ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts | Paper Citation Record

Observation bfcb7510-d51a-43c3-9196-bbc798a5fec8 · outbound

This paper cites Existing memory hierarchies fail to balance the trade-off between density, leakage, and access latency for these sparse, static weights.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Existing memory hierarchies fail to balance the trade-off between density, leakage, and access latency for these sparse, static weights

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:18.266378Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4b722ce4-fe5a-4d59-a953-1b1263c896b2 · outbound

This paper cites Consequently, the routing of tokens from their source to the assigned expert computing cores for a particular batch creates a non-deterministic traffic pattern.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Consequently, the routing of tokens from their source to the assigned expert computing cores for a particular batch creates a non-deterministic traffic pattern

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:18.324875Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dd4fbbad-7903-4d63-8fa0-8301bb73965d · outbound

This paper cites 1) imposes a synchronous gather operation.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts 1) imposes a synchronous gather operation

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:18.464815Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 39c6fb6a-bce3-458a-97bc-e25843d841a8 · outbound

This paper cites stationary.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts stationary

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:18.636671Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0b05ed8b-d4b9-4e86-adde-cb40cb7d245f · outbound

This paper cites an unresolved cited work.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Unresolved cited work

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:18.884745Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 78244fb3-0bf2-4021-9fcb-6e67ee9d36f9 · outbound

This paper cites Second, the number of links must not exceed that of a standard 2D mesh NoC.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Second, the number of links must not exceed that of a standard 2D mesh NoC

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:19.004736Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a038e01e-c53f-4d9b-9c97-29d550bbc257 · outbound

This paper cites robustly.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts robustly

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:19.105915Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 162aa294-f168-404d-acb9-e626d82d50e1 · outbound

This paper cites CHIME: Chiplet -based Heterogeneous Near -Memory Acceleration for Edge Multimodal LLM Inference,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts CHIME: Chiplet -based Heterogeneous Near -Memory Acceleration for Edge Multimodal LLM Inference,

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:20.534747Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 90716323-e213-4bff-b6d1-0aa956e71975 · outbound

This paper cites A survey of transformers,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts A survey of transformers,

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:19.322692Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5dd1275f-a549-47eb-8c04-32ed728db55a · outbound

This paper cites Mixture of Experts in Large Language Models,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Mixture of Experts in Large Language Models,

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:19.423795Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bb23b340-09ff-47a8-9fef-8676840ce54d · outbound

This paper cites A survey of architectural approaches for improving GPGPU performance, programmability and heterogeneity,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts A survey of architectural approaches for improving GPGPU performance, programmability and heterogeneity,

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:19.705032Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc6266ca-d5f4-44a5-94a9-db09ade7ab9b · outbound

This paper cites Focus Session: Hardware/Software Co -Design to Accelerate Generative AI Workloads on Heterogeneous Architectures,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Focus Session: Hardware/Software Co -Design to Accelerate Generative AI Workloads on Heterogeneous Architectures,

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:19.825218Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 460cf6e7-e59a-4830-b295-89a634f5602d · outbound

This paper cites A Survey on Memory -centric Computer Architectures,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts A Survey on Memory -centric Computer Architectures,

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:20.023012Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9b353477-79ba-4608-ad1b-587f8fe44963 · outbound

This paper cites X-Former: In-Memory Acceleration of Transformers,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts X-Former: In-Memory Acceleration of Transformers,

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:20.154748Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6ed8dd75-b7d9-4e5f-a3bb-41394626c266 · outbound

This paper cites HeTraX: Energy Efficient 3D Heterogeneous Manycore Architecture for Transformer Acceleration,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts HeTraX: Energy Efficient 3D Heterogeneous Manycore Architecture for Transformer Acceleration,

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:20.334749Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c145a185-b39c-4deb-b24e-ed8fc5edbc22 · outbound

This paper cites {DistServe}: Disaggregating prefill and decoding for goodput-optimized large language model serving,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts {DistServe}: Disaggregating prefill and decoding for goodput-optimized large language model serving,

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:21.946327Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e697229f-87a7-4dd0-ac2d-52cb9bae2534 · outbound

This paper cites H3D-Transformer: A Heterogeneous 3D (H3D) Computing Platform for Transformer Model Acceleration on Edge Devices,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts H3D-Transformer: A Heterogeneous 3D (H3D) Computing Platform for Transformer Model Acceleration on Edge Devices,

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:20.693397Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ca96cdfe-1c76-416b-8bf6-a9b568df6b13 · outbound

This paper cites Compute -in-Memory Chips for Deep Learning: Recent Trends and Prospects,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Compute -in-Memory Chips for Deep Learning: Recent Trends and Prospects,

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:20.854884Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 286ed6e2-92d8-4cf2-9ea8-7d8824cd68c9 · outbound

This paper cites Atleus: Accelerating Transformers on the Edge Enabled by 3D Heterogeneous Manycore Architectures,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Atleus: Accelerating Transformers on the Edge Enabled by 3D Heterogeneous Manycore Architectures,

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:20.961482Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a478050f-d827-423e-b18d-b31e5610d132 · outbound

This paper cites MixNet: A Runtime Reconfigurable Optical -Electrical Fabric for Distributed Mixture -of-Experts Training,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts MixNet: A Runtime Reconfigurable Optical -Electrical Fabric for Distributed Mixture -of-Experts Training,

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:20.991196Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a710d4d-b218-4bf5-b68f-68f05679743b · outbound

This paper cites Accelerating Distributed MoE Training and Inference with Lina.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Accelerating Distributed MoE Training and Inference with Lina

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:21.111262Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3d4913c5-33c4-4281-b8a2-79465c8a6012 · outbound

This paper cites A3D-MoE: Acceleration of Large Language Models with Mixture of Experts via 3D Heterogeneous Integration.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts A3D-MoE: Acceleration of Large Language Models with Mixture of Experts via 3D Heterogeneous Integration

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:21.231105Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c101cae5-b458-4bc7-b82d-70632d32cf91 · outbound

This paper cites Stratum: System -Hardware Co -Design with Tiered Monolithic 3D-Stackable DRAM for Efficient MoE Serving,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Stratum: System -Hardware Co -Design with Tiered Monolithic 3D-Stackable DRAM for Efficient MoE Serving,

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:21.371927Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cd84cf73-c326-4d08-b9a5-a29bf8c70b0f · outbound

This paper cites Superior QLC Retention (10 Years, 85°C) and Record Memory Window (12.2 V) by Gate Stack Engineering in Ferroelectric FET: from ‘MIFIS’ to.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Superior QLC Retention (10 Years, 85°C) and Record Memory Window (12.2 V) by Gate Stack Engineering in Ferroelectric FET: from ‘MIFIS’ to

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:26.033328Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9900ace6-1356-4eb1-8e94-67bfdc106e3c · outbound

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

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models,

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:25.099615Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e920f1cd-3045-468f-b17d-56d78cc20fd4 · outbound

This paper cites Newton: A DRAM-maker’s accelerator-in-memory (AiM) architecture for machine learning,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Newton: A DRAM-maker’s accelerator-in-memory (AiM) architecture for machine learning,

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:25.245562Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 07a24695-bb04-4b64-9e49-25d3a6dc0486 · outbound

This paper cites Sandisk unveils the future of memory architecture for AI introducing: High bandwidth flash.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Sandisk unveils the future of memory architecture for AI introducing: High bandwidth flash

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:25.444747Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 252ccc09-07a0-4935-b2aa-091cd3e8d1c4 · outbound

This paper cites High Bandwidth Flash: A New Memory for the Next Generation of AI Computing,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts High Bandwidth Flash: A New Memory for the Next Generation of AI Computing,

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:25.568972Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e526af7b-ecea-4b9f-bb23-532cce45eb50 · outbound

This paper cites H3 : H ybrid Architecture Using High Bandwidth Memory and High Bandwidth Flash for Cost-Efficient LLM Inference,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts H3 : H ybrid Architecture Using High Bandwidth Memory and High Bandwidth Flash for Cost-Efficient LLM Inference,

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:25.700187Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 023cc1d5-24e2-4fb8-95da-a8cdd378ec23 · outbound

This paper cites 3D Memories and Ferroelectrics,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts 3D Memories and Ferroelectrics,

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:25.803975Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ec06d9fa-9184-403c-a83b-3a86c4e91f1b · outbound

This paper cites A Vertical Channel -All-Around FeFET with Thermally Stable Oxide Semiconductor Achieving High ΔIon > 2µA/cell for 3D Stackable 4F 2 High Speed Memory,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts A Vertical Channel -All-Around FeFET with Thermally Stable Oxide Semiconductor Achieving High ΔIon > 2µA/cell for 3D Stackable 4F 2 High Speed Memory,

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:25.924739Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dde4c86c-6748-47dd-a843-7b7ccf9e2e97 · outbound

This paper cites The MMLU dataset provides a highly diverse set of multiple -choice questions spanning 57 distinct subjects (ranging from STEM to humanities).

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts The MMLU dataset provides a highly diverse set of multiple -choice questions spanning 57 distinct subjects (ranging from STEM to humanities)

Reference 32

Resolution: malformed identifier; no resolver link, observed 2026-08-04T04:18:19.200564Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 414fa6ef-e824-485e-8c16-503401c6494e · outbound

This paper cites AttAcc! Unleashing the Power of PIM for Batched Transformer-based Generative Model Inference,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts AttAcc! Unleashing the Power of PIM for Batched Transformer-based Generative Model Inference,

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:26.175455Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fefedb73-951f-4d2a-b6bb-1ce43aad398a · outbound

This paper cites Cloud TPU v6e.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Cloud TPU v6e

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:26.294764Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a3afd2fa-c40f-4ab9-8ae4-2b6980eee0cf · outbound

This paper cites In -Datacenter Performance Analysis of a Tensor Processing Unit,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts In -Datacenter Performance Analysis of a Tensor Processing Unit,

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:26.353056Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 01401004-dbff-4d3e-9923-e9b79a68e80c · outbound

This paper cites Universal Chiplet Interconnect Express (UCIe): An Open Industry Standard for Innovations With Chiplets at Package Level,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Universal Chiplet Interconnect Express (UCIe): An Open Industry Standard for Innovations With Chiplets at Package Level,

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:26.451944Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cb526c6e-5eb9-4620-a1e7-c0d1692b986b · outbound

This paper cites 3 -D NAND Technology Achievements and Future Scaling Perspectives,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts 3 -D NAND Technology Achievements and Future Scaling Perspectives,

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:26.530306Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d1edbf7a-96bd-4793-9a10-675edec38b76 · outbound

This paper cites Unleash Scaling Potential of 3D NAND with Innovative Xtacking® Architecture,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Unleash Scaling Potential of 3D NAND with Innovative Xtacking® Architecture,

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:26.636650Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3299ffef-b6cf-4745-b88f-a8710a38e42b · outbound

This paper cites Ferroelectric Field-Effect Transistor-Based 3-D NAND Architecture for Energy -Efficient on -Chip Training Accelerator,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Ferroelectric Field-Effect Transistor-Based 3-D NAND Architecture for Energy -Efficient on -Chip Training Accelerator,

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:26.937602Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 72c5b852-157f-45bd-9bda-d7f6e6113579 · outbound

This paper cites High-Endurance (>1011 cycles) and Thermally-Stable Sub-100nm TiO 2 Channel FeFET for Low -Power Memory Centric 3D -LSI Applications,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts High-Endurance (>1011 cycles) and Thermally-Stable Sub-100nm TiO 2 Channel FeFET for Low -Power Memory Centric 3D -LSI Applications,

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:27.132621Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f58fd3f-572e-4f3f-be65-9226a8b08c44 · outbound

This paper cites CMOS Directly Bonded to Array (CBA) Technology for Future 3D Flash Memory,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts CMOS Directly Bonded to Array (CBA) Technology for Future 3D Flash Memory,

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:27.290937Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fa7ed367-5e42-4c77-9fc7-8490997c7869 · outbound

This paper cites A Systematic Methodology for Characterizing Scalability of DNN Accelerators using SCALE -Sim,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts A Systematic Methodology for Characterizing Scalability of DNN Accelerators using SCALE -Sim,

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:27.472623Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b2c16ecb-94e2-4962-9ba3-57b773167fcd · outbound

This paper cites On generalized fat trees,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts On generalized fat trees,

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:27.580529Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 277e287d-06f8-47a6-b7a6-02cc3a31dcdd · outbound

This paper cites Simba: Scaling Deep -Learning Inference with Multi - Chip-Module-Based Architecture,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Simba: Scaling Deep -Learning Inference with Multi - Chip-Module-Based Architecture,

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:27.649967Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c4ae1700-ec96-42ac-9db1-f85d3147470c · outbound

This paper cites Learning -Based Application -Agnostic 3D NoC Design for Heterogeneous Manycore Systems,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Learning -Based Application -Agnostic 3D NoC Design for Heterogeneous Manycore Systems,

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:27.812666Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1476f098-0b2f-46be-92b4-482317e4afe3 · outbound

This paper cites A Simulated Annealing -Based Multiobjective Optimization Algorithm: AMOSA,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts A Simulated Annealing -Based Multiobjective Optimization Algorithm: AMOSA,

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:27.997055Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 957dc3a7-4439-4227-ad34-6d19838e6730 · outbound

This paper cites DNN+NeuroSim V2.0: An End -to-End Benchmarking Framework for Compute -in-Memory Accelerators for On -Chip Training,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts DNN+NeuroSim V2.0: An End -to-End Benchmarking Framework for Compute -in-Memory Accelerators for On -Chip Training,

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:28.129727Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e9643936-f20d-470c-b06f-6ecb041affb4 · outbound

This paper cites A detailed and flexible cycle -accurate Network-on-Chip simulator,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts A detailed and flexible cycle -accurate Network-on-Chip simulator,

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:28.296761Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f71c0d3-324f-4165-a28a-11446816320e · outbound

This paper cites Hotspot 6.0: Validation, accelacceleration and Extension,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Hotspot 6.0: Validation, accelacceleration and Extension,

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:28.423534Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7722b37-82fc-44a1-af1f-858e449b68f5 · outbound

This paper cites A Survey of Design and Optimization for Systolic Array - based DNN Accelerators,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts A Survey of Design and Optimization for Systolic Array - based DNN Accelerators,

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:28.555550Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e6886916-53da-489e-8e6d-96b1af4f1470 · outbound

This paper cites Ultra High Density SoIC with Sub-micron Bond Pitch,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Ultra High Density SoIC with Sub-micron Bond Pitch,

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:28.623547Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 21032911-52f6-4b5d-bf7c-7bed2b35a700 · outbound

This paper cites System on Integrated Chips (SoIC(TM) for 3D Heterogeneous Integration,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts System on Integrated Chips (SoIC(TM) for 3D Heterogeneous Integration,

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:28.768517Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a20fa63e-4428-41e3-af34-d7153cd62b58 · outbound

This paper cites Efficient Memory Management for Large Language Model Serving with PagedAttention,.

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts Efficient Memory Management for Large Language Model Serving with PagedAttention,

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:28.905462Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 40b88454-160b-42c3-92a1-c506ff20423d · outbound

This paper cites [Online].

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts [Online]

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:29.004045Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8ecd680-8232-4761-a169-ff91e9dd3c2f · outbound

This paper cites [Online].

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts [Online]

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:29.127717Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 153f92f2-6351-4ff1-a2b2-0842f9689ddb · outbound

This paper cites [Online].

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts [Online]

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:29.298878Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2975c53a-2d34-4e39-bb13-1a8c5f439bcb · outbound

This paper cites [Online].

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts [Online]

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:29.471278Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8dcb13b-c9a5-4dc9-a727-545f313e554e · outbound

This paper cites [Online].

ThAME: 3D Memory-Enabled Heterogeneous Accelerator for LLM Mixture of Experts [Online]

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-04T04:18:29.575504Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers