Paper Citation Record · LEDGER

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling

As of 18 August 2026, this Paper Citation Record lists 31 of 31 outbound references and 0 inbound Pith citation observations for arXiv:2608.07974.

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

pith.paper-citation-record.v1
2608.07974 v1

Coverage vector

measured 31 of 31 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-12T00:46:33.917982Z

measured 31 of 31 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-18T06:34:40.430872+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

31 of 31 outbound references displayed

verified exact1
verified fuzzy25
unresolved4
parse uncertain0
malformed identifier0
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 5a7863c9-3613-4aae-b2ad-00b46b11bbab · outbound

This paper cites End-to-end test-time training for long context,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling End-to-end test-time training for long context,

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-12T00:46:33.770748Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f679fd1-231b-4403-bd6d-15d17782c553 · outbound

This paper cites Cross-subject eeg signals-based emotion recognition using contrastive learning,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Cross-subject eeg signals-based emotion recognition using contrastive learning,

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.672962Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 8fb9bf48-7fd4-4cf9-8916-23b06228ecdc · outbound

This paper cites GPipe: Efficient training of giant neural networks using pipeline parallelism,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling GPipe: Efficient training of giant neural networks using pipeline parallelism,

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.658561Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 358fe6c8-a330-4435-928d-914abd502bea · outbound

This paper cites PipeDream: Generalized pipeline parallelism for DNN training,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling PipeDream: Generalized pipeline parallelism for DNN training,

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.643962Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation b1c1270b-d892-4ef8-8202-3b10ffda93c8 · outbound

This paper cites Confidant: Customizing transformer-based llms via collaborative training on mobile devices,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Confidant: Customizing transformer-based llms via collaborative training on mobile devices,

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.629303Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation c210203a-2b00-4f29-a1c5-7fb39533454a · outbound

This paper cites Memory-efficient pipeline-parallel DNN training,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Memory-efficient pipeline-parallel DNN training,

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.614684Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation f8e54f4a-740f-4742-8eb3-debc8a7ab215 · outbound

This paper cites DAPPLE: A pipelined data parallel approach for training large models,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling DAPPLE: A pipelined data parallel approach for training large models,

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.600541Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 55562b2e-781e-48f5-9d9e-7e25f6d21962 · outbound

This paper cites Efficient large-scale language model training on GPU clusters using Megatron-LM,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Efficient large-scale language model training on GPU clusters using Megatron-LM,

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.585318Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 5c0641d4-7a4d-424f-b6ab-d455d7a8d7a6 · outbound

This paper cites Chimera: Efficiently training large-scale neural networks with bidirectional pipelines,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Chimera: Efficiently training large-scale neural networks with bidirectional pipelines,

Reference 9

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.570046Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 0803ad90-64a3-41a2-b0b7-4cfff52ec579 · outbound

This paper cites Zero bubble (almost) pipeline parallelism,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Zero bubble (almost) pipeline parallelism,

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.554267Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation f7160761-c329-4055-b4d7-e815eac16991 · outbound

This paper cites Universal checkpointing: A flexible and efficient distributed checkpointing system for large-scale DNN training with reconfigurable parallelism,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Universal checkpointing: A flexible and efficient distributed checkpointing system for large-scale DNN training with reconfigurable parallelism,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.536594Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 137e6c87-ab0f-4c1c-a8fa-ca2f164aecbf · outbound

This paper cites Sparse checkpointing for fast and reliable MoE training,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Sparse checkpointing for fast and reliable MoE training,

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.522521Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation a12ebc72-7b7b-442b-9701-0ca6066ea048 · outbound

This paper cites Attack of the bubbles: Straggler-resilient pipeline parallelism for large model training,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Attack of the bubbles: Straggler-resilient pipeline parallelism for large model training,

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.508234Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation b4806688-c2e1-4e6b-99bf-d130af39f4e9 · outbound

This paper cites CollaPipe: Adaptive segment-optimized pipeline parallelism for collabo- rative LLM training in heterogeneous edge networks,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling CollaPipe: Adaptive segment-optimized pipeline parallelism for collabo- rative LLM training in heterogeneous edge networks,

Reference 14

Resolution: verified exact; raw_fallback, observed 2026-08-12T00:46:34.195442Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 66e15773-bc30-4426-85e4-2c440a04b878 · outbound

This paper cites SWARM parallelism: Training large models can be surprisingly communication- efficient,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling SWARM parallelism: Training large models can be surprisingly communication- efficient,

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.493719Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation dfa104de-ecbd-4799-9b99-5b76c6eba5c1 · outbound

This paper cites Petals: Collaborative infer- ence and fine-tuning of large models,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Petals: Collaborative infer- ence and fine-tuning of large models,

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.479123Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation dd5b1f40-690a-41e9-a12a-892798b2133b · outbound

This paper cites Decoupled parallel backpropagation with convergence guarantee,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Decoupled parallel backpropagation with convergence guarantee,

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.463654Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 0689e8ca-d738-45b1-adb3-981ce3602829 · outbound

This paper cites Depth-progressive monotonic learning without global backpropagation,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Depth-progressive monotonic learning without global backpropagation,

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.447949Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 6930df10-4428-4666-a38f-ff7cea643f24 · outbound

This paper cites Beyond- backpropagation training: Methods, applications, and perspectives,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Beyond- backpropagation training: Methods, applications, and perspectives,

Reference 19

Resolution: metadata mismatch; raw_fallback, observed 2026-08-12T00:46:34.099602Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation c593da75-0bfd-4d38-b9e6-3b3263e8a51d · outbound

This paper cites Direct feedback alignment provides learning in deep neural networks,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Direct feedback alignment provides learning in deep neural networks,

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.431998Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 454a38ba-a87b-4f56-8b6b-6ddcf6290282 · outbound

This paper cites Deep learning without weight transport,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Deep learning without weight transport,

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.416765Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation a431d332-d24a-40c9-b0eb-d72772b83c4d · outbound

This paper cites Fine-tuning language models with just forward passes,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Fine-tuning language models with just forward passes,

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.401637Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 1429a096-00f3-48d7-bdf8-df9dad67dba7 · outbound

This paper cites Curvzo: Adaptive curvature-guided sparse zeroth-order optimization for efficient llm fine-tuning,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Curvzo: Adaptive curvature-guided sparse zeroth-order optimization for efficient llm fine-tuning,

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.385676Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 7b2b594b-8bfb-400e-b380-3dd39c4f580e · outbound

This paper cites Noprop: Training neural networks without back-propagation or forward-propagation,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Noprop: Training neural networks without back-propagation or forward-propagation,

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.369090Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 8d820b95-65e2-477e-a9f7-ed06e5db332b · outbound

This paper cites Predictive coding approximates backprop along arbitrary computation graphs,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Predictive coding approximates backprop along arbitrary computation graphs,

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.353685Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 740843f9-c526-482d-b412-87ce05d8d282 · outbound

This paper cites Faster Multi-GPU Training with PPLL: A Pipeline Parallelism Framework Leveraging Local Learning.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Faster Multi-GPU Training with PPLL: A Pipeline Parallelism Framework Leveraging Local Learning

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-12T00:46:33.892664Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 01934c66-198a-4fd8-a292-44df2455137b · outbound

This paper cites Idle no more: Boosting distributed pipeline training via FluidPipe,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Idle no more: Boosting distributed pipeline training via FluidPipe,

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.337937Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 88c69243-1253-4f94-9b27-3d9a7853a1d7 · outbound

This paper cites Scpl: Enhancing neural network training throughput with decoupled local losses and model parallelism,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Scpl: Enhancing neural network training throughput with decoupled local losses and model parallelism,

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.321723Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation 0f8d825f-b61a-4c73-932b-2f94f343c122 · outbound

This paper cites Rethinking Local Learning: A Cheaper and Faster Recipe for LLM Post-Training.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Rethinking Local Learning: A Cheaper and Faster Recipe for LLM Post-Training

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-12T00:46:33.907700Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98bad4f2-a3be-4f38-90ea-7f29c49922f4 · outbound

This paper cites LoRA: Low-rank adaptation of large language models,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling LoRA: Low-rank adaptation of large language models,

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T00:46:34.306538Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.

Observation fb406d83-7019-478c-80da-0cf32e67f85a · outbound

This paper cites Convergence analysis of split federated learning on heterogeneous data,.

ZeroLock: Concurrent Memory-Efficient LLM Training via Modular Update Decoupling Convergence analysis of split federated learning on heterogeneous data,

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-12T00:46:33.917982Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.