Paper Citation Record · LEDGER

SCBench: A KV Cache-Centric Analysis of Long-Context Methods

As of 15 August 2026, this Paper Citation Record lists 100 of 108 outbound references and 18 inbound Pith citation observations for arXiv:2412.10319.

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

pith.paper-citation-record.v1
2412.10319 v2

Coverage vector

measured 100 of 108 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-11T16:14:06.284069Z

measured 118 of 118 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-15T06:32:42.880941+00:00

measured 18 of 18 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-11T00:38:46.746093Z

measured 1 of 1 external citation measurements

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

Source: pith, observed 2026-08-05T02:28:24.338817Z

Reference resolution

100 of 108 outbound references displayed

verified exact1
verified fuzzy18
unresolved81
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

1
pith, observed 2026-08-05T02:28:24.338817Z

Outbound references

Observation 18bfbac8-1a26-4962-a053-34f6e6c8b3d8 · outbound

This paper cites Star Attention: Efficient LLM Inference over Long Sequences.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Star Attention: Efficient LLM Inference over Long Sequences

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.493385Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fe095b35-d5b3-4ae7-9bb4-449b65ffcff4 · outbound

This paper cites Chan, Biao Zhang, Ankesh Anand, Zaheer Abbas, Azade Nova, John D Co-Reyes, Eric Chu, Feryal Behbahani, Aleksandra Faust, and Hugo Larochelle.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Chan, Biao Zhang, Ankesh Anand, Zaheer Abbas, Azade Nova, John D Co-Reyes, Eric Chu, Feryal Behbahani, Aleksandra Faust, and Hugo Larochelle

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.503321Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f3d474b6-4284-4ba3-956c-71b50c3013e0 · outbound

This paper cites GQA : Training generalized multi-query transformer models from multi-head checkpoints.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods GQA : Training generalized multi-query transformer models from multi-head checkpoints

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.508289Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7e4e5e93-2df0-4c68-8909-a77e8e23a17c · outbound

This paper cites Just read twice: closing the recall gap for recurrent language models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Just read twice: closing the recall gap for recurrent language models

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.513213Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 778e19cc-45a0-4def-ae16-b81f1c555fad · outbound

This paper cites Prompt caching.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Prompt caching

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.525308Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 80301961-0e49-4895-b67b-ec1c15b0f668 · outbound

This paper cites MT -bench-101: A fine-grained benchmark for evaluating large language models in multi-turn dialogues.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods MT -bench-101: A fine-grained benchmark for evaluating large language models in multi-turn dialogues

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.529601Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2d8fde69-847a-4c0c-848e-abd4a8714e7a · outbound

This paper cites L ong B ench: A bilingual, multitask benchmark for long context understanding.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods L ong B ench: A bilingual, multitask benchmark for long context understanding

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.534148Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 17b9cd6d-5fb8-4c25-84d3-7230803902e4 · outbound

This paper cites Codeplan: Repository-level coding using llms and planning.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Codeplan: Repository-level coding using llms and planning

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.538651Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation aa38c05d-75a8-4446-a94f-03b638986935 · outbound

This paper cites Longformer: The Long-Document Transformer.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Longformer: The Long-Document Transformer

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.545976Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0ed946b5-9a6f-4b91-99ed-b776ff242134 · outbound

This paper cites Peek across: Improving multi-document modeling via cross-document question-answering.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Peek across: Improving multi-document modeling via cross-document question-answering

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.552095Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 73932485-a9cf-43e2-b151-a82d3efaa9b2 · outbound

This paper cites PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.559074Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3c7b7770-b0a0-42dd-8a93-46b381e2c2c9 · outbound

This paper cites Magic PIG : LSH sampling for efficient LLM generation.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Magic PIG : LSH sampling for efficient LLM generation

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.569975Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d706461c-637b-47df-aaff-cc0366b35892 · outbound

This paper cites Generating Long Sequences with Sparse Transformers.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Generating Long Sequences with Sparse Transformers

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.581212Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7bf95e5c-4ba6-4af8-a9e8-a30482e60893 · outbound

This paper cites Prompt caching with claude.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Prompt caching with claude

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.587485Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 10688121-8db5-442a-8783-bade2f68d766 · outbound

This paper cites CORM: Cache Optimization with Recent Message for Large Language Model Inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods CORM: Cache Optimization with Recent Message for Large Language Model Inference

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.592172Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 40383359-a7e9-456b-8b68-d751478e04a3 · outbound

This paper cites Flashattention-2: Faster attention with better parallelism and work partitioning.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Flashattention-2: Faster attention with better parallelism and work partitioning

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.597645Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 177686b2-3b3b-456e-8163-95a03c989e42 · outbound

This paper cites Transformers are SSM s: Generalized models and efficient algorithms through structured state space duality.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Transformers are SSM s: Generalized models and efficient algorithms through structured state space duality

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.602037Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 271fafd0-d81b-455e-a805-d2108fa7a407 · outbound

This paper cites How Sparse Attention Approximates Exact Attention? Your Attention is Naturally $n^C$-Sparse.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods How Sparse Attention Approximates Exact Attention? Your Attention is Naturally $n^C$-Sparse

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.607282Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ec997631-019a-481a-a033-8cbb04b67af7 · outbound

This paper cites HashAttention: Semantic Sparsity for Faster Inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods HashAttention: Semantic Sparsity for Faster Inference

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.613937Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 798bc6bc-32cf-45df-979d-45d6da57d34b · outbound

This paper cites Domeccleston/sharegpt: Easily share permanent links to chatgpt conversations with your friends.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Domeccleston/sharegpt: Easily share permanent links to chatgpt conversations with your friends

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.619168Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 08aea4bc-d5c9-464e-a316-4f81c94c17c1 · outbound

This paper cites The Llama 3 Herd of Models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods The Llama 3 Herd of Models

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.623861Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 26e6dfca-560a-45fb-a6fc-12cb7371dd4f · outbound

This paper cites Finch, James D.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Finch, James D

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.629895Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d15a8e5f-af64-4ccf-9c5d-0fad4fefbf19 · outbound

This paper cites Moa: Mixture of sparse attention for automatic large language model compression.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Moa: Mixture of sparse attention for automatic large language model compression

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.635486Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f1c321da-c372-4aae-a834-93fbed9de7a6 · outbound

This paper cites Challenges in Deploying Long-Context Transformers: A Theoretical Peak Performance Analysis.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Challenges in Deploying Long-Context Transformers: A Theoretical Peak Performance Analysis

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.640877Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 511a06ce-6c18-463f-930b-602a4d9d3434 · outbound

This paper cites Model tells you what to discard: Adaptive kv cache compression for llms.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Model tells you what to discard: Adaptive kv cache compression for llms

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.647156Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d30ef9fe-2a14-4169-9ae6-8e330e7e30e3 · outbound

This paper cites Context caching.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Context caching

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.652480Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1a7dda80-cd62-457f-ac49-5dd93d54be83 · outbound

This paper cites Prompt cache: Modular attention reuse for low-latency inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Prompt cache: Modular attention reuse for low-latency inference

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.657245Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1c18c550-1c1e-4c77-9fb3-a4e1570cf463 · outbound

This paper cites ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.665083Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fdb52d04-7c0a-44a2-8ffb-3a156f50195b · outbound

This paper cites Llama-3 8b instruct gradient 4194k (v0.1), 2024.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Llama-3 8b instruct gradient 4194k (v0.1), 2024

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.672513Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 988a1d49-9e67-410c-bc1b-fb2154793c53 · outbound

This paper cites Mamba: Linear-time sequence modeling with selective state spaces.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Mamba: Linear-time sequence modeling with selective state spaces

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.678164Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2840c81b-80dc-4b34-87ca-5cc2f189b002 · outbound

This paper cites Efficiently modeling long sequences with structured state spaces.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Efficiently modeling long sequences with structured state spaces

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.683448Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 39ff0f01-ef68-4669-9e3f-bfa802775362 · outbound

This paper cites LM -infinite: Zero-shot extreme length generalization for large language models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods LM -infinite: Zero-shot extreme length generalization for large language models

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.690698Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 25bf92da-2bfa-4482-b0fe-fd924d4b3a86 · outbound

This paper cites Structured Prompting: Scaling In-Context Learning to 1,000 Examples.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Structured Prompting: Scaling In-Context Learning to 1,000 Examples

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.700159Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 25410f7b-8665-4a5a-b8cb-112703d47632 · outbound

This paper cites Liquid structural state-space models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Liquid structural state-space models

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.720243Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 42757b06-926f-4a90-8808-b624a3571c08 · outbound

This paper cites Block transformer: Global-to-local language modeling for fast inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Block transformer: Global-to-local language modeling for fast inference

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.735108Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 23d9cb6d-f612-4719-b520-b8d8907a0ddc · outbound

This paper cites Squeezed attention: Accelerating long context length llm inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Squeezed attention: Accelerating long context length llm inference

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.745921Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b651caa1-ee07-4f2a-9acd-2c460a63bf33 · outbound

This paper cites RULER : What s the real context size of your long-context language models? In First Conference on Language Modeling, 2024.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods RULER : What s the real context size of your long-context language models? In First Conference on Language Modeling, 2024

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.753247Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b41d645b-fac8-422c-a97e-7963cec17c38 · outbound

This paper cites Kakade, and eran malach.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Kakade, and eran malach

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.759270Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9cef39e2-c13b-4ca8-81e8-75edb810893e · outbound

This paper cites LLML ingua: Compressing prompts for accelerated inference of large language models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods LLML ingua: Compressing prompts for accelerated inference of large language models

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.766023Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8bb5c15d-81cb-48e1-91d0-6474f38e97f7 · outbound

This paper cites Abdi, Dongsheng Li, Chin-Yew Lin, Yuqing Yang, and Lili Qiu.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Abdi, Dongsheng Li, Chin-Yew Lin, Yuqing Yang, and Lili Qiu

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.783041Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8d3ddb03-18bc-483b-b4e9-cfdf9b52352b · outbound

This paper cites SWE -bench: Can language models resolve real-world github issues? In The Twelfth International Conference on Learning Representations, 2024.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods SWE -bench: Can language models resolve real-world github issues? In The Twelfth International Conference on Learning Representations, 2024

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.792099Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d56d8df0-3248-4452-b84f-35c3115b04bd · outbound

This paper cites RAGCache: Efficient Knowledge Caching for Retrieval-Augmented Generation.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods RAGCache: Efficient Knowledge Caching for Retrieval-Augmented Generation

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.801823Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c69b9e64-bd9a-49ad-892b-78695d0c61ba · outbound

This paper cites Hydragen: High-Throughput LLM Inference with Shared Prefixes.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Hydragen: High-Throughput LLM Inference with Shared Prefixes

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.806863Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4c4f7554-03f9-4849-a93e-f6ca12eb31bb · outbound

This paper cites Needle in a haystack - pressure testing llms, 2023.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Needle in a haystack - pressure testing llms, 2023

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.815247Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9d8f3ddb-240e-4c67-87e2-8732c273e502 · outbound

This paper cites MT -eval: A multi-turn capabilities evaluation benchmark for large language models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods MT -eval: A multi-turn capabilities evaluation benchmark for large language models

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.825768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b17a2e77-4efd-492c-b86d-abc664291541 · outbound

This paper cites Flexprefill: A context-aware sparse attention mechanism for efficient long-sequence inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Flexprefill: A context-aware sparse attention mechanism for efficient long-sequence inference

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.844002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 91252970-082f-436a-b988-b8afdb9de1d0 · outbound

This paper cites an unresolved cited work.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Unresolved cited work

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.849167Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 22c5b984-f9f3-48c7-94db-7e96b6bac95e · outbound

This paper cites Long-context LLMs Struggle with Long In-context Learning.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Long-context LLMs Struggle with Long In-context Learning

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.857578Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1e06a6e8-1e1f-48d2-9422-5b606cd61a2f · outbound

This paper cites Alpacaeval: An automatic evaluator of instruction-following models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Alpacaeval: An automatic evaluator of instruction-following models

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.863375Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d02dc9c8-06c0-4244-9eb8-390e0182d5dd · outbound

This paper cites Compressing context to enhance inference efficiency of large language models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Compressing context to enhance inference efficiency of large language models

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.869749Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6a8e5304-f31f-400a-99ea-92ac8fbcce98 · outbound

This paper cites Snap KV : LLM knows what you are looking for before generation.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Snap KV : LLM knows what you are looking for before generation

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.874814Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 658faf26-dac0-465a-b45c-4a31b19df96a · outbound

This paper cites Jamba: A Hybrid Transformer-Mamba Language Model.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Jamba: A Hybrid Transformer-Mamba Language Model

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.881945Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c225df22-e0e7-48de-b586-3e2de20edeba · outbound

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

SCBench: A KV Cache-Centric Analysis of Long-Context Methods DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.889712Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 527ca086-9439-4b0c-a600-1840b4307bc6 · outbound

This paper cites RetrievalAttention: Accelerating Long-Context LLM Inference via Vector Retrieval.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods RetrievalAttention: Accelerating Long-Context LLM Inference via Vector Retrieval

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.896609Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b80d5d13-cd3d-472b-90b4-69eccce7945f · outbound

This paper cites RepoQA: Evaluating Long Context Code Understanding.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods RepoQA: Evaluating Long Context Code Understanding

Reference 56

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.900988Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 528e4dd8-7aea-4814-8304-7e4008d8eec8 · outbound

This paper cites Liu, Kevin Lin, John Hewitt, Ashwin Paranjape, Michele Bevilacqua, Fabio Petroni, and Percy Liang.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Liu, Kevin Lin, John Hewitt, Ashwin Paranjape, Michele Bevilacqua, Fabio Petroni, and Percy Liang

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.908218Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f85d36f6-f265-4014-8850-9183cf7a44dc · outbound

This paper cites Scissorhands: Exploiting the persistence of importance hypothesis for LLM KV cache compression at test time.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Scissorhands: Exploiting the persistence of importance hypothesis for LLM KV cache compression at test time

Reference 58

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.915658Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 889c9ea3-8860-4e6d-9b37-0aba53f14722 · outbound

This paper cites KIVI : A tuning-free asymmetric 2bit quantization for KV cache.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods KIVI : A tuning-free asymmetric 2bit quantization for KV cache

Reference 59

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.925832Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 198d9c45-aebe-4cf7-94b8-19a502c710ed · outbound

This paper cites MoBA: Mixture of Block Attention for Long-Context LLMs.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods MoBA: Mixture of Block Attention for Long-Context LLMs

Reference 60

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.931089Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e1793aa7-df82-4c94-bfbb-173c9766c5d5 · outbound

This paper cites Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention

Reference 61

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.937098Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9c22a376-d722-4ce0-a32e-563170ac771f · outbound

This paper cites Dynamic memory compression: Retrofitting LLM s for accelerated inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Dynamic memory compression: Retrofitting LLM s for accelerated inference

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.817249Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 2deae104-22b9-4784-98a1-a821e77acbf0 · outbound

This paper cites In-context Learning and Induction Heads.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods In-context Learning and Induction Heads

Reference 63

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.947431Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 17ac8824-9b32-44bc-9055-29430d43b639 · outbound

This paper cites Learning to reason with llms, 2024 a.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Learning to reason with llms, 2024 a

Reference 64

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.955388Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2969f56b-40b0-46e9-ba54-2f456969c045 · outbound

This paper cites Prompt caching in the api.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Prompt caching in the api

Reference 65

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.768455Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 440829c8-ba64-465d-80c8-b0e52d2a7e6c · outbound

This paper cites Vicky Zhao, Lili Qiu, and Dongmei Zhang.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Vicky Zhao, Lili Qiu, and Dongmei Zhang

Reference 66

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.971476Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b4462b04-5a87-40b8-9c9f-3f69d08ad7f7 · outbound

This paper cites O'Brien, Carrie J.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods O'Brien, Carrie J

Reference 67

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.726914Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 8943bcd6-a291-471e-9472-6c713bc89683 · outbound

This paper cites Wind, Stanis aw Wo \'z niak, Zhenyuan Zhang, Qinghua Zhou, Jian Zhu, and Rui-Jie Zhu.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Wind, Stanis aw Wo \'z niak, Zhenyuan Zhang, Qinghua Zhou, Jian Zhu, and Rui-Jie Zhu

Reference 68

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.705590Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation f24dec16-aa6e-42a6-a9aa-ed107cb8f74e · outbound

This paper cites Mooncake: Trading more storage for less computation a KVCache-centric architecture for serving LLM chatbot.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Mooncake: Trading more storage for less computation a KVCache-centric architecture for serving LLM chatbot

Reference 69

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.619077Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 3ff945e3-5e2c-469c-bde2-1663884c73e6 · outbound

This paper cites Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

Reference 70

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:05.996739Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a65b7ef8-78f9-455b-ab0e-e1f3982b3294 · outbound

This paper cites Samba: Simple hybrid state space models for efficient unlimited context language modeling.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Samba: Simple hybrid state space models for efficient unlimited context language modeling

Reference 71

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.006947Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 494b5aa1-fe8b-4634-bfdb-73fd36a4771f · outbound

This paper cites Sparq attention: Bandwidth-efficient LLM inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Sparq attention: Bandwidth-efficient LLM inference

Reference 72

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.485959Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 4587d32f-fbb7-401d-9932-70542f847662 · outbound

This paper cites Fast Transformer Decoding: One Write-Head is All You Need.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Fast Transformer Decoding: One Write-Head is All You Need

Reference 73

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.017822Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 222e6612-2b0c-494e-8b0d-08ba82ccf2bf · outbound

This paper cites F lex G en: High-throughput generative inference of large language models with a single GPU.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods F lex G en: High-throughput generative inference of large language models with a single GPU

Reference 74

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.437291Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation b7c57564-bfa8-40c4-a9eb-b1c323735f24 · outbound

This paper cites Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters

Reference 75

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.031774Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab500c6e-b0ee-429e-a6cc-3c8966cf4871 · outbound

This paper cites Beyond the imitation game: Quantifying and extrapolating the capabilities of language models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Beyond the imitation game: Quantifying and extrapolating the capabilities of language models

Reference 76

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.419525Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 26a87e2a-968c-4121-9028-452a777bf42d · outbound

This paper cites ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference

Reference 77

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.050044Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05320c79-bdd3-476b-bc40-cd0afe5fecd7 · outbound

This paper cites Triforce: Lossless acceleration of long sequence generation with hierarchical speculative decoding.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Triforce: Lossless acceleration of long sequence generation with hierarchical speculative decoding

Reference 78

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.388689Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation fb10823b-db14-4a4f-a061-1da966191dec · outbound

This paper cites Retentive Network: A Successor to Transformer for Large Language Models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Retentive Network: A Successor to Transformer for Large Language Models

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.064028Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6adc473c-32e0-46bf-9f20-d770813298b6 · outbound

This paper cites You only cache once: Decoder-decoder architectures for language models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods You only cache once: Decoder-decoder architectures for language models

Reference 80

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.364731Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 569071c9-4b33-432f-900c-ea4174e6cbd3 · outbound

This paper cites QUEST : Query-aware sparsity for efficient long-context LLM inference.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods QUEST : Query-aware sparsity for efficient long-context LLM inference

Reference 81

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.080839Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bdfa4aaa-db5f-4930-88ae-08d1bfb603ac · outbound

This paper cites Codestral mamba, 2024.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Codestral mamba, 2024

Reference 82

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.312561Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation c5483b3e-6342-483b-85f1-8874d6af226b · outbound

This paper cites Qwen2.5: A party of foundation models, 2024.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Qwen2.5: A party of foundation models, 2024

Reference 83

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.285181Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 6686e7be-bda2-46e4-8e3c-7538d5f4cd44 · outbound

This paper cites an unresolved cited work.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Unresolved cited work

Reference 84

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.104081Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d2efaf0b-cc4c-46aa-893e-ef2e967eb006 · outbound

This paper cites Automatic prefix caching.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Automatic prefix caching

Reference 85

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.254447Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 2dece6c6-3f32-4f9d-8fa0-3a181e4607b1 · outbound

This paper cites Michelangelo: Long Context Evaluations Beyond Haystacks via Latent Structure Queries.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Michelangelo: Long Context Evaluations Beyond Haystacks via Latent Structure Queries

Reference 86

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.121591Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4524286c-7793-4a71-b162-50af9049f9d4 · outbound

This paper cites An Empirical Study of Mamba-based Language Models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods An Empirical Study of Mamba-based Language Models

Reference 87

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.129767Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9d6a03d-a463-4065-aa62-e37f107167b5 · outbound

This paper cites MINT : Evaluating LLM s in multi-turn interaction with tools and language feedback.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods MINT : Evaluating LLM s in multi-turn interaction with tools and language feedback

Reference 88

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.234551Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 96ffaab2-c48c-4ab8-af55-f441d82bd9cf · outbound

This paper cites Efficient streaming language models with attention sinks.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Efficient streaming language models with attention sinks

Reference 89

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.145396Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4fe1dbbf-88fe-4b1f-8e35-874bba0b8c44 · outbound

This paper cites Duoattention: Efficient long-context LLM inference with retrieval and streaming heads.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Duoattention: Efficient long-context LLM inference with retrieval and streaming heads

Reference 90

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.178682Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation ff884504-f70b-4e81-b39e-a6ec9928ad9b · outbound

This paper cites LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention

Reference 91

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.161696Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 353866e5-2091-459b-b15f-85bdcc7a0be0 · outbound

This paper cites CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion

Reference 92

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.173124Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 68bef232-e66d-465d-97c5-ca891e8d482a · outbound

This paper cites Tree of thoughts: Deliberate problem solving with large language models.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Tree of thoughts: Deliberate problem solving with large language models

Reference 93

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.189873Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0f393cad-5ce7-45e6-a2d5-9b7e224cf771 · outbound

This paper cites Cascade inference: Memory bandwidth efficient shared prefix batch decoding, 2024.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Cascade inference: Memory bandwidth efficient shared prefix batch decoding, 2024

Reference 94

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.117891Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 9a1d06c3-cacf-4c4b-a352-842dc6d3a7de · outbound

This paper cites HELMET: How to Evaluate Long-Context Language Models Effectively and Thoroughly.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods HELMET: How to Evaluate Long-Context Language Models Effectively and Thoroughly

Reference 95

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.222150Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8468a9e5-3470-401c-9df6-18a13595b0e3 · outbound

This paper cites Dialogue-based relation extraction.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Dialogue-based relation extraction

Reference 96

Resolution: verified exact; doi, observed 2026-08-11T16:14:06.420317Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation fe36ebce-ce0a-402f-b777-c622d49dfa7b · outbound

This paper cites KV cache compression, but what must we give in return? a comprehensive benchmark of long context capable approaches.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods KV cache compression, but what must we give in return? a comprehensive benchmark of long context capable approaches

Reference 97

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.252705Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 11bb78d4-17cc-4a5f-97d9-7ff4be83815c · outbound

This paper cites Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention

Reference 98

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.258692Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 05cdc766-3668-4039-ab4d-41075a5500cc · outbound

This paper cites Reddi, and Sanjiv Kumar.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Reddi, and Sanjiv Kumar

Reference 99

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.081543Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 20c3e22f-3acc-4b9c-8b39-c99bea5ed461 · outbound

This paper cites Infinitebench: Extending long context evaluation beyond 100 K tokens.

SCBench: A KV Cache-Centric Analysis of Long-Context Methods Infinitebench: Extending long context evaluation beyond 100 K tokens

Reference 100

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T16:14:08.060930Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation f847b6cc-af3e-4ab0-ad3d-f1195fb45c2b · outbound

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

SCBench: A KV Cache-Centric Analysis of Long-Context Methods H2o: Heavy-hitter oracle for efficient generative inference of large language models

Reference 101

Resolution: unresolved; no resolver link, observed 2026-08-11T16:14:06.284069Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 7b080277-58e4-4e45-8560-57bac9655726 · inbound

A Survey on Large Language Model Acceleration based on KV Cache Management cites this paper.

A Survey on Large Language Model Acceleration based on KV Cache Management SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 57

Resolution: unresolved; no resolver link, observed 2026-08-11T00:38:46.746093Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ef36e9a9-22d8-4947-a948-1d7fb234a758 · inbound

Cache Me If You Must: Adaptive Key-Value Quantization for Large Language Models cites this paper.

Cache Me If You Must: Adaptive Key-Value Quantization for Large Language Models SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-09T20:34:01.265626Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 73f6391c-4166-407d-973e-5850a203fe1f · inbound

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation cites this paper.

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-09T11:11:17.722512Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 59b9fc34-209f-4705-a7f5-c96c010a895c · inbound

A Survey of Scaling in Large Language Model Reasoning cites this paper.

A Survey of Scaling in Large Language Model Reasoning SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 101

Resolution: verified exact; arxiv_id, observed 2026-05-22T21:22:09.420903Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 1d7ae517-d409-4357-b5c4-6d3b18cdd479 · inbound

SeerAttention-R: Sparse Attention Adaptation for Long Reasoning cites this paper.

SeerAttention-R: Sparse Attention Adaptation for Long Reasoning SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-07T05:06:32.642988Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e5d9f31e-b613-4a3c-8b4c-a3c8fbe621df · inbound

Lag-Relative Sparse Attention In Long Context Training cites this paper.

Lag-Relative Sparse Attention In Long Context Training SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-07T04:10:46.759948Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a7cc8d2f-75ec-4ffa-8fc0-5f5277e9658c · inbound

ReasonCache: Accelerating Large Reasoning Model Serving through KV Cache Sharing cites this paper.

ReasonCache: Accelerating Large Reasoning Model Serving through KV Cache Sharing SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-05-19T03:17:00.821916Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 6805502c-372d-4a5c-9805-371e773c449f · inbound

IceCache: Memory-efficient KV-cache Management for Long-Sequence LLMs cites this paper.

IceCache: Memory-efficient KV-cache Management for Long-Sequence LLMs SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 7

Resolution: metadata mismatch; arxiv_id, observed 2026-05-11T09:51:01.223918Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation dd591b7a-ea57-429c-adbd-4eb123ac75b4 · inbound

Accuracy Is Speed: Towards Long-Context-Aware Routing for Distributed LLM Serving cites this paper.

Accuracy Is Speed: Towards Long-Context-Aware Routing for Distributed LLM Serving SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-05-10T08:17:37.454956Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 2cefd318-562d-4ea8-802c-06367d0fedfc · inbound

PermaFrost-Attack: Stealth Pretraining Seeding(SPS) for planting Logic Landmines During LLM Training cites this paper.

PermaFrost-Attack: Stealth Pretraining Seeding(SPS) for planting Logic Landmines During LLM Training SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 120

Resolution: verified exact; arxiv_id, observed 2026-05-09T21:43:30.254592Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation ab6cb96e-8163-4cda-b3d5-5d1bab322634 · inbound

PermaFrost-Attack: Stealth Pretraining Seeding(SPS) for planting Logic Landmines During LLM Training cites this paper.

PermaFrost-Attack: Stealth Pretraining Seeding(SPS) for planting Logic Landmines During LLM Training SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 131

Resolution: metadata mismatch; arxiv_id, observed 2026-05-11T14:31:06.204775Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation eb610b38-d85c-4f9d-b2ec-fe3f114e1d93 · inbound

Rethinking KV Cache Eviction via a Unified Information-Theoretic Objective cites this paper.

Rethinking KV Cache Eviction via a Unified Information-Theoretic Objective SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-05-09T01:54:34.022760Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation a5420cf9-fe87-4123-bd0a-133290fc9353 · inbound

Tangram: Unlocking Non-Uniform KV Cache for Efficient Multi-turn LLM Serving cites this paper.

Tangram: Unlocking Non-Uniform KV Cache for Efficient Multi-turn LLM Serving SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 26

Resolution: verified exact; arxiv_id, observed 2026-07-02T12:06:56.130137Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 5f1e4339-8f7e-41f6-b85b-7663bea7bd5b · inbound

End-to-End Context Compression at Scale cites this paper.

End-to-End Context Compression at Scale SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 48

Resolution: metadata mismatch; arxiv_id, observed 2026-07-03T01:17:31.591780Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation 200ced63-88ec-4fe5-803b-61f3175d24dd · inbound

What to Keep, What to Forget: A Rate--Distortion View of Memory Compaction in LLMs and Agents cites this paper.

What to Keep, What to Forget: A Rate--Distortion View of Memory Compaction in LLMs and Agents SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 65

Resolution: verified exact; local_arxiv, observed 2026-07-10T01:36:44.018523Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.

Observation e4281599-93db-4aae-92a1-5122263b59ed · inbound

Understanding Sparse Attention Selectivity in Long-Context Foundation Models via Counterfactual Evaluation cites this paper.

Understanding Sparse Attention Selectivity in Long-Context Foundation Models via Counterfactual Evaluation SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-04T23:14:59.988064Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e9ba8123-f1c0-410d-9398-a20799742211 · inbound

PhyCheck: Fine-Grained Evidence-Grounded Dataset for Physical Law Understanding in Video-LLMs cites this paper.

PhyCheck: Fine-Grained Evidence-Grounded Dataset for Physical Law Understanding in Video-LLMs SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-04T13:43:50.964348Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f0409599-70a5-44b9-9c7f-193024812f5b · inbound

PhyCheck: Fine-Grained Evidence-Grounded Dataset for Physical Law Understanding in Video-LLMs cites this paper.

PhyCheck: Fine-Grained Evidence-Grounded Dataset for Physical Law Understanding in Video-LLMs SCBench: A KV Cache-Centric Analysis of Long-Context Methods

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-07T00:11:48.319572Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.