Pith. sign in

Paper Citation Record · LEDGER

The risk of KV cache compression

As of 14 August 2026, this Paper Citation Record lists 27 of 27 outbound references and 1 inbound Pith citation observation for arXiv:2607.01520.

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

pith.paper-citation-record.v1
2607.01520 v1

Coverage vector

measured 27 of 27 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-07-03T20:53:44.487336Z

measured 28 of 28 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-14T06:32:32.682623+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-03T00:43:51.470200Z

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

27 of 27 outbound references displayed

  • verified exact4
  • verified fuzzy19
  • unresolved2
  • parse uncertain0
  • malformed identifier1
  • metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 578a657a-1bf4-44f9-aaf8-d5402241a5db · outbound

This paper cites Keyformer: KV cache reduction through key tokens selection for efficient generative inference.Proceedings of Machine Learning and Systems, 6:114–127, 2024.

The risk of KV cache compression Keyformer: KV cache reduction through key tokens selection for efficient generative inference.Proceedings of Machine Learning and Systems, 6:114–127, 2024

Reference 1

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.822137Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:2817039d2af9bb762aac1a1e0f291bd946998a8a7ab9b20fe1dfc6cba2e9cb89

Observation 51bb7488-6964-4ef3-996b-5b19e5155973 · outbound

This paper cites Support basis: Fast attention beyond bounded entries.

The risk of KV cache compression Support basis: Fast attention beyond bounded entries

Reference 2

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.839910Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:51bb6478a9b08fb4bd6d36f3ee498a75ad14fac8ae2b942fb60121f397a9fb10

Observation aead27ce-3fdb-4e1a-9145-2ddd2a68d0cf · outbound

This paper cites Fast attention requires bounded entries.Advances in Neural Information Processing Systems, 36:63117–63135, 2023.

The risk of KV cache compression Fast attention requires bounded entries.Advances in Neural Information Processing Systems, 36:63117–63135, 2023

Reference 3

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.848301Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:a4e4c2b278663e1f5e775a8cb81adb2b5fcc9f384b44dffdcf6bf95e28b136ab

Observation d536ce97-039e-4a41-997d-974242cccb7f · outbound

This paper cites Effective gene expression prediction from sequence by integrating long-range interactions.

The risk of KV cache compression Effective gene expression prediction from sequence by integrating long-range interactions

Reference 4

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.870120Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:630aa514769624ad5dc68f08dcba2b1b3865db1c7475b153d8cbfb3b0bd1d433

Observation c0f40314-6dca-4961-ad10-c511cb54bd71 · outbound

This paper cites wav2vec 2.0: A frame- work for self-supervised learning of speech representations.Advances in Neural Information Processing Systems, 33:12449–12460, 2020.

The risk of KV cache compression wav2vec 2.0: A frame- work for self-supervised learning of speech representations.Advances in Neural Information Processing Systems, 33:12449–12460, 2020

Reference 5

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.830450Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:45876e357501eec45afdef5289b0301c66dffe03ccc4de00c3fc084f1c2cff47

Observation dd0faa31-a4ff-4e92-abe4-b546675d8cee · outbound

This paper cites LongBench v2: Towards deeper understanding and reason- ing on realistic long-context multitasks.

The risk of KV cache compression LongBench v2: Towards deeper understanding and reason- ing on realistic long-context multitasks

Reference 6

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.842466Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:ea92e9f39d600a5e0e5351208f5aace11fa9902e3bb7c3bfc39ff92029134ca1

Observation 5142cba2-91ca-430f-8df8-18cd72207bac · outbound

This paper cites On a generalization of iterated and randomized rounding.Theory of Computing, 20(6):1–23, 2024.

The risk of KV cache compression On a generalization of iterated and randomized rounding.Theory of Computing, 20(6):1–23, 2024

Reference 7

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.837235Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:bff74fef10add3e60fd45dd7ba9ca368d5869a692d050d8502c09f154849fdb0

Observation 048142dd-0721-42e7-91d5-89333ea03a17 · outbound

This paper cites Language models are few-shot learners.Advances in Neural Information Processing Systems, 33:1877–1901.

The risk of KV cache compression Language models are few-shot learners.Advances in Neural Information Processing Systems, 33:1877–1901

Reference 8

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.835008Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:dd10ede0a679eac82c7d9352cd7969d3771d19390a199ecf2af6f3ae3bf8647e

Observation 061afa8e-aa2b-4d4d-8510-8fcd1d72339f · outbound

This paper cites an unresolved cited work.

The risk of KV cache compression Unresolved cited work

Reference 9

Resolution
unresolved
raw_fallback, observed 2026-07-05T01:40:34.819641Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:9ebac90b78bb88d0269310af73322104159f8ee95e8059ed4a0446efd9acd414

Observation dfcc7161-af29-4961-aed2-91c42007df65 · outbound

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

The risk of KV cache compression PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling

Reference 10

Resolution
verified exact
local_arxiv, observed 2026-07-03T20:58:57.304826Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:031c7d62cb34b93638ea39dda96b8f910bba91234297cd80a3eb26620929467d

Observation a197b57b-0434-4694-a800-232407301a4e · outbound

This paper cites Low-rank thinning.

The risk of KV cache compression Low-rank thinning

Reference 11

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.832559Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:f81fb389731945df81afde0ef784a805a740f245249cf2236b115398080d3f44

Observation f4e70011-78c6-471e-97a0-cbce841802a2 · outbound

This paper cites Super-samples from kernel herding.

The risk of KV cache compression Super-samples from kernel herding

Reference 12

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.859496Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:04cfa64d03907b5f49195a23424b9439252859d611e7fb640bec27e4551e7432

Observation 9c60fcfb-1806-4133-9fec-da92e7f39515 · outbound

This paper cites Transformer-xl: Attentive language models beyond a fixed-length context.

The risk of KV cache compression Transformer-xl: Attentive language models beyond a fixed-length context

Reference 13

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.834764Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:cc75da6abf1b324d422aab7508be16397d17d62341e083249acb1048110cea22

Observation d27bed6d-a67f-4966-8289-4c89b9d4ae2e · outbound

This paper cites BERT: Pre-training of deep bidirectional transformers for language understanding.

The risk of KV cache compression BERT: Pre-training of deep bidirectional transformers for language understanding

Reference 14

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.845809Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:1a2a663ff86a46e038be390af03e15c6a54b7d345dcd61eb06c4fbd27914dad3

Observation d17eb2a1-185d-42ec-84cf-2f24e3c3ebf9 · outbound

This paper cites A mathematical perspective on transformers.Bulletin of the American Mathematical Society, 62(3):427–479.

The risk of KV cache compression A mathematical perspective on transformers.Bulletin of the American Mathematical Society, 62(3):427–479

Reference 15

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.865328Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:b28acb4bfea137a4f9b690016b0ae3bc0cdde70f3ba09899b972bc02633e77ca

Observation ccd80069-5d95-4ada-8a30-80c86ac588f2 · outbound

This paper cites Compression Barriers for Autoregressive Transformers.

The risk of KV cache compression Compression Barriers for Autoregressive Transformers

Reference 16

Resolution
metadata mismatch
arxiv_id, observed 2026-07-03T20:58:57.292372Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:0319be505597f5bda728f493b6899273927a27f8785cf90e3017ad63fb412037

Observation 9a87ce7d-41d1-4cb6-9aec-3fbac7ab3b5d · outbound

This paper cites Revisiting frank-wolfe: Projection-free sparse convex optimization.

The risk of KV cache compression Revisiting frank-wolfe: Projection-free sparse convex optimization

Reference 17

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.867446Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:df80c21f73f4d7e3922a7a624ef33fa09201ae1c7e2717c3eed1f43a7cf2c165

Observation 3f5d7848-8c90-49d4-a047-2337750db863 · outbound

This paper cites Balancekv: Kv cache compression through discrepancy theory.

The risk of KV cache compression Balancekv: Kv cache compression through discrepancy theory

Reference 18

Resolution
verified exact
arxiv_id, observed 2026-07-03T20:58:57.301648Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:26ad03afb1438fa7ebad4534147d84f0ca217fc0ac15986ae6da4cce83b68f47

Observation 912cbfd2-6972-4e78-b60e-e1ded5a7707d · outbound

This paper cites SnapKV: LLM knows what you are looking for before generation.Advances in Neural Information Processing Systems, 37:22947–22970.

The risk of KV cache compression SnapKV: LLM knows what you are looking for before generation.Advances in Neural Information Processing Systems, 37:22947–22970

Reference 19

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.863191Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:a5056c4e53bfedc1ae71a276502be381b49523e25fea73b01a66a5dd13a6d9e2

Observation ad841e48-b221-415e-bf2c-98af9e3db4fc · outbound

This paper cites ScissorHands: Exploiting the persistence of importance hypothesis for LLM KV cache compression at test time.Advances in Neural Information Processing Systems, 36:52342–52364, 2023.

The risk of KV cache compression ScissorHands: Exploiting the persistence of importance hypothesis for LLM KV cache compression at test time.Advances in Neural Information Processing Systems, 36:52342–52364, 2023

Reference 20

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.852585Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:d7a801506f124311ba875bf5a2768dad90fcafc51ee7dca1f3a69d64f8ca9708

Observation d453dfd0-5a11-4164-91ec-6fcd21f190c7 · outbound

This paper cites YaRN: Efficient context window extension of large language models.

The risk of KV cache compression YaRN: Efficient context window extension of large language models

Reference 21

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.850958Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:5c1dcc2e6b42ef4aac5b93d043f34586ddaf32a0911939ab0cdfd36089d81cf0

Observation 1435f0d4-28ed-4752-af92-87cf0c3e6563 · outbound

This paper cites an unresolved cited work.

The risk of KV cache compression Unresolved cited work

Reference 22

Resolution
unresolved
raw_fallback, observed 2026-07-05T01:40:34.858178Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:669f55c7ca986e9bc07716bd051a2ca46b7aa0d4d674e577b1e75489d1e7236f

Observation 659d8172-708a-4bc4-8576-99f4f4362462 · outbound

This paper cites Attention is all you need.Advances in Neural Information Processing Systems, 30.

The risk of KV cache compression Attention is all you need.Advances in Neural Information Processing Systems, 30

Reference 23

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.861005Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:c040edb6de234cf797eac3b7fbcdeef451d4cdc621f36199b5cf28b38dca9ed7

Observation 9fd9baa8-16c7-4976-9b90-cfd44c6dfe53 · outbound

This paper cites Efficient streaming language models with attention sinks.

The risk of KV cache compression Efficient streaming language models with attention sinks

Reference 24

Resolution
verified fuzzy
raw_fallback, observed 2026-07-05T01:40:34.872612Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:9322b666b0ced40d0f1da5a70e9733c6afd4e1f874fb2d049be7ea6f4c5e0bcb

Observation 2ee25660-92da-4d64-8e8c-9adff4e93186 · outbound

This paper cites Qwen3 Technical Report.

The risk of KV cache compression Qwen3 Technical Report

Reference 25

Resolution
verified exact
local_arxiv, observed 2026-07-03T20:58:57.298418Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:54d79d30c58baf17a803733cec199d575c452b104f015ac71dc401384edd7501

Observation 0dd53361-4fe6-4d8c-a761-b94c69ca0f63 · outbound

This paper cites SubGen: Token Generation in Sublinear Time and Memory.

The risk of KV cache compression SubGen: Token Generation in Sublinear Time and Memory

Reference 26

Resolution
verified exact
arxiv_id, observed 2026-07-03T20:58:57.295575Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:0ec2cabe0af0bac6216db7eede10a0c37bb210cac80656ce99740578f5272548

Observation ced9066c-a915-4e36-b9fb-2c477b504cda · outbound

This paper cites alive set.

The risk of KV cache compression alive set

Reference 27

Resolution
malformed identifier
raw_fallback, observed 2026-07-05T01:40:34.853400Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.

source=pdf_text observed=2026-07-03T20:53:44.487336Z digest=sha256:274495fd2b8d9a4f2b59cfc7b997a3050fb10089e96bbdd77bde571ef1fb97d9

Pith citing papers

Observation ffda679e-359e-4a8e-891c-a3ae5763e150 · inbound

WitCert: Sound Runtime Risk Observability and Gating for KV-Cache Quantization cites this paper.

WitCert: Sound Runtime Risk Observability and Gating for KV-Cache Quantization The risk of KV cache compression

Reference 32

Resolution
unresolved
no resolver link, observed 2026-08-03T00:43:51.470200Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

source=pdf_text observed=2026-08-03T00:43:51.470200Z digest=sha256:122b5cba214220c712f2b3aa4b681551404d55d62a1d9a2a1dc4087343391c2a