On the public citation graph

TensorSocket: Shared Data Loading for Deep Learning Training

0 signed · 42 inferred · view as graph

Inferred edges come from two machine sources: sweeps of open repository metadata (mostly SSRN and Zenodo), and a bulk OpenAlex citation snapshot. Every row below is labelled with the one it came from. Neither is a complete citation index — absence here is not evidence of absence.

Is this your work? Sign in with ORCID and put your name on the claims that are yours.

Claim with ORCID

Reference health. See whether every reference in this work resolves to a known work and none has been retracted — and get a dated CiteStamped badge you can embed.

See reference health

What this work says

Outgoing claims: the papers this work cites, builds on, or argues with.

inferredcites →https://doi.org/10.1007/978-3-030-01246-5_2openalex-snapshot

Progressive Neural Architecture Search

inferredcites →https://doi.org/10.1109/cvpr.2009.5206848openalex-snapshot

ImageNet: A large-scale hierarchical image database

inferredcites →https://doi.org/10.1109/icassp.2015.7178964openalex-snapshot

Librispeech: An ASR corpus based on public domain audio books

inferredcites →https://doi.org/10.1109/icde55515.2023.00077openalex-snapshot

The Art of Losing to Win: Using Lossy Image Compression to Improve Data Loading in Deep Learning Pipelines

inferredcites →https://doi.org/10.1109/iiswc63097.2024.00013openalex-snapshot

Lotus: Characterization of Machine Learning Preprocessing Pipelines via Framework and Hardware Profiling

inferredcites →https://doi.org/10.1109/mc.2022.3148714openalex-snapshot

The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrink

inferredcites →https://doi.org/10.1145/1066157.1066201openalex-snapshot

QPipe

inferredcites →https://doi.org/10.1145/2541940.2541941openalex-snapshot

Quasar

inferredcites →https://doi.org/10.1145/3381831openalex-snapshot

Green AI

inferredcites →https://doi.org/10.1145/3448016.3457286openalex-snapshot

Expand your Training Limits! Generating Training Data for ML-based Data Management

inferredcites →https://doi.org/10.1145/3470496.3533044openalex-snapshot

Understanding data storage and ingestion for large-scale deep recommendation model training

inferredcites →https://doi.org/10.1145/3514221.3517848openalex-snapshot

Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines

inferredcites →https://doi.org/10.1145/3531146.3533234openalex-snapshot

Measuring the Carbon Intensity of AI in Cloud Instances

inferredcites →https://doi.org/10.1145/3572751.3572765openalex-snapshot

Characterizing I/O in Machine Learning with MLPerf Storage

inferredcites →https://doi.org/10.1145/3578356.3592589openalex-snapshot

Profiling and Monitoring Deep Learning Training Tasks

inferredcites →https://doi.org/10.1145/3592980.3595314openalex-snapshot

The Difficult Balance Between Modern Hardware and Conventional CPUs

inferredcites →https://doi.org/10.1145/3595360.3595851openalex-snapshot

Data Management and Visualization for Benchmarking Deep Learning Training Systems

inferredcites →https://doi.org/10.1145/3620678.3624666openalex-snapshot

tf.data service

inferredcites →https://doi.org/10.1145/3627703.3629578openalex-snapshot

Orion: Interference-aware, Fine-grained GPU Sharing for ML Applications

inferredcites →https://doi.org/10.1145/3642970.3655827openalex-snapshot

An Analysis of Collocation on GPUs for Deep Learning Training

inferredcites →https://doi.org/10.1145/3642970.3655830openalex-snapshot

Characterizing Training Performance and Energy for Foundation Models and Image Classifiers on Multi-Instance GPUs

inferredcites →https://doi.org/10.1145/3715275.3732006openalex-snapshot

Hype, Sustainability, and the Price of the Bigger-is-Better Paradigm in AI

inferredcites →https://doi.org/10.14778/2168651.2168654openalex-snapshot

SharedDB

inferredcites →https://doi.org/10.14778/2536360.2536364openalex-snapshot

Sharing data and work across concurrent analytical queries

inferredcites →https://doi.org/10.14778/2733085.2733095openalex-snapshot

ADDICT

inferredcites →https://doi.org/10.14778/3342263.3342276openalex-snapshot

Crossbow

inferredcites →https://doi.org/10.14778/3407790.3407816openalex-snapshot

Cerebro

inferredcites →https://doi.org/10.14778/3446095.3446100openalex-snapshot

Analyzing and mitigating data stalls in DNN training

inferredcites →https://doi.org/10.14778/3476311.3476374openalex-snapshot

tf.data

inferredcites →https://doi.org/10.14778/3579075.3579083openalex-snapshot

FastFlow: Accelerating Deep Learning Model Training with Smart Offloading of Input Data Pipeline

inferredcites →https://doi.org/10.14778/3636218.3636238openalex-snapshot

FusionFlow: Accelerating Data Preprocessing for Machine Learning with CPU-GPU Cooperation

inferredcites →https://doi.org/10.18653/v1/p18-1238openalex-snapshot

Conceptual Captions: A Cleaned, Hypernymed, Image Alt-text Dataset For Automatic Image Captioning

inferredcites →https://doi.org/10.18653/v1/p19-1355openalex-snapshot

Energy and Policy Considerations for Deep Learning in NLP

inferredcites →https://doi.org/10.48550/arxiv.1802.09596openalex-snapshot

Tunability: Importance of Hyperparameters of Machine Learning Algorithms

inferredcites →https://doi.org/10.48550/arxiv.2401.08895openalex-snapshot

cedar: Optimized and Unified Machine Learning Input Data Pipelines

inferredcites →https://doi.org/10.5281/zenodo.4414861openalex-snapshot

rwightman/pytorch-image-models: 3rd Party Vision Transformer Weights

inferredcites →https://doi.org/10.5281/zenodo.5624572openalex-snapshot

Contrastive Learning of Musical Representations

What points at this work

Incoming claims: where this work appears in other papers’ reference lists and arguments.

inferred← backgroundhttps://doi.org/10.5281/zenodo.23400672zenodo-depositor-declared

inferred← citeshttps://doi.org/10.1007/978-3-032-07373-0_21openalex-snapshot

Evaluating the Impact of Parallel Data Loading on Training Performance in PyTorch

inferred← citeshttps://doi.org/10.1007/s44443-026-01137-zopenalex-snapshot

Website fingerprinting data format (WFDF): Harnessing parallelisation performance on dataset encoding and decoding operations

inferred← citeshttps://doi.org/10.1145/3746467.3801513openalex-snapshot

Identifying the Optimal Number of DataLoader Workers for CPU-GPU Concurrency in Asynchronous Deep Training Pipelines

inferred← citeshttps://doi.org/10.1145/3746467.3801522openalex-snapshot

GPU Starvation from the Control Plane: A Systems Study of Reinforcement Learning in Deep Learning Pipelines

Editorial review for independent work. ICSAC — the Institute for Complexity Science and Advanced Computing — runs a structured editorial panel for researchers publishing outside journals, and publishes on CiteStamp rails. Submission is free and goes through ORCID: icsacinstitute.org/submit.