Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention | Paper Citation Record

Observation 8d7dd8e5-e608-48ad-a652-6c400b045177 · outbound

This paper cites Nvidia h200 sxm 141 gb,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Nvidia h200 sxm 141 gb,

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.149323Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d23cb950-4071-4111-9db1-c23f2d3ddbc6 · outbound

This paper cites TensorRT-LLM,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention TensorRT-LLM,

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.153945Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d994fae3-7ee8-4ec8-b00e-d5c596aeb7bd · outbound

This paper cites Lmcache: Turboboosting vllm with 7x faster access to 100x more kv caches,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Lmcache: Turboboosting vllm with 7x faster access to 100x more kv caches,

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.160343Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation daf5a3b2-9e11-4e41-8adf-64eae6a3906f · outbound

This paper cites Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities,

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.165570Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 50801cb7-55e5-45f7-8934-e0966f654ede · outbound

This paper cites Openai models,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Openai models,

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.170342Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05d2b9d8-8946-40c5-b0f0-fabe509eb385 · outbound

This paper cites Taming throughput-latency tradeoff in llm inference with sarathi-serve,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Taming throughput-latency tradeoff in llm inference with sarathi-serve,

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.175839Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 97c43c0c-55b5-4cba-ac65-1b894d4faa99 · outbound

This paper cites Claude Opus 5,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Claude Opus 5,

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.180450Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6def015d-e2ff-4748-916b-0361ec8adddd · outbound

This paper cites Introducing claude opus 4.6,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Introducing claude opus 4.6,

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.185398Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4d5c634b-397f-411d-b485-69997b07cd7d · outbound

This paper cites Indexcache: Accelerating sparse attention via cross-layer index reuse,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Indexcache: Accelerating sparse attention via cross-layer index reuse,

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.190238Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d889abdc-da70-4da1-b7bb-a0f4fc511191 · outbound

This paper cites SWE-chat: Coding Agent Interactions From Real Users in the Wild.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention SWE-chat: Coding Agent Interactions From Real Users in the Wild

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.194675Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e795f554-5ee1-49e2-980a-e83d6624a0de · outbound

This paper cites RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM Inference.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM Inference

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.200352Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9785c5b9-9ee9-42fa-8391-63d5b9581fda · outbound

This paper cites Magicpig: LSH sampling for efficient LLM generation,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Magicpig: LSH sampling for efficient LLM generation,

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.205506Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 27524ffa-a2f9-4341-8c8b-ed4d34ffd408 · outbound

This paper cites Llmservingsim: A hw/sw co-simulation infrastructure for llm inference serving at scale,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Llmservingsim: A hw/sw co-simulation infrastructure for llm inference serving at scale,

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.210220Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7fb6dc67-2ed9-4e65-b40a-bb1bd3beef83 · outbound

This paper cites 37.3 a 2nm all-digital 14.4gb/s/pin lpddr6 phy with quarter-rate clocking architecture and multi-level fifo-based speculative dfe,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention 37.3 a 2nm all-digital 14.4gb/s/pin lpddr6 phy with quarter-rate clocking architecture and multi-level fifo-based speculative dfe,

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.214380Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b00c4d4e-a972-48a4-8271-5d7730099d97 · outbound

This paper cites DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.218841Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0607bc4-f56d-4fcc-8a80-0e078adbc902 · outbound

This paper cites Deepseek-v3.2: Pushing the frontier of open large language models,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Deepseek-v3.2: Pushing the frontier of open large language models,

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.227132Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 41eab178-fbd1-4a7e-aa98-be1626eb6d5e · outbound

This paper cites DeepSeek-V3 Technical Report.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention DeepSeek-V3 Technical Report

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.232105Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4698871e-9424-4072-8744-99c442d1eaf6 · outbound

This paper cites SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.236994Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b9c54e13-78da-4267-8386-4ae5bc89ecfe · outbound

This paper cites com/EPIC-RPI/STARC, 2025, accessed: 2026-08-01.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention com/EPIC-RPI/STARC, 2025, accessed: 2026-08-01

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.242889Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f4c4fff2-a8d7-44ad-81e9-fc3a4fe0344f · outbound

This paper cites Starc: Selective token access with remapping and clustering for efficient llm decoding on pim systems,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Starc: Selective token access with remapping and clustering for efficient llm decoding on pim systems,

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.247307Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5d98f96b-0faf-44ae-b8f7-66296dc882d2 · outbound

This paper cites Mtia: First generation silicon targeting meta’s recommendation systems,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Mtia: First generation silicon targeting meta’s recommendation systems,

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.251826Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bc31ffd5-61dd-4a79-9e0d-42d1c00e7906 · outbound

This paper cites kv-cache-tester: Inference server cache performance test- ing suite,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention kv-cache-tester: Inference server cache performance test- ing suite,

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.261315Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0c3fe003-8250-4d43-8360-0e285464e383 · outbound

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

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Rdma over ethernet for distributed training at meta scale,

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.266049Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b3c33a8c-2718-48b3-b248-4acb9635012f · outbound

This paper cites Seerattention: Self-distilled attention gating for efficient long-context prefilling,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Seerattention: Self-distilled attention gating for efficient long-context prefilling,

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.272118Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dccc7dd2-bd06-474d-97f5-bd388c7914fc · outbound

This paper cites Glm-5: from vibe coding to agentic engineering,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Glm-5: from vibe coding to agentic engineering,

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.277907Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5a398715-204c-44e4-bc15-693948fe7d10 · outbound

This paper cites The Llama 3 Herd of Models.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention The Llama 3 Herd of Models

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.290966Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eb81e072-00eb-482a-a58b-409c75c5b507 · outbound

This paper cites Pim is all you need: A cxl-enabled gpu-free system for large language model inference,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Pim is all you need: A cxl-enabled gpu-free system for large language model inference,

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.297214Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e4511efe-4c01-4c32-b566-97888a99794f · outbound

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

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Fastermoe: modeling and optimizing training of large-scale dynamic pre-trained models,

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.307638Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7554c8b6-9059-4353-9e6f-0eecee72c60e · outbound

This paper cites Shyla: 3d-stacked nvm-dram hybrid llm-inference ar- chitecture exploiting data and memory heterogeneity,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Shyla: 3d-stacked nvm-dram hybrid llm-inference ar- chitecture exploiting data and memory heterogeneity,

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.312140Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0606b9f1-2d6e-4b29-bccd-e19b59a675d8 · outbound

This paper cites Transparent offloading and mapping (tom): enabling programmer-transparent near-data processing in gpu systems,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Transparent offloading and mapping (tom): enabling programmer-transparent near-data processing in gpu systems,

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.322232Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 92794a87-85ac-4989-aae7-21eccd6726f9 · outbound

This paper cites Hybridspec: Exploiting hybrid-bonding memory to accelerate llm serving through heterogeneous architecture and specu- lative decoding,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Hybridspec: Exploiting hybrid-bonding memory to accelerate llm serving through heterogeneous architecture and specu- lative decoding,

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.327773Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 65ce5f7e-a2de-4057-a3af-d8051eeb72cb · outbound

This paper cites codex_swebenchpro_traces,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention codex_swebenchpro_traces,

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.405442Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation ff04b0c3-e21d-481b-bce7-1aeceec7758c · outbound

This paper cites A cost-effective near-storage processing solution for offline inference of long-context llms,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention A cost-effective near-storage processing solution for offline inference of long-context llms,

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.337036Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4357d5ee-87d6-4671-bd19-378565f75dbc · outbound

This paper cites 2023, standard.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention 2023, standard

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.388832Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 292a8f94-8850-42e8-abe0-a5218dc819a0 · outbound

This paper cites Scalable processing-near-memory for 1m-token llm inference: Cxl-enabled kv-cache management beyond gpu limits,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Scalable processing-near-memory for 1m-token llm inference: Cxl-enabled kv-cache management beyond gpu limits,

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.373609Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 27e6b1ca-3cef-41fc-802e-ed0b75920cb8 · outbound

This paper cites Toward standardized near-data processing with unrestricted data placement for gpus,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Toward standardized near-data processing with unrestricted data placement for gpus,

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.351827Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bdb24d7b-fb77-4ad6-bc34-c67397c21fca · outbound

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

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Samsung pim/pnm for transfmer based ai : Energy efficiency on pim/pnm cluster,

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.357756Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 3455db69-509a-4875-bd3a-7c51965ffa3f · outbound

This paper cites A silicon-proven unified low-latency CXL controller and port-based routing switch for memory-centric fabrics,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention A silicon-proven unified low-latency CXL controller and port-based routing switch for memory-centric fabrics,

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.342205Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 465ec14d-4bfc-402e-8413-0ec92401e7ef · outbound

This paper cites Efficient memory management for large language model serving with pagedattention,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Efficient memory management for large language model serving with pagedattention,

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.370496Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 00f128ec-72ba-4720-aa6a-5f82733ceee4 · outbound

This paper cites MiniMax Sparse Attention.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention MiniMax Sparse Attention

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.375091Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b6655629-ac27-42ea-94e6-43883d1bb4c8 · outbound

This paper cites Hardware architecture and software stack for pim based on commercial dram technology,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Hardware architecture and software stack for pim based on commercial dram technology,

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.380038Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a83c82e6-8200-4657-bce0-f1358b096d23 · outbound

This paper cites Pond: Cxl-based memory pooling systems for cloud platforms,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Pond: Cxl-based memory pooling systems for cloud platforms,

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.384812Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f38912b-813f-451a-9ab3-85f17b37e38b · outbound

This paper cites Snapkv: Llm knows what you are looking for before generation,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Snapkv: Llm knows what you are looking for before generation,

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.327089Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 7f6498db-72a4-45f6-90a3-f75c2a6c6dfb · outbound

This paper cites Meridian: In-memory acceleration for rag with document attention decomposition,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Meridian: In-memory acceleration for rag with document attention decomposition,

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.311907Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 71aad5d4-c91f-49c2-858b-41b4ce41751c · outbound

This paper cites Chime: A case for efficient long-context attention- fc disaggregated inference with dimm-pim,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Chime: A case for efficient long-context attention- fc disaggregated inference with dimm-pim,

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.296688Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 869f7343-2506-4a36-b35b-e6b316f3042e · outbound

This paper cites Nvidia’s ad102 officially revealed, how close were the previ- ous estimates?.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Nvidia’s ad102 officially revealed, how close were the previ- ous estimates?

Reference 49

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.281663Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 829d515a-1328-484d-a2a7-e0dfb5104868 · outbound

This paper cites MoBA: Mixture of block attention for long- context LLMs,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention MoBA: Mixture of block attention for long- context LLMs,

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.266183Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 4b1472e6-8049-4b9b-a3fd-0d908548c702 · outbound

This paper cites Sac: Disaggregated kv cache system for sparse attention llms with cxl,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Sac: Disaggregated kv cache system for sparse attention llms with cxl,

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.249064Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation f4247f77-a36b-4cbb-bf91-6c1dde323abc · outbound

This paper cites Tpp: Transparent page placement for cxl-enabled tiered-memory,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Tpp: Transparent page placement for cxl-enabled tiered-memory,

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.416938Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 64f75a88-0540-4e21-9f37-6939e5b7260f · outbound

This paper cites HBM3E product brief,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention HBM3E product brief,

Reference 53

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.230476Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 7c1242cb-69e5-425b-9301-cc03127b2e7c · outbound

This paper cites g ed., May 2025, production Data Sheet.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention g ed., May 2025, production Data Sheet

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.214140Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 7e09f9a6-ee34-4308-aa07-d9b8b4f6cbd2 · outbound

This paper cites Early silicon of raptor: The first 3d-dram accelerator for generative inference,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Early silicon of raptor: The first 3d-dram accelerator for generative inference,

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.196710Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 9836d34e-ce16-4ebf-89ad-ae9c2ff349d2 · outbound

This paper cites NVIDIA DGX B200 datasheet,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention NVIDIA DGX B200 datasheet,

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.177834Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 1e5d862b-f97e-4795-b713-89da268bb6e0 · outbound

This paper cites NVIDIA DGX H200 datasheet,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention NVIDIA DGX H200 datasheet,

Reference 58

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.155632Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation bc8a10d5-e6e6-443c-b7bc-1192e7f07541 · outbound

This paper cites SLA-based planner,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention SLA-based planner,

Reference 59

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.133856Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 40a02988-eb34-4758-ae43-b55623d62339 · outbound

This paper cites Multi-process service: Appendix: Tools and in- terface reference,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Multi-process service: Appendix: Tools and in- terface reference,

Reference 60

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.114628Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 91f4f66e-9c28-4902-9863-85885036bb75 · outbound

This paper cites (2026) Overall architecture — NVIDIA Dynamo documentation.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention (2026) Overall architecture — NVIDIA Dynamo documentation

Reference 61

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.096891Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 28c3453f-11da-4719-85fc-e20312900b88 · outbound

This paper cites Planner design,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Planner design,

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.078680Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation d2f30c89-eb67-4ec1-9181-69f50c4b9246 · outbound

This paper cites Supported MIG profiles — NVIDIA multi- instance GPU user guide,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Supported MIG profiles — NVIDIA multi- instance GPU user guide,

Reference 63

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.057883Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 0a47aa5c-f6f7-4039-821b-fa75e4d27b88 · outbound

This paper cites Fine-grained dram: energy-efficient dram for extreme bandwidth systems,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Fine-grained dram: energy-efficient dram for extreme bandwidth systems,

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.471604Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 99c0a576-0a4e-4b7c-ba8f-b01174e431c9 · outbound

This paper cites Exegpt: Constraint-aware resource scheduling for llm inference,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Exegpt: Constraint-aware resource scheduling for llm inference,

Reference 65

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.476730Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1e245943-ff76-4cb3-a140-40ef552e5bfe · outbound

This paper cites ChatGPT (GPT-5),.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention ChatGPT (GPT-5),

Reference 66

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:58:00.013479Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation ace60453-3340-4802-81d5-92b0b982d7e2 · outbound

This paper cites Attacc! unleashing the power of pim for batched transformer-based generative model inference,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Attacc! unleashing the power of pim for batched transformer-based generative model inference,

Reference 68

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.490666Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4be5218b-1826-48f2-908a-3323625201bd · outbound

This paper cites A 192-gb 12-high 896-gb/s hbm3 dram with a tsv auto-calibration scheme and machine-learning-based layout opti- mization,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention A 192-gb 12-high 896-gb/s hbm3 dram with a tsv auto-calibration scheme and machine-learning-based layout opti- mization,

Reference 69

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.987872Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 3bfbb05d-b684-4ff6-8cb7-ac3d9e0cfbc4 · outbound

This paper cites An lpddr-based cxl-pnm platform for tco- efficient inference of transformer-based large language models,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention An lpddr-based cxl-pnm platform for tco- efficient inference of transformer-based large language models,

Reference 70

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.500163Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1341262d-1f59-4dc4-b3de-789a5a15811a · outbound

This paper cites Apple m2 die shot and architecture analysis – big cost increase and a15 based ip,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Apple m2 die shot and architecture analysis – big cost increase and a15 based ip,

Reference 71

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.969080Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 38d07b6e-9fa6-4c2c-a51e-0ade696dcac8 · outbound

This paper cites Splitwise: Efficient generative llm inference using phase splitting,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Splitwise: Efficient generative llm inference using phase splitting,

Reference 72

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.509113Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4eaec88c-7719-437a-afd7-fd967a84a59f · outbound

This paper cites Mooncake: A kvcache-centric disaggregated architecture for llm serving,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Mooncake: A kvcache-centric disaggregated architecture for llm serving,

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.513839Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation be2c78d2-20ee-4db1-91b9-59affbe87787 · outbound

This paper cites Efficient Interactive LLM Serving with Proxy Model-based Sequence Length Prediction.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Efficient Interactive LLM Serving with Proxy Model-based Sequence Length Prediction

Reference 74

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.518164Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4717908-7610-4d28-b5b2-78b1b4a751c9 · outbound

This paper cites Longsight: Compute-enabled memory to accelerate large-context llms via sparse attention,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Longsight: Compute-enabled memory to accelerate large-context llms via sparse attention,

Reference 75

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.527863Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f44faa9-2424-42e0-b960-19ece375ed0d · outbound

This paper cites Drex: Accurate and scalable dense retrieval acceleration via algorithmic-hardware codesign,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Drex: Accurate and scalable dense retrieval acceleration via algorithmic-hardware codesign,

Reference 76

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.532714Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 25ac2814-db2d-45f2-8542-7c5d76ec5873 · outbound

This paper cites Sparq attention: bandwidth-efficient llm inference,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Sparq attention: bandwidth-efficient llm inference,

Reference 77

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.951661Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation ecf9d468-0bbf-4022-b109-7beef9a8a122 · outbound

This paper cites AttAcc simulator,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention AttAcc simulator,

Reference 78

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.933904Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 4abada64-ac32-424b-9d45-198da8bbcdc5 · outbound

This paper cites LLMSimulator,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention LLMSimulator,

Reference 79

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.915842Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 9bdce448-1a96-4225-ae22-eda8614c7bd6 · outbound

This paper cites Toolformer: Language models can teach themselves to use tools,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Toolformer: Language models can teach themselves to use tools,

Reference 80

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.896406Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 78165bad-bf47-4527-a14a-fc563f2e7431 · outbound

This paper cites Ianus: Integrated accelerator based on npu-pim unified memory system,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Ianus: Integrated accelerator based on npu-pim unified memory system,

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.561610Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 218598e1-5126-43d3-8251-a7b889123e80 · outbound

This paper cites Dynamollm: Designing llm inference clusters for performance and energy efficiency,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Dynamollm: Designing llm inference clusters for performance and energy efficiency,

Reference 82

Resolution: malformed identifier; no resolver link, observed 2026-08-15T14:57:57.567291Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation da7ae6ab-79c6-4dbb-a06e-cce59b3194c7 · outbound

This paper cites Quest: query-aware sparsity for efficient long-context llm inference,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Quest: query-aware sparsity for efficient long-context llm inference,

Reference 83

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.877514Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 667ab8f0-9d40-440b-b974-95062c1b1d3e · outbound

This paper cites Astra-sim2.0: Modeling hierarchical networks and disaggregated systems for large-model training at scale,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Astra-sim2.0: Modeling hierarchical networks and disaggregated systems for large-model training at scale,

Reference 84

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.859074Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 839c0458-9fd5-4f3c-a139-8460017ad2ea · outbound

This paper cites MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention MemExplorer: Navigating the Heterogeneous Memory Design Space for Agentic Inference NPUs

Reference 85

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.583167Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e503753f-a17d-4fc5-8d8e-ab0748aa2ea3 · outbound

This paper cites Efficient streaming language models with attention sinks,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Efficient streaming language models with attention sinks,

Reference 86

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.840387Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 548abcf4-5487-455f-8d76-53761d9825d2 · outbound

This paper cites Hisparse: Turbocharging sparse attention with hierarchical memory,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Hisparse: Turbocharging sparse attention with hierarchical memory,

Reference 87

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.817534Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 06aa777f-2162-43e5-90a6-415a16325328 · outbound

This paper cites Strata: Hierarchical Context Caching for Long Context Language Model Serving.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Strata: Hierarchical Context Caching for Long Context Language Model Serving

Reference 88

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.598073Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 61352f64-28ee-4604-9d51-645e6cc9726d · outbound

This paper cites Beluga: A cxl-based memory architecture for scalable and efficient llm kvcache management,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Beluga: A cxl-based memory architecture for scalable and efficient llm kvcache management,

Reference 90

Resolution: verified exact; doi, observed 2026-08-15T14:57:57.723542Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 5a859f43-3852-4bb0-b0a9-ed6b4028d22d · outbound

This paper cites ReAct: Synergizing reasoning and acting in language models,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention ReAct: Synergizing reasoning and acting in language models,

Reference 91

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.799077Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 850db07b-93a5-45a3-9604-7db34f945c25 · outbound

This paper cites Tract: Disaggregated llm serving with cxl shared memory kv cache at rack-scale,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Tract: Disaggregated llm serving with cxl shared memory kv cache at rack-scale,

Reference 92

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.623910Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a5ed225-56ef-45e6-8cbb-da43b1be8389 · outbound

This paper cites Available: https://arxiv.org/abs/2601.06288.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Available: https://arxiv.org/abs/2601.06288

Reference 93

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.608756Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1737c7af-5fd7-4fad-b374-3efcad806983 · outbound

This paper cites Duplex: A device for large language models with mixture of experts, grouped query attention, and continuous batching,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Duplex: A device for large language models with mixture of experts, grouped query attention, and continuous batching,

Reference 94

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.634586Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e1732f4f-713b-41b3-a128-263a10bb69cd · outbound

This paper cites H2o: heavy-hitter oracle for efficient generative inference of large language models,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention H2o: heavy-hitter oracle for efficient generative inference of large language models,

Reference 95

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.766301Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 71d4a0a6-e8ec-4b87-a3d9-5714e972d590 · outbound

This paper cites Deepep: an efficient expert-parallel communication library,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Deepep: an efficient expert-parallel communication library,

Reference 96

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.749108Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation d5be8cc2-997c-4877-83ef-d897fb643fe6 · outbound

This paper cites Patterns behind chaos: Forecasting data movement for efficient large-scale moe llm inference,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Patterns behind chaos: Forecasting data movement for efficient large-scale moe llm inference,

Reference 97

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.782718Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 41ca0024-80fe-42d2-b55e-9edb60e1fb5b · outbound

This paper cites Sglang: efficient execution of structured language model programs,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Sglang: efficient execution of structured language model programs,

Reference 98

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.715187Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 3907fcb6-17cb-48ca-8e76-7207edf8feee · outbound

This paper cites Distserve: disaggregating prefill and decoding for goodput-optimized large language model serving,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Distserve: disaggregating prefill and decoding for goodput-optimized large language model serving,

Reference 99

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.695869Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation e239bc84-b2d3-4b0d-ae2d-c14ff1bd4364 · outbound

This paper cites Octopus: Enhancing CXL memory pods via sparse topology,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Octopus: Enhancing CXL memory pods via sparse topology,

Reference 100

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.678165Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 0c47f604-1cf0-4652-8f5f-2db614f4fe2f · outbound

This paper cites InfLLM-v2: Dense-sparse switchable attention for seamless short-to-long adaptation,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention InfLLM-v2: Dense-sparse switchable attention for seamless short-to-long adaptation,

Reference 101

Resolution: verified fuzzy; raw_fallback, observed 2026-08-15T14:57:59.732854Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation 8e9f74b2-3a44-4952-8547-279579b2989a · outbound

This paper cites Megascale-infer: Efficient mixture-of-experts model serving with disaggregated expert parallelism,.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Megascale-infer: Efficient mixture-of-experts model serving with disaggregated expert parallelism,

Reference 105

Resolution: verified exact; doi, observed 2026-08-15T14:57:57.707092Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.

Observation f2a2bbea-f4d4-4576-85e9-cffcebd4f8ee · outbound

This paper cites Available: https://doi.org/10.1145/3579371.3589348.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Available: https://doi.org/10.1145/3579371.3589348

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.256627Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 278d6a21-d664-440c-900d-9e99ab57a0c5 · outbound

This paper cites Efficient Interactive LLM Serving with Proxy Model-based Sequence Length Prediction.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention Efficient Interactive LLM Serving with Proxy Model-based Sequence Length Prediction

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.523025Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 577d7638-0e16-430e-a748-bf2dae12ae21 · outbound

This paper cites GLM-5: from Vibe Coding to Agentic Engineering.

Heterogeneous LLM Serving with General-Purpose Processing-Near-Memory for Retrieval-Based Sparse Attention GLM-5: from Vibe Coding to Agentic Engineering

Reference 2026

Resolution: unresolved; no resolver link, observed 2026-08-15T14:57:57.283172Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Coverage vector

Reference resolution

Outbound references

Pith citing papers