Paper Citation Record · LEDGER

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines

As of 17 August 2026, this Paper Citation Record lists 39 of 39 outbound references and 1 inbound Pith citation observation for arXiv:2412.14335.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2412.14335 v2

Coverage vector

measured 39 of 39 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-11T12:23:19.004428Z

measured 40 of 40 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-05-18T00:27:43.845408Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: arxiv_reference, observed 2026-05-18T00:30:33.076250Z

Reference resolution

39 of 39 outbound references displayed

verified exact1
verified fuzzy27
unresolved11
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 890877d4-5f77-49ac-b521-e89d0ff47993 · outbound

This paper cites A bridging model for parallel computation,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines A bridging model for parallel computation,

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.468135Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d1a0b163-db15-42ec-8fb3-c4784c7ce237 · outbound

This paper cites Pytorch fsdp: Experiences on scaling fully sharded data parallel,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Pytorch fsdp: Experiences on scaling fully sharded data parallel,

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.500285Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3c7b4e4f-8200-4234-ad70-3c5a017746e5 · outbound

This paper cites NanoFlow: Towards Optimal Large Language Model Serving Throughput.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines NanoFlow: Towards Optimal Large Language Model Serving Throughput

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.596808Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c32deaf5-bd94-4184-bdab-1ec72303a5b3 · outbound

This paper cites The Llama 3 Herd of Models.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines The Llama 3 Herd of Models

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.637910Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1ae43603-382f-4fc8-968c-5d3e08f8b168 · outbound

This paper cites {ARK}:{GPU-driven} code execution for distributed deep learning,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines {ARK}:{GPU-driven} code execution for distributed deep learning,

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.799105Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation e6ebe5e3-331a-4d90-98ad-863da059bdbe · outbound

This paper cites 11.1 amd instincttm mi300 series modular chiplet package – hpc and ai accelerator for exa-class systems,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines 11.1 amd instincttm mi300 series modular chiplet package – hpc and ai accelerator for exa-class systems,

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.783250Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 6a1f410e-90f4-4d51-b122-aa53bde69838 · outbound

This paper cites AMD Instinct™MI300X Accelerator: Packaging and Architecture Co-Optimization,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines AMD Instinct™MI300X Accelerator: Packaging and Architecture Co-Optimization,

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.772487Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 1dd6be83-e635-4c87-9d31-354fad14baf0 · outbound

This paper cites The AMD CDNA™ 3 architecture,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines The AMD CDNA™ 3 architecture,

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.761505Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation c75bc5a0-a566-4c90-846e-9ad51407dc64 · outbound

This paper cites HSA Runtime API and runtime for ROCm,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines HSA Runtime API and runtime for ROCm,

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.750820Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation ba5d4042-d53e-4154-9a86-8291b703c223 · outbound

This paper cites HIP: C++ Heterogeneous-Compute Interface for Portability,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines HIP: C++ Heterogeneous-Compute Interface for Portability,

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.739086Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 243312cf-886f-423c-86bb-63307a79a35f · outbound

This paper cites Tale of two cs: Computation vs. communication scaling for future transformers on future hardware,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Tale of two cs: Computation vs. communication scaling for future transformers on future hardware,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.727367Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 35802812-fcae-49fc-96f9-f86b8d45709e · outbound

This paper cites AMD ROCm™ Software,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines AMD ROCm™ Software,

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.716251Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation f858a095-6ea6-4c4f-8ac3-3a7610b2952d · outbound

This paper cites ROCm/rocBLAS: Next generation BLAS implementation for ROCm platform,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines ROCm/rocBLAS: Next generation BLAS implementation for ROCm platform,

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.703069Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 66cd9b53-7e39-4089-b23f-8874fe049d04 · outbound

This paper cites ROCm Communication Collectives Library (RCCL).

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines ROCm Communication Collectives Library (RCCL)

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.690679Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 6927e377-6619-403c-9cde-3c0acd2c1728 · outbound

This paper cites ROCm: HIPStream,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines ROCm: HIPStream,

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.679660Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 5cdb2bd9-e3db-4143-9029-4dceaab0961b · outbound

This paper cites rocprof — ROC Profiler Documentation,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines rocprof — ROC Profiler Documentation,

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.668818Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 2c930b0c-a81c-455d-aa56-f678e87621f8 · outbound

This paper cites AMD Instinct™ MI300X Accelerator Perfor- mance Validation Guide,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines AMD Instinct™ MI300X Accelerator Perfor- mance Validation Guide,

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.658083Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation a8f051b4-d58a-4f55-9839-7cdb3ccaf05a · outbound

This paper cites ROCm: ROCR-Runtime,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines ROCm: ROCR-Runtime,

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.647582Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 19addbb6-178a-457f-b8f0-8dbceae946c7 · outbound

This paper cites T3: Transparent tracking & triggering for fine-grained overlap of compute & collectives,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines T3: Transparent tracking & triggering for fine-grained overlap of compute & collectives,

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.637138Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation f1ba0c67-19ca-41da-a15f-04e42cd467ac · outbound

This paper cites DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.741039Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 918f3ca6-b710-4094-97a6-603550ca0fb0 · outbound

This paper cites TimeGraph: GPU Scheduling for Real-Time Multi-Tasking Environments,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines TimeGraph: GPU Scheduling for Real-Time Multi-Tasking Environments,

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.626380Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 79aa09f4-c0dc-4557-bc57-6e0ea243859b · outbound

This paper cites MSC- CLang: Microsoft Collective Communication Language,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines MSC- CLang: Microsoft Collective Communication Language,

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.615472Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation b6541107-eac4-40d7-9ec4-7267ff685389 · outbound

This paper cites Available: https://github.com/NVIDIA/nccl.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Available: https://github.com/NVIDIA/nccl

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.604358Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation d7372df3-5cb9-4ae8-828a-430e2ab53fb6 · outbound

This paper cites TACCL: Guiding Collective Algorithm Synthesis using Communication Sketches,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines TACCL: Guiding Collective Algorithm Synthesis using Communication Sketches,

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.593400Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 5c965cdf-de6a-4e2e-a15c-201a20225f60 · outbound

This paper cites MSCCL++: A GPU-driven communication stack for scalable AI applications.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines MSCCL++: A GPU-driven communication stack for scalable AI applications

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.581937Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 33a9f01b-36cd-481c-abf5-2b04b1c5de54 · outbound

This paper cites Deep Learning Recommendation Model for Personalization and Recommendation Systems.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Deep Learning Recommendation Model for Personalization and Recommendation Systems

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.838572Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c3c9734d-8040-4f14-ac61-bcea09cbcb98 · outbound

This paper cites Megatron-LM: Training Multi-Billion Parameter Language Mod- els Using Model Parallelism,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Megatron-LM: Training Multi-Billion Parameter Language Mod- els Using Model Parallelism,

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.570814Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 93de37ab-023d-4849-9713-109d1645eaee · outbound

This paper cites Enabling Compute-Communication Overlap in Distributed Deep Learning Training Platforms,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Enabling Compute-Communication Overlap in Distributed Deep Learning Training Platforms,

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.917412Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation febf812d-6ccd-40b8-b15f-3c2bfb32a186 · outbound

This paper cites The Case for GPGPU Spatial Multitasking,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines The Case for GPGPU Spatial Multitasking,

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.559159Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation f1ed9c9e-0e3f-42c4-9330-7a15f54fc25a · outbound

This paper cites Improving GPGPU energy-efficiency through concurrent kernel execution and DVFS,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Improving GPGPU energy-efficiency through concurrent kernel execution and DVFS,

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.546408Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 0de2b38d-dcb5-4768-81ea-ba7fe669f6cb · outbound

This paper cites Improving GPGPU Concurrency with Elastic Kernels,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Improving GPGPU Concurrency with Elastic Kernels,

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.533927Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 6ba035e1-6067-4d66-82c6-0c07a7857b84 · outbound

This paper cites Rammer: Enabling Holistic Deep Learning Compiler Optimizations with rTasks,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Rammer: Enabling Holistic Deep Learning Compiler Optimizations with rTasks,

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.521377Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation a3a604e8-475b-4bff-abd0-253409ab8303 · outbound

This paper cites Global Optimizations & Lightweight Dynamic Logic for Concurrency.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Global Optimizations & Lightweight Dynamic Logic for Concurrency

Reference 33

Resolution: verified exact; local_arxiv, observed 2026-08-11T12:23:19.250779Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation fb53f232-3849-4ca5-ae25-0ac468d29a7e · outbound

This paper cites An In-Network Architecture for Accelerating Shared-Memory Multiprocessor Collec- tives,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines An In-Network Architecture for Accelerating Shared-Memory Multiprocessor Collec- tives,

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.508949Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 1925a9da-25db-43c4-948d-e6befcaffe78 · outbound

This paper cites Introducing Async Tensor Parallelism in PyTorch,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Introducing Async Tensor Parallelism in PyTorch,

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T12:23:19.496119Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 0a16a08d-0815-4c5e-98c6-19f098eae38b · outbound

This paper cites Optimizing Distributed ML Communication with Fused Computation-Collective Operations.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Optimizing Distributed ML Communication with Fused Computation-Collective Operations

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.976636Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 25d52610-b9b8-4db3-917c-45ac4402b6f3 · outbound

This paper cites Breaking the Computation and Communication Abstraction Barrier in Distributed Machine Learning Workloads,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Breaking the Computation and Communication Abstraction Barrier in Distributed Machine Learning Workloads,

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.988507Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 54b67e22-4c11-4532-a4be-47ce55e0e45e · outbound

This paper cites Overlap Communication with Dependent Computation via Decomposition in Large Deep Learning Models,.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines Overlap Communication with Dependent Computation via Decomposition in Large Deep Learning Models,

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:19.004428Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation babe03a7-517b-4b39-9cf5-f1d3e8580eb1 · outbound

This paper cites PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel.

Optimizing ML Concurrent Computation and Communication with GPU DMA Engines PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-11T12:23:18.559962Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation a6d06ad2-aa27-4edc-8a66-4da6323f8d25 · inbound

DMA-Latte: Expanding the Reach of DMA Offloads to Latency-bound ML Communication cites this paper.

DMA-Latte: Expanding the Reach of DMA Offloads to Latency-bound ML Communication Optimizing ML Concurrent Computation and Communication with GPU DMA Engines

Reference 8

Resolution: verified exact; arxiv_id, observed 2026-05-18T00:30:33.078708Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.