Paper Citation Record · LEDGER

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression

As of 11 August 2026, this Paper Citation Record lists 83 of 83 outbound references and 5 inbound Pith citation observations for arXiv:2502.01941.

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

pith.paper-citation-record.v1
2502.01941 v4

Coverage vector

measured 83 of 83 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-23T04:15:36.906263Z

measured 88 of 88 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-10T06:31:04.303077+00:00

measured 5 of 5 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-07T00:11:49.083358Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-05-18T11:32:35.182969Z

Reference resolution

83 of 83 outbound references displayed

verified exact47
verified fuzzy29
unresolved1
parse uncertain0
malformed identifier3
metadata mismatch3

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 7b5438cc-5188-451a-b320-77221f6971c5 · outbound

This paper cites an unresolved cited work.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Unresolved cited work

Reference 1

Resolution: unresolved; raw_fallback, observed 2026-05-23T04:17:32.143515Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation b077ab99-2696-402b-b9ad-82785e5dd459 · outbound

This paper cites Language models are few-shot learners.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Language models are few-shot learners

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.139640Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 77165aeb-cbc2-4b4b-b2ae-03f002129989 · outbound

This paper cites PaLM: Scaling Language Modeling with Pathways.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression PaLM: Scaling Language Modeling with Pathways

Reference 3

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.261128Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation b1b0f01c-7880-4fb2-957d-e53ff84d6f29 · outbound

This paper cites UL2: Unifying Language Learning Paradigms.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression UL2: Unifying Language Learning Paradigms

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.271731Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 5f6b275e-e184-4daa-a3fb-0eb005c5ca88 · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression LLaMA: Open and Efficient Foundation Language Models

Reference 5

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.287924Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 40a24a2e-f74e-4601-823c-b624eecf2b76 · outbound

This paper cites Llama 2: Open Foundation and Fine-Tuned Chat Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Llama 2: Open Foundation and Fine-Tuned Chat Models

Reference 6

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.086641Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 44d245d4-468b-4d5a-a0a0-098e2c607ae4 · outbound

This paper cites Flashattention: Fast and memory-efficient exact attention with io-awareness.Advances in Neural Information Processing Systems, 35:16344–16359.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Flashattention: Fast and memory-efficient exact attention with io-awareness.Advances in Neural Information Processing Systems, 35:16344–16359

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.131522Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 7b853ade-d56d-4056-bed6-482413615709 · outbound

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

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression FlashAttention-2: Faster attention with better parallelism and work partitioning

Reference 8

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.135536Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 0cca177b-4137-4514-828b-1d363660cb45 · outbound

This paper cites DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.073977Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 8c24fd32-6ce9-44a3-8621-1640f881ce0d · outbound

This paper cites Efficient streaming language models with attention sinks.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Efficient streaming language models with attention sinks

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.127419Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation fc1e02be-9b19-4383-9083-8ffa9ea61b13 · outbound

This paper cites Extending Context Window of Large Language Models via Positional Interpolation.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Extending Context Window of Large Language Models via Positional Interpolation

Reference 12

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.190945Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation b764decf-ed5a-4f8e-957a-1c577e4c796a · outbound

This paper cites Effective long-context scaling of foundation models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Effective long-context scaling of foundation models

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.122852Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation a333b96d-2939-4aac-b077-2e2d7ce75b62 · outbound

This paper cites URL https://aclanthology.org/2024.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression URL https://aclanthology.org/2024

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.178295Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 678278be-bf3c-445d-89ef-6419c3645431 · outbound

This paper cites Lon- glora: Efficient fine-tuning of long-context large language models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Lon- glora: Efficient fine-tuning of long-context large language models

Reference 15

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.190050Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 8c289ec1-4369-4c78-8433-33df3cfbfa3a · outbound

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

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression YaRN: Efficient context window extension of large language models

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.193666Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 65ff9b19-b360-43dc-99f6-7af8daeb1c89 · outbound

This paper cites Introducing jamba: Ai21’s groundbreaking ssm-transformer model.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Introducing jamba: Ai21’s groundbreaking ssm-transformer model

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.167256Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 156354e5-793e-42b3-8a4f-540d4736a3d2 · outbound

This paper cites Announcing grok-1.5.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Announcing grok-1.5

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.170548Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 16736013-d419-4ab3-970a-b158d3832721 · outbound

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

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

Reference 19

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.094340Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation e24b0851-9c1e-494e-b04c-d9a48f70c34f · outbound

This paper cites Introducing the next generation of claude.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Introducing the next generation of claude

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.159385Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 8ede3234-a2ce-46e0-b953-c9ac2ee95e63 · outbound

This paper cites Deepseek-v2: A strong, economical, and efficient mixture-of-experts language model.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Deepseek-v2: A strong, economical, and efficient mixture-of-experts language model

Reference 21

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.155384Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 5292182d-541b-4b05-ad58-94f2257098e7 · outbound

This paper cites DeepSeek-V3 Technical Report.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression DeepSeek-V3 Technical Report

Reference 22

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.146691Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation b56ed318-f89a-418f-9919-cd1f11175a3b · outbound

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

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression H2o: Heavy-hitter oracle for efficient generative inference of large language models

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.163630Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation f20e445f-a05a-4c33-827e-7eef40ee0ec0 · outbound

This paper cites SnapKV: LLM Knows What You are Looking for Before Generation.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression SnapKV: LLM Knows What You are Looking for Before Generation

Reference 24

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.266013Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 3e63bea4-b701-4874-ac4c-15cd9ef7dcb0 · outbound

This paper cites Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs

Reference 26

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.299460Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 073260d5-2cbc-4713-9d53-45eef363d853 · outbound

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

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling

Reference 27

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.250437Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 9086b944-b156-4ae5-a31b-348224200e75 · outbound

This paper cites LazyLLM: Dynamic token pruning for efficient long context LLM inference.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression LazyLLM: Dynamic token pruning for efficient long context LLM inference

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.182385Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 35e38757-05ef-4846-815e-4eba4dbf94ba · outbound

This paper cites PyramidInfer: Pyramid KV Cache Compression for High-throughput LLM Inference.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression PyramidInfer: Pyramid KV Cache Compression for High-throughput LLM Inference

Reference 29

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.141533Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 2c2e0035-369c-4a26-875a-fdef2da4008b · outbound

This paper cites Keyformer: Kv cache reduction through key tokens selection for efficient generative inference.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Keyformer: Kv cache reduction through key tokens selection for efficient generative inference

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.097602Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 3f3f716d-30a5-4488-ad69-7793989aea36 · outbound

This paper cites Scissorhands: Exploiting the persistence of impor- tance hypothesis for llm kv cache compression at test time.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Scissorhands: Exploiting the persistence of impor- tance hypothesis for llm kv cache compression at test time

Reference 31

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.072523Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 09281a72-4df8-41b0-8cf2-5bb3844dd782 · outbound

This paper cites Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference

Reference 33

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.054275Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 4db05ef1-a8fe-4218-bafb-d72580e2f6af · outbound

This paper cites LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding

Reference 34

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.123790Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation fc7823f8-f28e-4b38-997c-1d039f92ef9f · outbound

This paper cites LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks

Reference 35

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.206056Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 7983a355-9573-4ed5-acf8-2721eb1ef274 · outbound

This paper cites Needle In A Haystack - pressure testing LLMs.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Needle In A Haystack - pressure testing LLMs

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.118630Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation f4e994b8-cde8-4bdb-9ac4-4a13a0027cf0 · outbound

This paper cites Measuring Massive Multitask Language Understanding.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Measuring Massive Multitask Language Understanding

Reference 37

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.080161Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 4e00a9ae-3484-4586-a1f3-82e96ff5caf1 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Training Verifiers to Solve Math Word Problems

Reference 38

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.163386Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation e44249db-2ac5-4d6a-b2cd-3da0dc942d05 · outbound

This paper cites doi: 10.18653/v1/n19-1421.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression doi: 10.18653/v1/n19-1421

Reference 39

Resolution: metadata mismatch; doi, observed 2026-05-23T04:17:30.799284Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 39ba5651-68bc-451e-96fd-7e8e5c52fb68 · outbound

This paper cites Evaluating Large Language Models Trained on Code.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Evaluating Large Language Models Trained on Code

Reference 40

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.185490Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 08793890-fbb0-432d-a98f-17054d4bb3d3 · outbound

This paper cites Jailbreakv: A bench- mark for assessing the robustness of multimodal large language models against jailbreak attacks.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Jailbreakv: A bench- mark for assessing the robustness of multimodal large language models against jailbreak attacks

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.114002Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 5e9f0d06-ce18-47d1-89bc-2ae00c1817b8 · outbound

This paper cites LongGenBench: Long-context generation benchmark.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression LongGenBench: Long-context generation benchmark

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.110186Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 0552de9e-cabe-4c3e-bba8-1babc7dae067 · outbound

This paper cites The Llama 3 Herd of Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression The Llama 3 Herd of Models

Reference 43

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.217847Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 63a0d9b4-2ab5-404f-a4a5-a8c49f239c90 · outbound

This paper cites Mistral 7B.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Mistral 7B

Reference 44

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.168665Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T22:08:12.954417+00:00.

Observation efd852f4-5519-4f3b-86e5-9a19c3a810d1 · outbound

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

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 45

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.228380Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation a434a9d8-1738-4708-9d61-5a4d3acf6ae9 · outbound

This paper cites Chunkkv: Semantic-preserving kv cache compression for efficient long-context llm inference.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Chunkkv: Semantic-preserving kv cache compression for efficient long-context llm inference

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.147601Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 5ef1b7d3-eecf-4189-9b42-25c9599d41f7 · outbound

This paper cites Chunkkv: Semantic-preserving KV cache compression for efficient long-context LLM inference.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Chunkkv: Semantic-preserving KV cache compression for efficient long-context LLM inference

Reference 47

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.306776Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 984cd469-f043-48cf-a50b-b6518b0db3ee · outbound

This paper cites A framework for few-shot language model evaluation, 12 2023.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression A framework for few-shot language model evaluation, 12 2023

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.089816Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 50cf01a6-9e88-4c4c-881d-8a4e96c92be7 · outbound

This paper cites Many-Shot In-Context Learning.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Many-Shot In-Context Learning

Reference 49

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.157855Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 14f376dc-1eba-4cc6-a911-3a61bf58c658 · outbound

This paper cites Efficiently scaling transformer inference.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Efficiently scaling transformer inference

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.085372Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 03a35b08-88be-4b7e-a73c-b936e9307193 · outbound

This paper cites Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters

Reference 51

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.046864Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 52382218-a884-463d-a4b7-5909ac37b9cf · outbound

This paper cites Cam: Cache merging for memory-efficient llms inference.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Cam: Cache merging for memory-efficient llms inference

Reference 52

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.076180Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 673fdbb5-1d42-483d-a485-aa4b3444a007 · outbound

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

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion

Reference 53

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.277193Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation f827319a-be95-42c8-9a27-420ba7675998 · outbound

This paper cites Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference

Reference 54

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.067489Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 75e56063-6998-4558-8897-5826333d6890 · outbound

This paper cites SCOPE: Optimizing Key-Value Cache Compression in Long-context Generation.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression SCOPE: Optimizing Key-Value Cache Compression in Long-context Generation

Reference 55

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.238739Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 1568f826-7b6a-4846-b1cf-8bc21689a904 · outbound

This paper cites Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them

Reference 56

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.061104Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation c79d223c-ceef-4c26-8a6d-3c70c232ac82 · outbound

This paper cites Layer-Condensed KV Cache for Efficient Inference of Large Language Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Layer-Condensed KV Cache for Efficient Inference of Large Language Models

Reference 57

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.312626Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation e9629a70-76bc-479b-b9ba-0bfaaaeab813 · outbound

This paper cites You Only Cache Once: Decoder-Decoder Architectures for Language Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression You Only Cache Once: Decoder-Decoder Architectures for Language Models

Reference 58

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.196543Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 60eeb619-40af-4ecb-9862-e62d815b8235 · outbound

This paper cites Reducing Transformer Key-Value Cache Size with Cross-Layer Attention.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Reducing Transformer Key-Value Cache Size with Cross-Layer Attention

Reference 59

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.245179Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 0c27c169-0ae3-49b1-9f7d-caf7525136ff · outbound

This paper cites MiniCache: KV Cache Compression in Depth Dimension for Large Language Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression MiniCache: KV Cache Compression in Depth Dimension for Large Language Models

Reference 60

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.255973Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation cfd0daea-9c9f-4a62-a48c-12226667a05a · outbound

This paper cites Prompt compression and contrastive conditioning for controllability and toxicity reduction in language models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Prompt compression and contrastive conditioning for controllability and toxicity reduction in language models

Reference 61

Resolution: malformed identifier; raw_fallback, observed 2026-05-23T04:17:32.101599Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 0c009ca4-dc05-4485-8a41-88af0982531f · outbound

This paper cites Recurrentgpt: Interactive generation of (arbitrarily) long text.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Recurrentgpt: Interactive generation of (arbitrarily) long text

Reference 63

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.093830Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation b0caa610-cc6d-4cc6-956b-e59d36829da2 · outbound

This paper cites Recursively Summarizing Enables Long-Term Dialogue Memory in Large Language Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Recursively Summarizing Enables Long-Term Dialogue Memory in Large Language Models

Reference 64

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.117469Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation babc9ffb-2951-4421-bef6-5da5067c5981 · outbound

This paper cites LLMLingua: Com- pressing prompts for accelerated inference of large language models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression LLMLingua: Com- pressing prompts for accelerated inference of large language models

Reference 65

Resolution: verified exact; doi, observed 2026-05-23T04:17:30.780993Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 22f4079b-52a3-4fd6-9b6a-9de02334a195 · outbound

This paper cites LongLLMLingua: Accelerating and enhancing LLMs in long context scenarios via prompt compression.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression LongLLMLingua: Accelerating and enhancing LLMs in long context scenarios via prompt compression

Reference 66

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.105845Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 831b3c63-bfc9-4832-81de-909b07600c4c · outbound

This paper cites Extending context window of large language models via semantic compression.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Extending context window of large language models via semantic compression

Reference 67

Resolution: verified exact; doi, observed 2026-05-23T04:17:30.794077Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 46b54c63-8fa5-465c-a675-e25e4c161eb4 · outbound

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

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Measuring Mathematical Problem Solving With the MATH Dataset

Reference 68

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.233341Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation bf8f4c97-1980-438e-aaef-3d0985f3fa43 · outbound

This paper cites MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms

Reference 69

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.282686Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 06984b8b-98b9-438c-aa10-b52545e32b64 · outbound

This paper cites Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models

Reference 70

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.211761Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 1b3b9b87-cb69-445e-b9f9-e1b62e74d189 · outbound

This paper cites KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark of Long Context Capable Approaches.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark of Long Context Capable Approaches

Reference 71

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.317978Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 785fb6fb-b315-4da5-b16e-d3d677dfdb89 · outbound

This paper cites TruthfulQA: Measuring How Models Mimic Human Falsehoods.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression TruthfulQA: Measuring How Models Mimic Human Falsehoods

Reference 72

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.136136Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 876e2dd4-3c44-4b80-b1e8-056742970f2a · outbound

This paper cites ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection

Reference 73

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.103467Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation d0f1a838-e778-437b-94aa-e7b27e86f070 · outbound

This paper cites Towards understanding and mitigating social biases in language models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Towards understanding and mitigating social biases in language models

Reference 74

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.081152Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 6d4a7877-8291-4b47-9f15-9ee8526a3d70 · outbound

This paper cites Promptbench: Towards evaluating the robustness of large language models on adversarial prompts.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Promptbench: Towards evaluating the robustness of large language models on adversarial prompts

Reference 75

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.068277Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation d6d8ef6b-c5e0-4782-a59d-b6bfef31afc8 · outbound

This paper cites ” do anything now”: Characterizing and evaluating in-the-wild jailbreak prompts on large language models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression ” do anything now”: Characterizing and evaluating in-the-wild jailbreak prompts on large language models

Reference 76

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.186342Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 4f2c6903-5852-4bdb-99db-82117721ce08 · outbound

This paper cites Multilingual Jailbreak Challenges in Large Language Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Multilingual Jailbreak Challenges in Large Language Models

Reference 77

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.294021Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 1db2b9b9-1446-4933-9494-f75d653fc7f4 · outbound

This paper cites Should We Really Edit Language Models? On the Evaluation of Edited Language Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Should We Really Edit Language Models? On the Evaluation of Edited Language Models

Reference 78

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.174411Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 4bad6f04-1a37-4bf2-a87b-866b9b3cc6ec · outbound

This paper cites Program Synthesis with Large Language Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Program Synthesis with Large Language Models

Reference 79

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.201487Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 3e30bb0f-e511-45ba-b73e-943899986e8b · outbound

This paper cites $\infty$Bench: Extending Long Context Evaluation Beyond 100K Tokens.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression $\infty$Bench: Extending Long Context Evaluation Beyond 100K Tokens

Reference 80

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.180588Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 88107afe-17d4-4b54-9d8b-5767b3aab3df · outbound

This paper cites Yifan Li, Yifan Du, Kun Zhou, Jinpeng Wang, Xin Zhao, and Ji-Rong Wen.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Yifan Li, Yifan Du, Kun Zhou, Jinpeng Wang, Xin Zhao, and Ji-Rong Wen

Reference 81

Resolution: metadata mismatch; doi, observed 2026-05-23T04:17:30.789091Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 20812d0f-3b1c-40d3-b55f-2f68fbd2ea98 · outbound

This paper cites L-Eval: Instituting Standardized Evaluation for Long Context Language Models.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression L-Eval: Instituting Standardized Evaluation for Long Context Language Models

Reference 82

Resolution: metadata mismatch; arxiv_id, observed 2026-05-23T04:17:31.131311Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 344f8ad2-c376-4410-93cf-99cace6355a9 · outbound

This paper cites Landmark Attention: Random-Access Infinite Context Length for Transformers.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Landmark Attention: Random-Access Infinite Context Length for Transformers

Reference 83

Resolution: verified exact; arxiv_id, observed 2026-05-23T04:17:31.223364Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation c2ec18a0-3a2a-4786-8ede-ff292fad4e11 · outbound

This paper cites How long can open-source LLMs truly promise on context length?.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression How long can open-source LLMs truly promise on context length?

Reference 84

Resolution: verified fuzzy; raw_fallback, observed 2026-05-23T04:17:32.151500Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation ae667d52-869c-4ca1-a8e9-d7465654e812 · outbound

This paper cites Lost in the Middle: How Language Models Use Long Contexts.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Lost in the Middle: How Language Models Use Long Contexts

Reference 85

Resolution: malformed identifier; doi_truncated, observed 2026-05-23T04:17:30.804347Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation e9bbe8dc-94b1-42ea-869f-70fe802fe765 · outbound

This paper cites RULER: What's the Real Context Size of Your Long-Context Language Models?.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression RULER: What's the Real Context Size of Your Long-Context Language Models?

Reference 86

Resolution: verified exact; local_arxiv, observed 2026-05-23T04:17:31.152439Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 587a2bad-2291-44b1-af91-680f468ca9e8 · outbound

This paper cites Long range arena : A benchmark for efficient transformers.

Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression Long range arena : A benchmark for efficient transformers

Reference 87

Resolution: malformed identifier; raw_fallback, observed 2026-05-23T04:17:32.174421Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Pith citing papers

Observation a4dcc34e-ae39-4407-ad67-96607a15e5f3 · inbound

The Pitfalls of KV Cache Compression cites this paper.

The Pitfalls of KV Cache Compression Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression

Reference 6

Resolution: metadata mismatch; local_arxiv, observed 2026-05-18T11:32:35.185624Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation e96c7aaf-e00f-42ae-aac4-f8733be98775 · inbound

Position: LLM Inference Should Be Evaluated as Energy-to-Token Production cites this paper.

Position: LLM Inference Should Be Evaluated as Energy-to-Token Production Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression

Reference 57

Resolution: verified exact; local_arxiv, observed 2026-05-13T05:27:19.414579Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.

Observation 01cfe5ec-1c64-4207-9a82-387f1bf6f5b2 · inbound

FastTPS: An Optimized Method for LLM Token Phase for AI accelerators cites this paper.

FastTPS: An Optimized Method for LLM Token Phase for AI accelerators Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression

Reference 52

Resolution: unresolved; no resolver link, observed 2026-07-14T06:16:09.070414Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e470aa28-4602-4f27-9855-a8ebde3b7820 · 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 Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-04T13:43:56.043069Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6e5d5ff8-1e0e-42b3-9390-f96aef9626d9 · 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 Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression

Reference 79

Resolution: unresolved; no resolver link, observed 2026-08-07T00:11:49.083358Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.