Paper Citation Record · LEDGER

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving

As of 22 July 2026, this Paper Citation Record lists 24 of 24 outbound references and 4 inbound Pith citation observations for arXiv:2604.19157.

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

pith.paper-citation-record.v1
2604.19157 v1

Coverage vector

measured 24 of 24 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-10T03:09:51.453839Z

measured 28 of 28 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-07-21T06:31:05.380196+00:00

measured 4 of 4 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-07-01T07:06:53.318182Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-07-01T08:55:35.168254Z

Reference resolution

24 of 24 outbound references displayed

verified exact8
verified fuzzy6
unresolved0
parse uncertain0
malformed identifier2
metadata mismatch8

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 1fc4751d-1b7d-4c8a-8a95-2813420122f9 · outbound

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

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Gqa: Training generalized multi-query transformer models from multi-head checkpoints

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-05-22T18:05:01.676258Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation f460f266-c1ff-4e33-ab90-f3182d1b90e5 · outbound

This paper cites Small Language Models are the Future of Agentic AI.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Small Language Models are the Future of Agentic AI

Reference 2

Resolution: metadata mismatch; arxiv_id, observed 2026-05-16T11:55:51.260368Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c4b094ad-1f6e-4d08-acbc-fe61eb2c4a6f · outbound

This paper cites Evaluating Large Language Models Trained on Code.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Evaluating Large Language Models Trained on Code

Reference 3

Resolution: metadata mismatch; local_arxiv, observed 2026-05-10T03:14:08.305431Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation a7bacdf5-d0eb-4f28-ad29-f829ca98612c · outbound

This paper cites Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-05-10T03:14:08.300321Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 0dbeb0ac-ca5b-4fbc-8c29-b34318db1bef · outbound

This paper cites Extreme Compression of Large Language Models via Additive Quantization.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Extreme Compression of Large Language Models via Additive Quantization

Reference 5

Resolution: metadata mismatch; arxiv_id, observed 2026-05-10T03:14:08.292383Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation ef201b2f-224a-428c-9713-7d9d694b6f01 · outbound

This paper cites GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers

Reference 6

Resolution: verified exact; arxiv_id, observed 2026-05-10T17:18:35.295691Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation fefe4adf-a0c3-466a-8ea5-efce1be0a46d · outbound

This paper cites Measuring Mathematical Problem Solving With the MATH Dataset.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Measuring Mathematical Problem Solving With the MATH Dataset

Reference 7

Resolution: verified exact; arxiv_id, observed 2026-05-10T13:00:39.484061Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 6b836c25-2ba9-4800-ab23-4d090d932d1e · outbound

This paper cites LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code

Reference 8

Resolution: metadata mismatch; arxiv_id, observed 2026-05-10T17:34:43.229714Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation aea58b42-45b8-479f-9d0f-0ffb9a29cf81 · outbound

This paper cites Challenges and Applications of Large Language Models.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Challenges and Applications of Large Language Models

Reference 9

Resolution: verified exact; arxiv_id, observed 2026-05-10T03:14:08.308026Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 8196466f-24c9-400c-b563-0b79058b682c · outbound

This paper cites GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-05-10T03:14:08.289804Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation b2271486-0af3-487c-b0b0-404416548090 · outbound

This paper cites CommVQ: Commutative Vector Quantization for KV Cache Compression.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving CommVQ: Commutative Vector Quantization for KV Cache Compression

Reference 11

Resolution: metadata mismatch; arxiv_id, observed 2026-05-10T03:14:08.310691Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation a269f207-fc79-4bc9-9885-ea6962de1d77 · outbound

This paper cites an unresolved cited work.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Unresolved cited work

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-05-22T18:05:01.668836Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 02801590-f8a8-4050-a820-085c98a8088d · outbound

This paper cites ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-05-10T03:14:08.320530Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 428a8b2e-970b-4868-87d5-f10b13a74b04 · outbound

This paper cites Eigen attention: Attention in low-rank space for KV cache compression.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Eigen attention: Attention in low-rank space for KV cache compression

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-05-22T18:05:01.672591Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 8bdd13dc-c879-463a-aa79-1e6dd9d912e4 · outbound

This paper cites doi: 10.18653/v1/2024.findings-emnlp.899.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving doi: 10.18653/v1/2024.findings-emnlp.899

Reference 15

Resolution: verified exact; doi, observed 2026-05-10T03:14:07.459764Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 276d13e8-a14d-4186-8b71-1d60016fb61b · outbound

This paper cites Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-05-22T18:05:01.674363Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation a3927740-fa61-46d0-ad94-a19809036269 · outbound

This paper cites Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization

Reference 17

Resolution: verified exact; local_arxiv, observed 2026-05-10T03:14:08.323070Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 3a621870-3670-40ae-9e51-78081747bd6b · outbound

This paper cites Accurate and efficient 2-bit kv cache quantization with dynamic channel-wise precision boost.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Accurate and efficient 2-bit kv cache quantization with dynamic channel-wise precision boost

Reference 18

Resolution: metadata mismatch; arxiv_id, observed 2026-05-10T03:14:08.313255Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 57c7eb06-e0a0-472b-97f3-d188dbf1ce25 · outbound

This paper cites Qwen3 Technical Report.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Qwen3 Technical Report

Reference 19

Resolution: metadata mismatch; local_arxiv, observed 2026-05-10T03:14:08.315559Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation f87840a7-2d24-4dce-933d-851e85e5026e · outbound

This paper cites TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate

Reference 20

Resolution: metadata mismatch; arxiv_id, observed 2026-05-20T08:09:22.519591Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 8ad863f4-fc04-4a61-b2af-7a9521f986ee · outbound

This paper cites 13 Preprint.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving 13 Preprint

Reference 21

Resolution: malformed identifier; raw_fallback, observed 2026-05-22T18:05:01.665105Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation a90c3243-2111-423a-9a76-6edfb4764912 · outbound

This paper cites (K)” and “(K&V).

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving (K)” and “(K&V)

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-05-22T18:05:01.670806Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation c145495f-f2e2-43b1-b9c9-1bd97862e92a · outbound

This paper cites These two metrics can diverge significantly under memory-pressure conditions, as we explain below.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving These two metrics can diverge significantly under memory-pressure conditions, as we explain below

Reference 23

Resolution: malformed identifier; raw_fallback, observed 2026-05-22T18:05:01.678233Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 3236f2e8-e37f-4425-b44a-fe3e6ab5a63d · outbound

This paper cites Under review.

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving Under review

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-05-22T18:05:01.666830Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Pith citing papers

Observation 3264530b-d55c-46b6-9809-a799a5cfc4d5 · inbound

TriAxialKV: Toward Extreme Low-Precision KV-Cache Quantization for Agentic Inference Tasks cites this paper.

TriAxialKV: Toward Extreme Low-Precision KV-Cache Quantization for Agentic Inference Tasks SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving

Reference 20

Resolution: verified exact; local_arxiv, observed 2026-05-20T14:33:21.298670Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 5701b047-9b3a-4cac-af2f-11ccf86a9784 · inbound

OSCAR: Offline Spectral Covariance-Aware Rotation for 2-bit KV Cache Quantization cites this paper.

OSCAR: Offline Spectral Covariance-Aware Rotation for 2-bit KV Cache Quantization SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving

Reference 12

Resolution: verified exact; local_arxiv, observed 2026-05-20T12:18:16.555876Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 34b8ac7a-a67f-4d1c-a3bb-fd0fd08793b6 · inbound

Demystifying the Design Space and Best Practices for Heterogeneous LLM Inference and Serving cites this paper.

Demystifying the Design Space and Best Practices for Heterogeneous LLM Inference and Serving SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving

Reference 28

Resolution: verified exact; local_arxiv, observed 2026-06-30T14:04:45.295164Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.

Observation 5b1ca8d8-5672-4340-9fe7-fccf202243d0 · inbound

Demystifying the Design Space and Best Practices for Heterogeneous LLM Inference and Serving cites this paper.

Demystifying the Design Space and Best Practices for Heterogeneous LLM Inference and Serving SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving

Reference 28

Resolution: verified exact; local_arxiv, observed 2026-07-01T08:55:35.169438Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-07-21T06:31:05.380196+00:00.