RoundPipe achieves near-zero-bubble pipeline parallelism for LLM training on consumer GPUs by dynamically dispatching computation stages round-robin, yielding 1.48-2.16x speedups and enabling 235B model fine-tuning on 8x RTX 4090.
super hub Mixed citations
PyTorch: An Imperative Style, High-Performance Deep Learning Library
Mixed citation behavior. Most common role is background (53%).
158
Pith papers citing it
Background
53% of classified citations
abstract
Deep learning frameworks have often focused on either usability or speed, but not both. PyTorch is a machine learning library that shows that these two goals are in fact compatible: it provides an imperative and Pythonic programming style that supports code as a model, makes debugging easy and is consistent with other popular scientific computing libraries, while remaining efficient and supporting hardware accelerators such as GPUs. In this paper, we detail the principles that drove the implementation of PyTorch and how they are reflected in its architecture. We emphasize that every aspect of PyTorch is a regular Python program under the full control of its user. We also explain how the careful and pragmatic implementation of the key components of its runtime enables them to work together to achieve compelling performance. We demonstrate the efficiency of individual subsystems, as well as the overall speed of PyTorch on several common benchmarks.
hub tools
citation-role summary
background 18
method 11
dataset 1
citation-polarity summary
claims ledger
- abstract Deep learning frameworks have often focused on either usability or speed, but not both. PyTorch is a machine learning library that shows that these two goals are in fact compatible: it provides an imperative and Pythonic programming style that supports code as a model, makes debugging easy and is consistent with other popular scientific computing libraries, while remaining efficient and supporting hardware accelerators such as GPUs. In this paper, we detail the principles that drove the implementation of PyTorch and how they are reflected in its architecture. We emphasize that every aspect o
co-cited works
representative citing papers
Looped transformers with recall and outer normalization produce reachable, input-dependent fixed points with stable gradients, enabling generalization, while those without recall cannot; a new internal recall variant performs competitively or better.
First in-orbit demonstration of a DRL-trained AI satellite attitude controller that performs robust inertial pointing after sim-to-real transfer.
A two-pass optimization framework with polynomial-based simulation discovers heralded ballistic circuits for 3-5 qubit graph states achieving up to 7.5x higher success probabilities than fusion baselines, including first known circuits for some 5-qubit states.
Task vectors from weight differences allow arithmetic operations to edit pre-trained models, improving multiple tasks simultaneously and enabling analogical inference on unseen tasks.
Quantitative Bayesian inference using a deep-learning emulator detects 0.018-0.020 M_sun of helium in the Type Ic supernova 2014L.
ffortissimo is a JAX-based freeform forward-modeling pipeline that fits complex dust distributions and infers scattering properties in KLIP-reduced images of circumstellar disks such as HR 4796A.
A matrix-free, GPU-compatible PyTorch implementation of phase-field fracture with explicit dynamics, custom differentiable implicit damage solve, benchmarks on dynamic and quasi-static cases, and inverse recovery of fracture energy G_c via L-BFGS.
A Set-Transformer architecture with self-attention encodes Pauli-string correlations, optimizes via commutation objective, and finds symmetries with near-deterministic success on physical models like Ising and Toric code.
A new lattice method recasts SIGW integrals as FFT convolutions to compute fully non-Gaussian spectra in seconds with ~10% error on a radiation-dominated background.
Hybrid TimesFM plus ridge regression on covariates forecasts 1-MeV electron flux with average R² of 0.9 on out-of-sample 2024 data, outperforming linear regression, CNN, LSTM and Transformer models.
A neural network trained on simulations infers stripping times for Sagittarius stream stars from phase-space data, measuring a 0.3 dex/Gyr metallicity gradient and estimating ages for globular clusters such as Pal 12 and NGC 2419.
Events trigger on-the-fly LoRA module generation via hypernetworks over a shared team policy in MARL, paired with a Neural Manifold Diversity metric, enabling sequential role reassignment while preserving reward maximization.
Vibrational mode graphs from molecular dynamics enable sequence-free protein function prediction via graph neural networks, with entrainment improving signals for collective dynamics.
Dingo-Pop uses a transformer to perform amortized, end-to-end population inference from GW strain data in seconds, bypassing per-event Monte Carlo sampling.
Learning in low-rank RNNs reduces to an exact low-dimensional ODE system in overlap space, where loss-invisible overlaps encode training history without affecting function.
Dynamical magnetotropic susceptibility k(ω) acts as a probe of uniform spin and charge fluctuations, with its static scaling in α-RuCl3 arising specifically from dominant Kitaev interactions in the models examined.
Transformer networks sample up to 180x180 2D Ising systems and 64x64 Edwards-Anderson systems by generating spin groups with probability approximations, yielding ~20x higher effective sample size than prior neural samplers at criticality.
A single end-to-end Transformer model unifies stellar labels from heterogeneous spectroscopic surveys into a self-consistent scale without post-hoc recalibration.
Parametric neural networks learn likelihood ratios to infer top-philic scalar resonances from dip patterns caused by signal-background interference in hadron collider data.
A graph-conditioned meta-optimizer learns QAOA parameter trajectories from one problem class and transfers them to others, yielding better initializations than standard methods in an empirical study of 64 settings.
SMC forgets its initial condition geometrically in the jump chain and as 1/ℓ in continuous genetic distance, justifying independent-locus approximations.
Concept Graph Convolutions perform message passing on node concepts to increase interpretability of graph neural networks without losing task performance.
A neural network learns non-stationary anisotropic correlations from gridded CTM outputs and transfers the structure via LatticeKrig basis functions to station data for refined fine-scale NO2 predictions with uncertainty.
citing papers explorer
-
Stability and Generalization in Looped Transformers
Looped transformers with recall and outer normalization produce reachable, input-dependent fixed points with stable gradients, enabling generalization, while those without recall cannot; a new internal recall variant performs competitively or better.
-
Editing Models with Task Arithmetic
Task vectors from weight differences allow arithmetic operations to edit pre-trained models, improving multiple tasks simultaneously and enabling analogical inference on unseen tasks.
-
Learning reveals invisible structure in low-rank RNNs
Learning in low-rank RNNs reduces to an exact low-dimensional ODE system in overlap space, where loss-invisible overlaps encode training history without affecting function.
-
Concept Graph Convolutions: Message Passing in the Concept Space
Concept Graph Convolutions perform message passing on node concepts to increase interpretability of graph neural networks without losing task performance.
-
RLGT: A reinforcement learning framework for extremal graph theory
RLGT is a modular reinforcement learning framework for extremal graph theory that handles undirected, directed, looped, and multi-colored graphs to facilitate future research.
-
KernelBench: Can LLMs Write Efficient GPU Kernels?
KernelBench shows that even the best current LLMs generate correct and faster-than-baseline GPU kernels in fewer than 20 percent of realistic ML workloads.
-
Diffusion Models Beat GANs on Image Synthesis
Diffusion models with architecture improvements and classifier guidance achieve superior FID scores to GANs on unconditional and conditional ImageNet image synthesis.
-
When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning
Position-Weighted On-Policy Self-Distillation (PW-OPSD) weights later tokens more heavily after a diagnostic shows position predicts teacher reliability better than entropy, yielding +1.0 and +1.1 Avg@12 gains on AIME 2024/2025.
-
POETS: Uncertainty-Aware LLM Optimization via Compute-Efficient Policy Ensembles
POETS uses compute-efficient LLM policy ensembles to implicitly perform KL-regularized Thompson sampling, delivering O(sqrt(T gamma_T)) regret bounds and state-of-the-art sample efficiency in scientific discovery tasks such as protein search and quantum circuit design.
-
What If We Let Forecasting Forget? A Sparse Bottleneck for Cross-Variable Dependencies
MS-FLOW uses a capacity-limited sparse routing mechanism to model only critical inter-variable dependencies in time series data, achieving state-of-the-art accuracy on 12 benchmarks with fewer but more reliable connections.
-
Why Does Agentic Safety Fail to Generalize Across Tasks?
Agentic safety fails to generalize across tasks because the task-to-safe-controller mapping has a higher Lipschitz constant than the task-to-controller mapping alone, as proven in linear-quadratic control and demonstrated in quadcopter and LLM experiments.
-
TCL: Enabling Fast and Efficient Cross-Hardware Tensor Program Optimization via Continual Learning
TCL delivers 16.8x faster tuning on CPU and 12.48x on GPU with modestly lower inference latency by combining RDU active sampling, a lightweight Mamba cost model, and cross-platform continual knowledge distillation.
-
Scalable Variational Bayesian Fine-Tuning of LLMs via Orthogonalized Low-Rank Adapters
PoLAR-VBLL combines orthogonalized low-rank adapters with variational Bayesian last-layer inference to enable scalable, well-calibrated uncertainty quantification in fine-tuned LLMs.
-
Mitigating Data Scarcity in Spaceflight Applications for Offline Reinforcement Learning Using Physics-Informed Deep Generative Models
MI-VAE generates physics-constrained synthetic trajectories from scarce real data to improve offline RL policy performance on planetary lander tasks over standard VAEs.
-
The hidden risks of temporal resampling in clinical reinforcement learning
Resampling clinical time series into uniform bins for offline RL reduces performance by up to 60% and causes retrospective evaluations to overestimate returns by 1.5-3x versus unprocessed data.
-
Learning from Historical Activations in Graph Neural Networks
HISTOGRAPH applies unified layer-wise attention followed by node-wise attention over historical GNN activations to improve graph classification, especially in deep models.
-
Understanding the Staged Dynamics of Transformers in Learning Latent Structure
Transformers learn latent structure components in discrete stages during training, composing rules more robustly than decomposing complex examples, with identified layer plasticity windows.
-
Thermodynamically consistent machine learning model for excess Gibbs energy
HANNA is a thermodynamically consistent ML model for predicting excess Gibbs energy from molecular structures, trained on various binary mixture data and extended to multi-component mixtures using geometric projection.
-
Scalable Equilibrium Propagation via Intermediate Error Signals for Deep Convolutional CRNNs
Introduces layer-wise learning signals combining knowledge distillation and local errors into Equilibrium Propagation, enabling scalable training of deep VGG-style CRNNs with SOTA results on CIFAR-10 and CIFAR-100.
-
Unsupervised risk factor identification across cancer types and data modalities via explainable artificial intelligence
New unsupervised method adapts the multivariate logrank statistic into a differentiable loss for training any neural network on any data modality to discover prognostically distinct patient clusters, demonstrated on myeloma lab data and lung cancer CT images with post-hoc explainability.
-
SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics
SmolVLA is a small efficient VLA model that achieves performance comparable to 10x larger models while training on one GPU and deploying on consumer hardware via community data and chunked asynchronous action prediction.
-
MONAI: An open-source framework for deep learning in healthcare
MONAI is a community-supported PyTorch framework that extends deep learning to medical data with domain-specific architectures, transforms, and deployment tools.
-
Continual Learning for Sequential Personalization of Small Language Models: A Stability Monitoring Analysis
Checkpoint monitoring during sequential LoRA adaptation of SLMs reveals instability patterns via reference set diagnostics that standard task metrics can miss.
-
Reducing Experimental Testing in Space Propulsion Film Cooling Analyses by Pixelwise Generative Image Interpolation
A feed-forward neural network with positional encoding generates film cooling images from 30% fewer experimental measurements while achieving RMSE below 8% and SSIM above 93%.
-
Physics-Informed Graph Neural Network Surrogates for Turbulent Nanoparticle Dispersion in Dental Clinical Environments
ELGIN is a graph-based physics-informed surrogate model that predicts carrier flow and polydisperse particle motion in dental aerosol scenarios, achieving lower tracking errors and 37x speedup versus full OpenFOAM CFD in a preliminary single-case test.
-
ERPPO: Entropy Regularization-based Proximal Policy Optimization
ERPPO adds a DSA-based ambiguity estimator to MAPPO and switches between L1 and L2 entropy regularization to improve exploration and stability in non-stationary multi-dimensional observations.
-
AdaMeZO: Adam-style Zeroth-Order Optimizer for LLM Fine-tuning Without Maintaining the Moments
AdaMeZO adapts Adam moment estimates to zeroth-order LLM fine-tuning without extra memory storage, outperforming MeZO with up to 70% fewer forward passes.
-
Toeplitz MLP Mixers are Low Complexity, Information-Rich Sequence Models
Toeplitz MLP Mixers replace attention with masked Toeplitz multiplications for sub-quadratic complexity while retaining more sequence information and outperforming on copying and in-context tasks.
-
Revisiting Neural Activation Coverage for Uncertainty Estimation
Neural activation coverage can be adapted to provide uncertainty estimates in regression that the authors' experiments show are more meaningful than Monte-Carlo Dropout.
-
TabEmb: Joint Semantic-Structure Embedding for Table Annotation
TabEmb decouples LLM-based semantic column embeddings from graph-based structural modeling to produce joint representations that improve table annotation tasks.
-
PRiMeFlow: Capturing Complex Expression Heterogeneity in Perturbation Response Modelling
PRiMeFlow applies flow matching in gene expression space with a U-Net velocity field and pretraining-finetuning to model perturbation-induced heterogeneity, showing strong benchmark performance on PerturBench and the ARC Virtual Cell Challenge.
-
General Explicit Network (GEN): A novel deep learning architecture for solving partial differential equations
GEN is a neural network that solves PDEs by constructing explicit function approximations from basis functions based on prior PDE knowledge, yielding more robust and extensible solutions than standard PINNs.
-
Efficient compression of neural networks and datasets
Refined probabilistic and smooth l0 pruning techniques approximate minimum description length for neural networks, achieving high compression with minimal accuracy loss and empirically verifying better sample efficiency and generalization on image and text tasks.
-
Learning to model pediatric asthma exacerbation from multiple risk factors: a case study in coastal Virginia
A case study develops a sparse dictionary learning approach to model pediatric asthma exacerbations from multiple risk factors and reports consensus on relative risks across statistical and machine learning models.
-
Identifying Gems from Roman RAPIDly
Machine learning models RuBR_comb, RuBR_loc, and RuBR_DA for real-bogus classification of transients using combined simulated data and domain adaptation for the Roman RAPID pipeline.
-
Libra: Efficient Resource Management for Agentic RL Post-Training
Libra optimizes GPU allocation across rollout and training in agentic RL via an elastic hybrid pool and C-MLFQ scheduler based on tool-return causal signals, claiming up to 3.0x throughput and 2.5x faster reward convergence on 48 A800 GPUs.
-
QuChaTeR: A Hybrid Quantum-Chaotic Temporal Framework for Earthquake Prediction
QuChaTeR hybridizes chaotic maps and variational quantum circuits with recurrent networks and wavelets to achieve faster convergence and better performance than classical and quantum-inspired baselines on real seismic datasets.
-
Dynamics-Encoded Deep Learning for Robust System Identification and Parameter Estimation
Dynamics-encoded deep learning approaches are developed for system identification and parameter estimation in dynamical systems using numerical discretization schemes.
- Block-Based Double Decoders
- LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels