficient GPU execution, enabling scalability to billion-parameter LLMs. Evaluations on LLaMA-2 and LLaMA-3 show that DuoGPT outperforms state-of-the-art structured pruning methods by up to 9.17% accuracy at an iso-speedup of 1.39$\times$ compared to the baseline dense model.
We introduce Confucius3-Math, an open-source large language model with 14B parameters that (1) runs efficiently on a single consumer-grade GPU; (2) achieves SOTA performances on a range of mathematical reasoning tasks, outperforming many models with significantly larger sizes. In particular, as part of our mission to enhancing education and knowledge dissemination with AI, Confucius3-Math is specifically committed to mathematics learning for Chinese K-12 students and educators. Built via post-training with large-scale reinforcement learning (RL), Confucius3-Math aligns with national curriculum and excels at solving main-stream Chinese K-12 mathematical problems with low cost. In this report we share our development recipe, the challenges we encounter and the techniques we develop to overcome them. In particular, we introduce three technical innovations: Targeted Entropy Regularization, Recent Sample Recovery and Policy-Specific Hardness Weighting. These innovations encompass a new entropy regularization, a novel data scheduling policy, and an improved group-relative advantage estimator. Collectively, they significantly stabilize the RL training, improve data efficiency, and boost performance. Our work demonstrates the feasibility of building strong reasoning models in a particular domain at low cost. We open-source our model and code at https://github.com/netease-youdao/Confucius3-Math.
Computational astrochemical models are essential for helping us interpret and understand the observations of different astrophysical environments. In the age of high-resolution telescopes such as JWST and ALMA, the substructure of many objects can be resolved, raising the need for astrochemical modeling at these smaller scales, meaning that the simulations of these objects need to include both the physics and chemistry to accurately model the observations. The computational cost of the simulations coupling both the three-dimensional hydrodynamics and chemistry is enormous, creating an opportunity for surrogate models that can effectively substitute the chemical solver. In this work we present surrogate models that can replace the original chemical code, namely Latent Augmented Neural Ordinary Differential Equations. We train these surrogate architectures on three datasets of increasing physical complexity, with the last dataset derived directly from a three-dimensional simulation of a molecular cloud using a Photodissociation Region (PDR) code, 3D-PDR. We show that these surrogate models can provide speedup and reproduce the original observable column density maps of the dataset. This enables the rapid inference of the chemistry (on the GPU), allowing for the faster statistical inference of observations or increasing the resolution in hydrodynamical simulations of astrophysical environments.
Long sequences are critical for applications like RAG, long document summarization, multi-modality, etc., and modern LLMs, like Llama 4 Scout, support max sequence length of up to 10 million tokens. However, outside of enterprise labs, long sequence training is challenging for the AI community with limited system support in the open-source space. Out-of-box, even on a modern NVIDIA H100 80GB GPU cluster, training Llama 8B model with sequence over 32K runs out of memory on a basic Hugging Face (HF) model due to two reasons: i) LLM training workloads are not optimized to fully leverage a single GPU memory, ii) existing solutions for leveraging multiple GPU memory are not easily available to HF models, making long sequence training inaccessible. We address this with Arctic Long Sequence Training (ALST). It offers a combination of attention-agnostic single GPU and multi-GPU memory optimizations, that enables it to support out-of-box training of multi-million sequence length for a wide variety of HF models. ALST supports training Meta's Llama 8B model with 500K sequence length on a single H100 GPU, 3.7M on a single 8xH100 GPU node, and over 15M on a 4 node cluster, an increase of over 400x compared to the 32K baseline for the latter. ALST is fully compatible with HF models and open-sourced via Deepspeed https://www.deepspeed.ai/tutorials/ulysses-alst-sequence-pallellism/ and Arctic Training https://github.com/snowflakedb/ArcticTraining/blob/main/projects/sequence-parallelism/README.md.
We present a parallel GPU-accelerated solver for branch Model Predictive Control problems. Based on iterative LQR methods, our solver exploits the tree-sparse structure and implements temporal parallelism using the parallel scan algorithm. Consequently, the proposed solver enables parallelism across both the prediction horizon and the scenarios. In addition, we utilize an augmented Lagrangian method to handle general inequality constraints. We compare our solver with state-of-the-art numerical solvers in two automated driving applications. The numerical results demonstrate that, compared to CPU-based solvers, our solver achieves competitive performance for problems with short horizons and small-scale trees, while outperforming other solvers on large-scale problems.
Dynamic facial emotion is essential for believable AI-generated avatars; however, most systems remain visually inert, limiting their utility in high-stakes simulations such as virtual training for investigative interviews with abused children. We introduce and evaluate a real-time architecture fusing Unreal Engine 5 MetaHuman rendering with NVIDIA Omniverse Audio2Face to translate vocal prosody into high-fidelity facial expressions on photorealistic child avatars. We implemented a distributed two-PC setup that decouples language processing and speech synthesis from GPU-intensive rendering, designed to support low-latency interaction in desktop and VR environments. A between-subjects study ($N=70$) using audio+visual and visual-only conditions assessed perceptual impacts as participants rated emotional clarity, facial realism, and empathy for two avatars expressing joy, sadness, and anger. Results demonstrate that avatars could express emotions recognizably, with sadness and joy achieving high identification rates. However, anger recognition significantly dropped without audio, highlighting the importance of congruent vocal cues for high-arousal emotions. Interestingly, removing audio boosted perceived facial realism, suggesting that audiovisual desynchrony remains a key design challenge. These findings confirm the technical feasibility of generating emotionally expressive avatars and provide guidance for improving non-verbal communication in sensitive training simulations.
Gaussian Splatting have demonstrated remarkable novel view synthesis performance at high rendering frame rates. Optimization-based inverse rendering within complex capture scenarios remains however a challenging problem. A particular case is modelling complex surface light interactions for highly reflective scenes, which results in intricate high frequency specular radiance components. We hypothesize that such challenging settings can benefit from increased representation power. We hence propose a method that tackles this issue through a geometrically and physically grounded Gaussian Splatting borne radiance field, where normals and material properties are spatially variable in the primitive's local space. Using per-primitive texture maps for this purpose, we also propose to harness the GPU hardware to accelerate rendering at test time via unified material texture atlas.
It has become mainstream to deploy Convolutional Neural Network (CNN) models on ubiquitous Internet of Things (IoT) devices with the help of the cloud to provide users with a variety of high-quality services. Most existing methods have two limitations: (i) low robustness in handling corrupted image data collected by IoT devices; and (ii) high consumption of computational and transmission resources. To this end, we propose the Grouped NonLinear transformation generation method (GroupNL), which generates diversified feature maps by utilizing data-agnostic Nonlinear Transformation Functions (NLFs) to improve the robustness of the CNN model. Specifically, partial convolution filters are designated as seed filters in a convolutional layer, and a small set of feature maps, i.e., seed feature maps, are first generated based on vanilla convolution operation. Then, we split seed feature maps into several groups, each with a set of different NLFs, to generate corresponding diverse feature maps with in-place nonlinear processing. Moreover, GroupNL effectively reduces the parameter transmission between multiple nodes during model training by setting the hyperparameters of NLFs to random initialization and not updating them during model training, and reduces the computing resources by using NLFs to generate feature maps instead of most feature maps generated based on sliding windows. Experimental results on CIFAR-10, GTSRB, CIFAR-10-C, Icons50, and ImageNet-1K datasets in NVIDIA RTX GPU platforms show that the proposed GroupNL outperforms other state-of-the-art methods in model robust and training acceleration. Specifically, on the Icons-50 dataset, the accuracy of GroupNL-ResNet-18 achieves approximately 2.86% higher than the vanilla ResNet-18. GroupNL improves training speed by about 53% compared to vanilla CNN when trained on a cluster of 8 NVIDIA RTX 4090 GPUs on the ImageNet-1K dataset.
We introduce a novel framework for learning vector representations of tree-structured geometric data focusing on 3D vascular networks. Our approach employs two sequentially trained Transformer-based autoencoders. In the first stage, the Vessel Autoencoder captures continuous geometric details of individual vessel segments by learning embeddings from sampled points along each curve. In the second stage, the Vessel Tree Autoencoder encodes the topology of the vascular network as a single vector representation, leveraging the segment-level embeddings from the first model. A recursive decoding process ensures that the reconstructed topology is a valid tree structure. Compared to 3D convolutional models, this proposed approach substantially lowers GPU memory requirements, facilitating large-scale training. Experimental results on a 2D synthetic tree dataset and a 3D coronary artery dataset demonstrate superior reconstruction fidelity, accurate topology preservation, and realistic interpolations in latent space. Our scalable framework, named VeTTA, offers precise, flexible, and topologically consistent modeling of anatomical tree structures in medical imaging.
Location-based services often require users to share sensitive locational data, raising privacy concerns due to potential misuse or exploitation by untrusted servers. In response, we present VeLoPIR, a versatile location-based private information retrieval (PIR) system designed to preserve user privacy while enabling efficient and scalable query processing. VeLoPIR introduces three operational modes-interval validation, coordinate validation, and identifier matching-that support a broad range of real-world applications, including information and emergency alerts. To enhance performance, VeLoPIR incorporates multi-level algorithmic optimizations with parallel structures, achieving significant scalability across both CPU and GPU platforms. We also provide formal security and privacy proofs, confirming the system's robustness under standard cryptographic assumptions. Extensive experiments on real-world datasets demonstrate that VeLoPIR achieves up to 11.55 times speed-up over a prior baseline. The implementation of VeLoPIR is publicly available at https://github.com/PrivStatBool/VeLoPIR.
Spiking Neural Networks (SNNs) have the potential to drastically reduce the energy requirements of AI systems. However, mainstream accelerators like GPUs and TPUs are designed for the high arithmetic intensity of standard ANNs so are not well-suited to SNN simulation. FPGAs are well-suited to applications with low arithmetic intensity as they have high off-chip memory bandwidth and large amounts of on-chip memory. Here, we present a novel RISC-V-based soft vector processor (FeNN), tailored to simulating SNNs on FPGAs. Unlike most dedicated neuromorphic hardware, FeNN is fully programmable and designed to be integrated with applications running on standard computers from the edge to the cloud. We demonstrate that, by using stochastic rounding and saturation, FeNN can achieve high numerical precision with low hardware utilisation and that a single FeNN core can simulate an SNN classifier faster than both an embedded GPU and the Loihi neuromorphic system.
The widespread adoption of outsourced neural network inference presents significant privacy challenges, as sensitive user data is processed on untrusted remote servers. Secure inference offers a privacy-preserving solution, but existing frameworks suffer from high computational overhead and communication costs, rendering them impractical for real-world deployment. We introduce SecONNds, a non-intrusive secure inference framework optimized for large ImageNet-scale Convolutional Neural Networks. SecONNds integrates a novel fully Boolean Goldreich-Micali-Wigderson (GMW) protocol for secure comparison -- addressing Yao's millionaires' problem -- using preprocessed Beaver's bit triples generated from Silent Random Oblivious Transfer. Our novel protocol achieves an online speedup of 17$\times$ in nonlinear operations compared to state-of-the-art solutions while reducing communication overhead. To further enhance performance, SecONNds employs Number Theoretic Transform (NTT) preprocessing and leverages GPU acceleration for homomorphic encryption operations, resulting in speedups of 1.6$\times$ on CPU and 2.2$\times$ on GPU for linear operations. We also present SecONNds-P, a bit-exact variant that ensures verifiable full-precision results in secure computation, matching the results of plaintext computations. Evaluated on a 37-bit quantized SqueezeNet model, SecONNds achieves an end-to-end inference time of 2.8 s on GPU and 3.6 s on CPU, with a total communication of just 420 MiB. SecONNds' efficiency and reduced computational load make it well-suited for deploying privacy-sensitive applications in resource-constrained environments. SecONNds is open source and can be accessed from: https://github.com/shashankballa/SecONNds.
Spiking neural networks (SNNs) offer an effective approach to solving constraint satisfaction problems (CSPs) by leveraging their temporal, event-driven dynamics. Moreover, neuromorphic hardware platforms provide the potential for achieving significant energy efficiency in implementing such models. Building upon these foundations, we present an enhanced, fully spiking pipeline for solving CSPs on the SpiNNaker neuromorphic hardware platform. Focusing on the use case of Sudoku puzzles, we demonstrate that the adoption of a constraint stabilization strategy, coupled with a neuron idling mechanism and a built-in validation process, enables this application to be realized through a series of additional layers of neurons capable of performing control logic operations, verifying solutions, and memorizing the network’s state. Simulations conducted in the GPU-enhanced Neuronal Networks (GeNN) environment validate the contributions of each pipeline component before deployment on SpiNNaker. This approach offers three key advantages: i) Improved success rates for solving CSPs, particularly for challenging instances from the hard class, surpassing state-of-the-art SNN-based solvers. ii) Reduced data transmission overhead by transmitting only the final activity state from SpiNNaker instead of all generated spikes. iii) Substantially decreased spike extraction time. Compared to previous work focused on the same use case, our approach achieves a significant reduction in the number of extracted spikes (54.63% to 99.98%) and extraction time (88.56% to 96.41%).
Masked-based autoregressive models have demonstrated promising image generation capability in continuous space. However, their potential for video generation remains under-explored. In this paper, we propose \textbf{VideoMAR}, a concise and efficient decoder-only autoregressive image-to-video model with continuous tokens, composing temporal frame-by-frame and spatial masked generation. We first identify temporal causality and spatial bi-directionality as the first principle of video AR models, and propose the next-frame diffusion loss for the integration of mask and video generation. Besides, the huge cost and difficulty of long sequence autoregressive modeling is a basic but crucial issue. To this end, we propose the temporal short-to-long curriculum learning and spatial progressive resolution training, and employ progressive temperature strategy at inference time to mitigate the accumulation error. Furthermore, VideoMAR replicates several unique capacities of language models to video generation. It inherently bears high efficiency due to simultaneous temporal-wise KV cache and spatial-wise parallel generation, and presents the capacity of spatial and temporal extrapolation via 3D rotary embeddings. On the VBench-I2V benchmark, VideoMAR surpasses the previous state-of-the-art (Cosmos I2V) while requiring significantly fewer parameters ($9.3\%$), training data ($0.5\%$), and GPU resources ($0.2\%$).
Vine copulas are sophisticated models for multivariate distributions and are increasingly used in machine learning. To facilitate their integration into modern ML pipelines, we introduce the vine computational graph, a DAG that abstracts the multilevel vine structure and associated computations. On this foundation, we devise new algorithms for conditional sampling, efficient sampling-order scheduling, and constructing vine structures for customized conditioning variables. We implement these ideas in torchvinecopulib, a GPU-accelerated Python library built upon PyTorch, delivering improved scalability for fitting, sampling, and density evaluation. Our experiments illustrate how gradient flowing through the vine can improve Vine Copula Autoencoders and that incorporating vines for uncertainty quantification in deep learning can outperform MC-dropout, deep ensembles, and Bayesian Neural Networks in sharpness, calibration, and runtime. By recasting vine copula models as computational graphs, our work connects classical dependence modeling with modern deep-learning toolchains and facilitates the integration of state-of-the-art copula methods in modern machine learning pipelines.
The rise of GPU-based high-performance computing (HPC) has driven the widespread adoption of parallel programming models such as CUDA. Yet, the inherent complexity of parallel programming creates a demand for the automated sequential-to-parallel approaches. However, data scarcity poses a significant challenge for machine learning-based sequential-to-parallel code translation. Although recent back-translation methods show promise, they still fail to ensure functional equivalence in the translated code. In this paper, we propose a novel Mutual-Supervised Learning (MSL) framework for sequential-to-parallel code translation to address the functional equivalence issue. MSL consists of two models, a Translator and a Tester. Through an iterative loop consisting of Co-verify and Co-evolve steps, the Translator and the Tester mutually generate data for each other and improve collectively. The Tester generates unit tests to verify and filter functionally equivalent translated code, thereby evolving the Translator, while the Translator generates translated code as augmented input to evolve the Tester. Experimental results demonstrate that MuSL significantly enhances the performance of the base model: when applied to Qwen2.5-Coder, it not only improves Pass@1 by up to 28.91% and boosts Tester performance by 68.90%, but also outperforms the previous state-of-the-art method CodeRosetta by 1.56 and 6.92 in BLEU and CodeBLEU scores, while achieving performance comparable to DeepSeek-R1 and GPT-4.1. Our code is available at https://github.com/kcxain/musl.
Fuzzy clustering, which allows an article to belong to multiple clusters with soft membership degrees, plays a vital role in analyzing publication data. This problem can be formulated as a constrained optimization model, where the goal is to minimize the discrepancy between the similarity observed from data and the similarity derived from a predicted distribution. While this approach benefits from leveraging state-of-the-art optimization algorithms, tailoring them to work with real, massive databases like OpenAlex or Web of Science - containing about 70 million articles and a billion citations - poses significant challenges. We analyze potentials and challenges of the approach from both mathematical and computational perspectives. Among other things, second-order optimality conditions are established, providing new theoretical insights, and practical solution methods are proposed by exploiting the structure of the problem. Specifically, we accelerate the gradient projection method using GPU-based parallel computing to efficiently handle large-scale data.
The emerging memristor crossbar array based computing circuits exhibit computing speeds and energy efficiency far surpassing those of traditional digital processors. This type of circuits can complete high-dimensional matrix operations in an extremely short time through analog computing, making it naturally applicable to linear detection and maximum likelihood detection in massive multiple-input multiple-output (MIMO) systems. However, the challenge of employing memristor crossbar arrays to efficiently implement other nonlinear detection algorithms, such as the successive interference cancellation (SIC) algorithm, remains unresolved. In this paper we propose a memristor-based circuit design for massive MIMO SIC detector. The proposed circuit comprises several judiciously designed analog matrix computing modules and hybrid analog-digital slicers, which enables the proposed circuit to perform the SIC algorithm with a hybrid analog-digital computing architecture. We show that the computing speed and the computational energy-efficiency of the proposed detector circuit are 43 times faster and 110 times higher, respectively, than those of a traditional 8-core digital signal processor (DSP), and also advantageous over the benchmark high-performance field programmable gate array (FPGA) and graphics processing unit (GPU).
Diffusion models are a class of generative models that have been recently used for speech enhancement with remarkable success but are computationally expensive at inference time. Therefore, these models are impractical for processing streaming data in real-time. In this work, we adapt a sliding window diffusion framework to the speech enhancement task. Our approach progressively corrupts speech signals through time, assigning more noise to frames close to the present in a buffer. This approach outputs denoised frames with a delay proportional to the chosen buffer size, enabling a trade-off between performance and latency. Empirical results demonstrate that our method outperforms standard diffusion models and runs efficiently on a GPU, achieving an input-output latency in the order of 0.3 to 1 seconds. This marks the first practical diffusion-based solution for online speech enhancement.