Paper Citation Record · LEDGER

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism

As of 4 August 2026, this Paper Citation Record lists 35 of 35 outbound references and 0 inbound Pith citation observations for arXiv:2605.11005.

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

pith.paper-citation-record.v1
2605.11005 v1

Coverage vector

measured 35 of 35 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-13T01:08:45.674652Z

measured 35 of 35 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-04T06:34:03.388597+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

35 of 35 outbound references displayed

verified exact13
verified fuzzy17
unresolved3
parse uncertain1
malformed identifier0
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation aebdcfe5-452f-4924-b6c2-c0b6590d1602 · outbound

This paper cites gpt-oss-120b & gpt-oss-20b Model Card.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism gpt-oss-120b & gpt-oss-20b Model Card

Reference 1

Resolution: verified exact; local_arxiv, observed 2026-05-13T01:52:06.281680Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 7dcc398b-bd0d-4ff5-8772-068acdc300f8 · outbound

This paper cites How to scale your model.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism How to scale your model

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.298164Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation ad84c46f-19a8-4e99-afbe-af49b4e3bad8 · outbound

This paper cites Palm: Scal- ing language modeling with pathways.Journal of Machine Learning Research, 24(240):1–113.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Palm: Scal- ing language modeling with pathways.Journal of Machine Learning Research, 24(240):1–113

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.308017Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 440705cd-1d65-408d-8836-c2f16e5a10ce · outbound

This paper cites DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-05-13T01:52:06.290764Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation c6cac2c0-d909-4658-acc6-6e3f731cd78a · outbound

This paper cites Flashattention: Fast and memory- efficient exact attention with io-awareness.Advances in neural information processing systems, 35:16344– 16359.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Flashattention: Fast and memory- efficient exact attention with io-awareness.Advances in neural information processing systems, 35:16344– 16359

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.296073Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation f8fea870-9ac5-4cce-b290-d947184e47ed · outbound

This paper cites DeepSeek-V3.2: Efficient Rea- soning & Agentic AI.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism DeepSeek-V3.2: Efficient Rea- soning & Agentic AI

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.302243Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 358437bf-f23a-425e-a1fe-dcc5291b1a13 · outbound

This paper cites DeepSeek-V3 Technical Report.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism DeepSeek-V3 Technical Report

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-05-13T01:52:06.285496Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 34496c21-5e53-4b4c-8895-eb305d063f37 · outbound

This paper cites DeepSeek-V4-Pro.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism DeepSeek-V4-Pro

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.305885Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 5a95d605-3335-4f26-acd1-a9de55694189 · outbound

This paper cites an unresolved cited work.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Unresolved cited work

Reference 9

Resolution: parse uncertain; raw_fallback, observed 2026-05-13T15:37:56.300363Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 562c5fe4-d1a5-46a2-a578-6b5445df2af2 · outbound

This paper cites Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity.Journal of Machine Learning Research, 23(120):1–39.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity.Journal of Machine Learning Research, 23(120):1–39

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.304154Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 9f0d9c14-fcc4-44d8-a35c-6ae26fb3c8b0 · outbound

This paper cites Rdma over ethernet for distributed training at meta scale.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Rdma over ethernet for distributed training at meta scale

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.314748Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 24811c06-76d2-497e-b814-611c2103168f · outbound

This paper cites Gurobi Optimizer Refer- ence Manual.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Gurobi Optimizer Refer- ence Manual

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.327526Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 31c25f9a-cd6d-457d-833b-dd72254ad402 · outbound

This paper cites Faster- moe: modeling and optimizing training of large-scale dynamic pre-trained models.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Faster- moe: modeling and optimizing training of large-scale dynamic pre-trained models

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.323301Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation f2b8a846-bafa-4f14-8e93-07b7e100dc46 · outbound

This paper cites Tutel: Adaptive Mixture-of-Experts at Scale.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Tutel: Adaptive Mixture-of-Experts at Scale

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-05-13T01:52:06.277646Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 291ec63e-62cd-43b9-9d5e-69350b2893f7 · outbound

This paper cites Mixtral of Experts.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Mixtral of Experts

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-05-13T01:52:06.272279Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 469beba8-fdb0-4bfa-ab62-388712c83e9f · outbound

This paper cites Lancet: Ac- celerating mixture-of-experts training via whole graph computation-communication overlapping.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Lancet: Ac- celerating mixture-of-experts training via whole graph computation-communication overlapping

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.311364Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 64e9bf8d-1191-42e3-8bf3-5331e5fc557a · outbound

This paper cites GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding

Reference 17

Resolution: verified exact; local_arxiv, observed 2026-05-13T01:52:06.267382Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation d6d8e3c2-96c1-4d9b-9405-f224e717c459 · outbound

This paper cites Effi- cient large-scale language model training on gpu clusters using megatron-lm.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Effi- cient large-scale language model training on gpu clusters using megatron-lm

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.325404Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation a71ff0ab-06ba-490f-9062-702bbad22494 · outbound

This paper cites an unresolved cited work.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Unresolved cited work

Reference 19

Resolution: unresolved; raw_fallback, observed 2026-05-13T15:37:56.329776Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation e3b72b4b-e879-4d75-b947-8afbbf433e52 · outbound

This paper cites Gpudirect.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Gpudirect

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.319948Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 37455ec6-c554-4fda-b123-161ca2bd9e29 · outbound

This paper cites Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving.https://qwen.ai/blog?id= qwen3.6-max-preview, April 2026.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Qwen3.6-Max-Preview: Smarter, Sharper, Still Evolving.https://qwen.ai/blog?id= qwen3.6-max-preview, April 2026

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.321638Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 52c09602-f2cc-4157-85dc-24756a24e5eb · outbound

This paper cites Zero: Memory optimizations to- ward training trillion parameter models.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Zero: Memory optimizations to- ward training trillion parameter models

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.313069Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 0e3c1a85-a240-4e36-aef7-4a68a2634908 · outbound

This paper cites Deepspeed- moe: Advancing mixture-of-experts inference and training to power next-generation ai scale.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Deepspeed- moe: Advancing mixture-of-experts inference and training to power next-generation ai scale

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.333609Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 6c647e50-7c77-44a3-9442-323c1842a05e · outbound

This paper cites Deepspeed: System optimiza- tions enable training deep learning models with over 100 billion parameters.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Deepspeed: System optimiza- tions enable training deep learning models with over 100 billion parameters

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.318384Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 4c38a204-3f27-4a73-ae70-095a475b2331 · outbound

This paper cites Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer

Reference 25

Resolution: verified exact; local_arxiv, observed 2026-05-13T01:52:06.252088Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 92273d03-1919-459f-ab1b-c8582ff96caf · outbound

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

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism

Reference 26

Resolution: verified exact; local_arxiv, observed 2026-05-13T01:52:06.257115Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 5b00ccb4-b1bf-497b-8796-be7cb6477e33 · outbound

This paper cites an unresolved cited work.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Unresolved cited work

Reference 27

Resolution: unresolved; raw_fallback, observed 2026-05-13T15:37:56.316616Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 3835351e-ca87-49cd-b9d7-c69c61818247 · outbound

This paper cites Qwen3.5: Accelerating productivity with native multimodal agents, February 2026.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Qwen3.5: Accelerating productivity with native multimodal agents, February 2026

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-05-13T15:37:56.331755Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation fe4922be-c04c-4182-b474-c20a7d6b2731 · outbound

This paper cites Step-3 is large yet affordable: Model-system co-design for cost-effective decoding.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Step-3 is large yet affordable: Model-system co-design for cost-effective decoding

Reference 29

Resolution: verified exact; arxiv_id, observed 2026-05-13T01:52:06.240580Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation f3d9f39d-42f0-4ce3-8b2b-310170deda5c · outbound

This paper cites HeterMoE: Efficient Training of Mixture-of-Experts Models on Heterogeneous GPUs.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism HeterMoE: Efficient Training of Mixture-of-Experts Models on Heterogeneous GPUs

Reference 30

Resolution: verified exact; arxiv_id, observed 2026-05-13T01:52:06.246426Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 307c199e-9c62-4bb9-9766-259a6702aa66 · outbound

This paper cites Qwen3 Technical Report.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Qwen3 Technical Report

Reference 31

Resolution: verified exact; local_arxiv, observed 2026-05-13T01:52:06.262175Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 672894c3-8bfd-494b-8479-6e3d1980d680 · outbound

This paper cites an unresolved cited work.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Unresolved cited work

Reference 32

Resolution: unresolved; raw_fallback, observed 2026-05-13T15:37:56.309595Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 8d9fa804-e7ad-4d8b-a705-c9d3bf5b0ede · outbound

This paper cites Comet: Fine-grained Computation-communication Overlapping for Mixture-of-Experts.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism Comet: Fine-grained Computation-communication Overlapping for Mixture-of-Experts

Reference 33

Resolution: verified exact; arxiv_id, observed 2026-05-13T01:52:06.229386Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 1e115d64-2bec-4eb5-bf32-d2fdb2272dad · outbound

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

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel

Reference 34

Resolution: verified exact; local_arxiv, observed 2026-05-13T01:52:06.234637Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Observation 503f1b1f-4b25-41c8-a928-54e506617e8d · outbound

This paper cites MegaScale-Infer: Serving Mixture-of-Experts at Scale with Disaggregated Expert Parallelism.

DisagMoE: Computation-Communication overlapped MoE Training via Disaggregated AF-Pipe Parallelism MegaScale-Infer: Serving Mixture-of-Experts at Scale with Disaggregated Expert Parallelism

Reference 35

Resolution: metadata mismatch; arxiv_id, observed 2026-05-13T01:52:06.223867Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.

Pith citing papers

No inbound Pith citation observations are available.