Paper Citation Record · LEDGER

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing

As of 19 August 2026, this Paper Citation Record lists 28 of 28 outbound references and 0 inbound Pith citation observations for arXiv:2509.09420.

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

pith.paper-citation-record.v1
2509.09420 v1

Coverage vector

measured 28 of 28 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-04T19:16:55.968542Z

measured 28 of 28 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-19T06:32:44.657259+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

28 of 28 outbound references displayed

verified exact0
verified fuzzy0
unresolved28
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation c023fad7-e42f-4af1-91e1-f30712da59c3 · outbound

This paper cites DeepSeek-V3 Technical Report.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing DeepSeek-V3 Technical Report

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:53.978741Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc95a5cd-9a8a-4919-8671-0fd2e916564b · outbound

This paper cites Mixtral of Experts.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Mixtral of Experts

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.034317Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9223967-e2e5-404c-91f1-74b5f48ccdbc · outbound

This paper cites 1.1 computing’s energy problem (and what we can do about it),.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing 1.1 computing’s energy problem (and what we can do about it),

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.123728Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 648d718b-9cb3-415e-8a45-de80bf678ca6 · outbound

This paper cites Near- memory computing on fpgas with 3d-stacked memories: Ap- plications, architectures, and optimizations,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Near- memory computing on fpgas with 3d-stacked memories: Ap- plications, architectures, and optimizations,

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.207966Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8c005d75-46d8-4b02-8e66-8267716ce092 · outbound

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

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.346068Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 868f603e-6201-4cf9-b30d-af78ff293680 · outbound

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

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.437114Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 35deddcf-3cdb-4a0d-97fc-b175cceb5950 · outbound

This paper cites Adapmoe: Adaptive sensitivity-based expert gating and man- agement for efficient moe inference,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Adapmoe: Adaptive sensitivity-based expert gating and man- agement for efficient moe inference,

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.528051Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ad2c06d2-4f6e-4a83-b3c2-cd3417e7323b · outbound

This paper cites Task Scheduling for Efficient Inference of Large Language Models on Single Moderate GPU Systems.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Task Scheduling for Efficient Inference of Large Language Models on Single Moderate GPU Systems

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.612324Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 21df77bd-7575-4408-817d-5eea93f53ce4 · outbound

This paper cites DAOP: Data-Aware Offloading and Predictive Pre-Calculation for Efficient MoE Inference.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing DAOP: Data-Aware Offloading and Predictive Pre-Calculation for Efficient MoE Inference

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.707195Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 50eddfa0-7cf9-4198-978a-26934580ec3e · outbound

This paper cites HOBBIT: A Mixed Precision Expert Offloading System for Fast MoE Inference.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing HOBBIT: A Mixed Precision Expert Offloading System for Fast MoE Inference

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.812983Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b750e429-49c5-437c-9696-c99f7cb15091 · outbound

This paper cites Qwen2.5 Technical Report.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Qwen2.5 Technical Report

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:54.942933Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cd6e5e2e-36bb-47d3-a2b7-56376da3465f · outbound

This paper cites DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.009024Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9964d58e-d9b0-44da-8b7a-ef29cd5b4439 · outbound

This paper cites HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing HybriMoE: Hybrid CPU-GPU Scheduling and Cache Management for Efficient MoE Inference

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.064846Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 39938ced-e2cb-4a12-b64c-37d5f53e01ee · outbound

This paper cites EPS-MoE: Expert Pipeline Scheduler for Cost-Efficient MoE Inference.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing EPS-MoE: Expert Pipeline Scheduler for Cost-Efficient MoE Inference

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.123761Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 13b6405c-ff34-49f8-811e-c463f0119c21 · outbound

This paper cites MoETuner: Optimized Mixture of Expert Serving with Balanced Expert Placement and Token Routing.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing MoETuner: Optimized Mixture of Expert Serving with Balanced Expert Placement and Token Routing

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.187513Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b511cd5d-fc47-4c18-b843-9628e6c05bcf · outbound

This paper cites Shortcut-connected Expert Parallelism for Accelerating Mixture-of-Experts.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Shortcut-connected Expert Parallelism for Accelerating Mixture-of-Experts

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.265118Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fdf83db3-38a7-4543-b181-a7275d6628fc · outbound

This paper cites Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity,

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.306591Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b14e6847-d7fa-40b6-a19c-e66ec4f544c2 · outbound

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

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Fastermoe: modeling and optimizing training of large-scale dynamic pre-trained models,

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.337181Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5725ad46-9614-418c-a293-1b4b20c56505 · outbound

This paper cites Netmoe: Accelerating moe training through dynamic sample placement,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Netmoe: Accelerating moe training through dynamic sample placement,

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.387250Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 23b3792a-05cb-4a57-af79-7107bfd2c085 · outbound

This paper cites Samsung pim/pnm for transfmer based ai: Energy efficiency on pim/pnm cluster,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Samsung pim/pnm for transfmer based ai: Energy efficiency on pim/pnm cluster,

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.469355Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1246a6e4-ffb2-4a65-8ab5-7feea4864155 · outbound

This paper cites A stacked embedded dram array for lpddr4/4x using hybrid bond- ing 3d integration with 34gb/s/1gb 0.88 pj/b logic-to-memory interface,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing A stacked embedded dram array for lpddr4/4x using hybrid bond- ing 3d integration with 34gb/s/1gb 0.88 pj/b logic-to-memory interface,

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.511861Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ada569a1-975f-409f-b4ec-24430e250a57 · outbound

This paper cites 184qps/w 64mb/mm 2 3d logic-to-dram hybrid bonding with process-near-memory engine for recommendation system,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing 184qps/w 64mb/mm 2 3d logic-to-dram hybrid bonding with process-near-memory engine for recommendation system,

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.584274Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7a647bae-3091-4ebf-b727-a4aa758f498f · outbound

This paper cites Exploiting similarity opportunities of emerging vision ai models on hybrid bonding architecture,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Exploiting similarity opportunities of emerging vision ai models on hybrid bonding architecture,

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.636411Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 625e7d78-ba1c-46a1-93b1-226d046f6478 · outbound

This paper cites H2-llm: Hardware-dataflow co- exploration for heterogeneous hybrid-bonding-based low-batch llm inference,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing H2-llm: Hardware-dataflow co- exploration for heterogeneous hybrid-bonding-based low-batch llm inference,

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.703749Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c2fdafc8-2394-4fb8-9d3f-6f5e16e651b8 · outbound

This paper cites Qwen2 Technical Report.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Qwen2 Technical Report

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.758949Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6040e835-9af4-4422-a13c-ecb910596444 · outbound

This paper cites Collec- tive communication on architectures that support simultaneous communication over multiple links,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Collec- tive communication on architectures that support simultaneous communication over multiple links,

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.802746Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 518b3eac-fd3e-4f57-8556-34896c311998 · outbound

This paper cites Astra- sim: Enabling sw/hw co-design exploration for distributed dl training platforms,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Astra- sim: Enabling sw/hw co-design exploration for distributed dl training platforms,

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.913639Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7b94c5cc-7628-465d-be71-6e634afd80d3 · outbound

This paper cites Judging llm-as- a-judge with mt-bench and chatbot arena,.

HD-MoE: Hybrid and Dynamic Parallelism for Mixture-of-Expert LLMs with 3D Near-Memory Processing Judging llm-as- a-judge with mt-bench and chatbot arena,

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-04T19:16:55.968542Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.