Paper Citation Record · LEDGER

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation

As of 22 August 2026, this Paper Citation Record lists 19 of 19 outbound references and 0 inbound Pith citation observations for arXiv:2607.16213.

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

pith.paper-citation-record.v1
2607.16213 v1

Coverage vector

measured 19 of 19 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-02T13:41:14.943191Z

measured 19 of 19 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-22T06:32:14.747728+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

19 of 19 outbound references displayed

verified exact0
verified fuzzy0
unresolved19
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 4ddb8403-48ff-4dda-b0d1-affb8e4e8467 · outbound

This paper cites Longbench: A bilingual, multitask benchmark for long context understanding.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Longbench: A bilingual, multitask benchmark for long context understanding

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:13.209747Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 13086b78-df85-4b18-b8c0-cdfc426cae45 · outbound

This paper cites Fu, et al.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Fu, et al

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:13.302600Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 94ecb4e7-2f3a-4445-872f-9e74d2363ccb · outbound

This paper cites Pyramidkv: Dynamic kv cache compression based on pyramidal information funneling.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Pyramidkv: Dynamic kv cache compression based on pyramidal information funneling

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:13.470701Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c76fa2f0-3653-4940-bcd3-03591c473169 · outbound

This paper cites Diffrate: Differentiable compression rate for efficient vision transformers.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Diffrate: Differentiable compression rate for efficient vision transformers

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:13.608935Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3712b03b-74e6-4869-b3d6-d7d003ba1d16 · outbound

This paper cites Attention score is not all you need for token importance indicator in kv cache reduction: Value also matters.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Attention score is not all you need for token importance indicator in kv cache reduction: Value also matters

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:13.771791Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 94f232a7-fa34-4b92-b924-0e05aac6e87c · outbound

This paper cites Zipcache: Accurate and efficient kv cache quantization with salient token identification.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Zipcache: Accurate and efficient kv cache quantization with salient token identification

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:13.884121Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 96179e24-68ba-4cd2-addd-0b260f49312c · outbound

This paper cites Snapkv: Llm knows what you are looking for before generation.Advances in Neural Information Processing Systems, 37:22947–22970, 2024.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Snapkv: Llm knows what you are looking for before generation.Advances in Neural Information Processing Systems, 37:22947–22970, 2024

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:13.982069Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dbdda27f-c85c-4294-a543-b616f3e26e70 · outbound

This paper cites Chunkkv: Semantic-preserving kv cache compression for efficient long-context llm inference.arXiv preprint arXiv:2502.00299, 2025.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Chunkkv: Semantic-preserving kv cache compression for efficient long-context llm inference.arXiv preprint arXiv:2502.00299, 2025

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.105777Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 07a117ec-92ba-48bf-8993-4bb8b4a2be53 · outbound

This paper cites Kivi: A tuning-free asymmetric 2bit quantization for kv cache.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Kivi: A tuning-free asymmetric 2bit quantization for kv cache

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.217911Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 821376e1-de73-44f6-ab69-3209c680b971 · outbound

This paper cites Lava: Layer-wise kv cache eviction with dynamic budget allocation.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Lava: Layer-wise kv cache eviction with dynamic budget allocation

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.369320Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5767be33-4450-447c-a244-ddfd3e54a60a · outbound

This paper cites Keepkv: Eliminating output perturbation in kv cache compression for efficient llms inference.arXiv, 2024.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Keepkv: Eliminating output perturbation in kv cache compression for efficient llms inference.arXiv, 2024

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.533602Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0aa3385f-b70b-4f5b-badb-a5172cf1cd28 · outbound

This paper cites Look-m: Look-once optimization in kv cache for efficient multimodal long-context inference.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Look-m: Look-once optimization in kv cache for efficient multimodal long-context inference

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.559159Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1fa5edbe-e771-4ea0-a968-15a03391fdd9 · outbound

This paper cites D2o: Dynamic discriminative operations for efficient long-context inference of large language models.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation D2o: Dynamic discriminative operations for efficient long-context inference of large language models

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.563792Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20f4bf9f-c2c4-407c-b717-ddafc6fdf7b1 · outbound

This paper cites Model Tells You Where to Merge: Adaptive KV Cache Merging for LLMs on Long-Context Tasks.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Model Tells You Where to Merge: Adaptive KV Cache Merging for LLMs on Long-Context Tasks

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.592412Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1248b30e-01d2-4d69-a5db-8cca5f56bc93 · outbound

This paper cites Efficient streaming language models with attention sinks.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Efficient streaming language models with attention sinks

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.642097Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f8894808-f39f-4332-837d-5881805375b9 · outbound

This paper cites Evolkv: Evolutionary kv cache compression for llm inference.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Evolkv: Evolutionary kv cache compression for llm inference

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.721932Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a3e3234b-415f-42da-a7a9-a8ec76d88ee5 · outbound

This paper cites Weightedkv: Attention scores weighted key-value cache merging for large language models.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Weightedkv: Attention scores weighted key-value cache merging for large language models

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.801772Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 43bcf879-3278-4f13-8c9d-b4c3f36ac0f4 · outbound

This paper cites Ems: Adaptive evict-then-merge strategy for head-wise kv cache compression.arXiv, 2024.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation Ems: Adaptive evict-then-merge strategy for head-wise kv cache compression.arXiv, 2024

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.886026Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 485124aa-59af-458f-9d4e-be4efd3c0ad7 · outbound

This paper cites How is the ground truth for fake news established?.

SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation How is the ground truth for fake news established?

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-02T13:41:14.943191Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.