EMS: Adaptive Evict-then-Merge Strategy for Head-wise KV Cache Compression Based on Global-Local Importance Article Swipe
Yingxin Li
,
Ye Li
,
Yuan Meng
,
Xinzhu Ma
,
Zihan Geng
,
Shu‐Tao Xia
,
Zhi Wang
·
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2412.08521
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2412.08521
As large language models (LLMs) continue to advance, the demand for higher quality and faster processing of long contexts across various applications is growing. KV cache is widely adopted as it stores previously generated key and value tokens, effectively reducing redundant computations during inference. However, as memory overhead becomes a significant concern, efficient compression of KV cache has gained increasing attention. Most existing methods perform compression from two perspectives: identifying important tokens and designing compression strategies. However, these approaches often produce biased distributions of important tokens due to the influence of accumulated attention scores or positional encoding. Furthermore, they overlook the sparsity and redundancy across different heads, which leads to difficulties in preserving the most effective information at the head level. To this end, we propose EMS to overcome these limitations, while achieving better KV cache compression under extreme compression ratios. Specifically, we introduce a Global-Local score that combines accumulated attention scores from both global and local KV tokens to better identify the token importance. For the compression strategy, we design an adaptive and unified Evict-then-Merge framework that accounts for the sparsity and redundancy of KV tokens across different heads. Additionally, we implement the head-wise parallel compression through a zero-class mechanism to enhance efficiency. Extensive experiments demonstrate our SOTA performance even under extreme compression ratios. EMS consistently achieves the lowest perplexity, improves scores by over 1.28 points across four LLMs on LongBench under a 256 cache budget, and preserves 95% retrieval accuracy with a cache budget less than 2% of the context length in the Needle-in-a-Haystack task.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2412.08521
- https://arxiv.org/pdf/2412.08521
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4405307151
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4405307151Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2412.08521Digital Object Identifier
- Title
-
EMS: Adaptive Evict-then-Merge Strategy for Head-wise KV Cache Compression Based on Global-Local ImportanceWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2024Year of publication
- Publication date
-
2024-12-11Full publication date if available
- Authors
-
Yingxin Li, Ye Li, Yuan Meng, Xinzhu Ma, Zihan Geng, Shu‐Tao Xia, Zhi WangList of authors in order
- Landing page
-
https://arxiv.org/abs/2412.08521Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2412.08521Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2412.08521Direct OA link when available
- Concepts
-
Merge (version control), Computer science, Cache, Real-time computing, Computer network, Parallel computingTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
0Total citation count in OpenAlex
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4405307151 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2412.08521 |
| ids.doi | https://doi.org/10.48550/arxiv.2412.08521 |
| ids.openalex | https://openalex.org/W4405307151 |
| fwci | |
| type | preprint |
| title | EMS: Adaptive Evict-then-Merge Strategy for Head-wise KV Cache Compression Based on Global-Local Importance |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T11478 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9990000128746033 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1705 |
| topics[0].subfield.display_name | Computer Networks and Communications |
| topics[0].display_name | Caching and Content Delivery |
| topics[1].id | https://openalex.org/T11181 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.992900013923645 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1705 |
| topics[1].subfield.display_name | Computer Networks and Communications |
| topics[1].display_name | Advanced Data Storage Technologies |
| topics[2].id | https://openalex.org/T11269 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9926000237464905 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1702 |
| topics[2].subfield.display_name | Artificial Intelligence |
| topics[2].display_name | Algorithms and Data Compression |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C197129107 |
| concepts[0].level | 2 |
| concepts[0].score | 0.7727051973342896 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q1921621 |
| concepts[0].display_name | Merge (version control) |
| concepts[1].id | https://openalex.org/C41008148 |
| concepts[1].level | 0 |
| concepts[1].score | 0.7389147281646729 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[1].display_name | Computer science |
| concepts[2].id | https://openalex.org/C115537543 |
| concepts[2].level | 2 |
| concepts[2].score | 0.7305139303207397 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q165596 |
| concepts[2].display_name | Cache |
| concepts[3].id | https://openalex.org/C79403827 |
| concepts[3].level | 1 |
| concepts[3].score | 0.37791818380355835 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q3988 |
| concepts[3].display_name | Real-time computing |
| concepts[4].id | https://openalex.org/C31258907 |
| concepts[4].level | 1 |
| concepts[4].score | 0.3658849596977234 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q1301371 |
| concepts[4].display_name | Computer network |
| concepts[5].id | https://openalex.org/C173608175 |
| concepts[5].level | 1 |
| concepts[5].score | 0.3191361129283905 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q232661 |
| concepts[5].display_name | Parallel computing |
| keywords[0].id | https://openalex.org/keywords/merge |
| keywords[0].score | 0.7727051973342896 |
| keywords[0].display_name | Merge (version control) |
| keywords[1].id | https://openalex.org/keywords/computer-science |
| keywords[1].score | 0.7389147281646729 |
| keywords[1].display_name | Computer science |
| keywords[2].id | https://openalex.org/keywords/cache |
| keywords[2].score | 0.7305139303207397 |
| keywords[2].display_name | Cache |
| keywords[3].id | https://openalex.org/keywords/real-time-computing |
| keywords[3].score | 0.37791818380355835 |
| keywords[3].display_name | Real-time computing |
| keywords[4].id | https://openalex.org/keywords/computer-network |
| keywords[4].score | 0.3658849596977234 |
| keywords[4].display_name | Computer network |
| keywords[5].id | https://openalex.org/keywords/parallel-computing |
| keywords[5].score | 0.3191361129283905 |
| keywords[5].display_name | Parallel computing |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2412.08521 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | |
| locations[0].pdf_url | https://arxiv.org/pdf/2412.08521 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2412.08521 |
| locations[1].id | doi:10.48550/arxiv.2412.08521 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2412.08521 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5101402498 |
| authorships[0].author.orcid | https://orcid.org/0000-0003-1184-755X |
| authorships[0].author.display_name | Yingxin Li |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Li, Yingxin |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5100339237 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-6601-0864 |
| authorships[1].author.display_name | Ye Li |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Li, Ye |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5050528589 |
| authorships[2].author.orcid | https://orcid.org/0000-0001-6468-8623 |
| authorships[2].author.display_name | Yuan Meng |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Meng, Yuan |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5030101527 |
| authorships[3].author.orcid | https://orcid.org/0000-0003-0504-0186 |
| authorships[3].author.display_name | Xinzhu Ma |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Ma, Xinzhu |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5101409146 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-2690-3374 |
| authorships[4].author.display_name | Zihan Geng |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Geng, Zihan |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5034104790 |
| authorships[5].author.orcid | https://orcid.org/0000-0002-8639-982X |
| authorships[5].author.display_name | Shu‐Tao Xia |
| authorships[5].author_position | middle |
| authorships[5].raw_author_name | Xia, Shutao |
| authorships[5].is_corresponding | False |
| authorships[6].author.id | https://openalex.org/A5100376384 |
| authorships[6].author.orcid | https://orcid.org/0000-0002-1095-9632 |
| authorships[6].author.display_name | Zhi Wang |
| authorships[6].author_position | last |
| authorships[6].raw_author_name | Wang, Zhi |
| authorships[6].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2412.08521 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | EMS: Adaptive Evict-then-Merge Strategy for Head-wise KV Cache Compression Based on Global-Local Importance |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T11478 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9990000128746033 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1705 |
| primary_topic.subfield.display_name | Computer Networks and Communications |
| primary_topic.display_name | Caching and Content Delivery |
| related_works | https://openalex.org/W4234886518, https://openalex.org/W2389591058, https://openalex.org/W2382112581, https://openalex.org/W3124036233, https://openalex.org/W4229787472, https://openalex.org/W2486541857, https://openalex.org/W2108840191, https://openalex.org/W2759366996, https://openalex.org/W2110679372, https://openalex.org/W2354081742 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2412.08521 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2412.08521 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2412.08521 |
| primary_location.id | pmh:oai:arXiv.org:2412.08521 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | |
| primary_location.pdf_url | https://arxiv.org/pdf/2412.08521 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2412.08521 |
| publication_date | 2024-12-11 |
| publication_year | 2024 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 49, 144, 198, 233, 243 |
| abstract_inverted_index.2% | 248 |
| abstract_inverted_index.As | 0 |
| abstract_inverted_index.KV | 24, 55, 134, 157, 185 |
| abstract_inverted_index.To | 121 |
| abstract_inverted_index.an | 171 |
| abstract_inverted_index.as | 29, 45 |
| abstract_inverted_index.at | 117 |
| abstract_inverted_index.by | 223 |
| abstract_inverted_index.in | 111, 253 |
| abstract_inverted_index.is | 22, 26 |
| abstract_inverted_index.it | 30 |
| abstract_inverted_index.of | 16, 54, 83, 90, 184, 249 |
| abstract_inverted_index.on | 230 |
| abstract_inverted_index.or | 94 |
| abstract_inverted_index.to | 6, 87, 109, 127, 159, 201 |
| abstract_inverted_index.we | 124, 142, 169, 191 |
| abstract_inverted_index.256 | 234 |
| abstract_inverted_index.95% | 239 |
| abstract_inverted_index.EMS | 126, 215 |
| abstract_inverted_index.For | 165 |
| abstract_inverted_index.and | 13, 35, 72, 102, 155, 173, 182, 237 |
| abstract_inverted_index.due | 86 |
| abstract_inverted_index.for | 10, 179 |
| abstract_inverted_index.has | 57 |
| abstract_inverted_index.key | 34 |
| abstract_inverted_index.our | 207 |
| abstract_inverted_index.the | 8, 88, 100, 113, 118, 162, 166, 180, 193, 218, 250, 254 |
| abstract_inverted_index.two | 67 |
| abstract_inverted_index.1.28 | 225 |
| abstract_inverted_index.LLMs | 229 |
| abstract_inverted_index.Most | 61 |
| abstract_inverted_index.SOTA | 208 |
| abstract_inverted_index.both | 153 |
| abstract_inverted_index.end, | 123 |
| abstract_inverted_index.even | 210 |
| abstract_inverted_index.four | 228 |
| abstract_inverted_index.from | 66, 152 |
| abstract_inverted_index.head | 119 |
| abstract_inverted_index.less | 246 |
| abstract_inverted_index.long | 17 |
| abstract_inverted_index.most | 114 |
| abstract_inverted_index.over | 224 |
| abstract_inverted_index.than | 247 |
| abstract_inverted_index.that | 147, 177 |
| abstract_inverted_index.they | 98 |
| abstract_inverted_index.this | 122 |
| abstract_inverted_index.with | 242 |
| abstract_inverted_index.cache | 25, 56, 135, 235, 244 |
| abstract_inverted_index.large | 1 |
| abstract_inverted_index.leads | 108 |
| abstract_inverted_index.local | 156 |
| abstract_inverted_index.often | 79 |
| abstract_inverted_index.score | 146 |
| abstract_inverted_index.task. | 256 |
| abstract_inverted_index.these | 77, 129 |
| abstract_inverted_index.token | 163 |
| abstract_inverted_index.under | 137, 211, 232 |
| abstract_inverted_index.value | 36 |
| abstract_inverted_index.which | 107 |
| abstract_inverted_index.while | 131 |
| abstract_inverted_index.(LLMs) | 4 |
| abstract_inverted_index.across | 19, 104, 187, 227 |
| abstract_inverted_index.better | 133, 160 |
| abstract_inverted_index.biased | 81 |
| abstract_inverted_index.budget | 245 |
| abstract_inverted_index.demand | 9 |
| abstract_inverted_index.design | 170 |
| abstract_inverted_index.during | 42 |
| abstract_inverted_index.faster | 14 |
| abstract_inverted_index.gained | 58 |
| abstract_inverted_index.global | 154 |
| abstract_inverted_index.heads, | 106 |
| abstract_inverted_index.heads. | 189 |
| abstract_inverted_index.higher | 11 |
| abstract_inverted_index.length | 252 |
| abstract_inverted_index.level. | 120 |
| abstract_inverted_index.lowest | 219 |
| abstract_inverted_index.memory | 46 |
| abstract_inverted_index.models | 3 |
| abstract_inverted_index.points | 226 |
| abstract_inverted_index.scores | 93, 151, 222 |
| abstract_inverted_index.stores | 31 |
| abstract_inverted_index.tokens | 71, 85, 158, 186 |
| abstract_inverted_index.widely | 27 |
| abstract_inverted_index.adopted | 28 |
| abstract_inverted_index.becomes | 48 |
| abstract_inverted_index.budget, | 236 |
| abstract_inverted_index.context | 251 |
| abstract_inverted_index.enhance | 202 |
| abstract_inverted_index.extreme | 138, 212 |
| abstract_inverted_index.methods | 63 |
| abstract_inverted_index.perform | 64 |
| abstract_inverted_index.produce | 80 |
| abstract_inverted_index.propose | 125 |
| abstract_inverted_index.quality | 12 |
| abstract_inverted_index.ratios. | 140, 214 |
| abstract_inverted_index.through | 197 |
| abstract_inverted_index.tokens, | 37 |
| abstract_inverted_index.unified | 174 |
| abstract_inverted_index.various | 20 |
| abstract_inverted_index.However, | 44, 76 |
| abstract_inverted_index.accounts | 178 |
| abstract_inverted_index.accuracy | 241 |
| abstract_inverted_index.achieves | 217 |
| abstract_inverted_index.adaptive | 172 |
| abstract_inverted_index.advance, | 7 |
| abstract_inverted_index.combines | 148 |
| abstract_inverted_index.concern, | 51 |
| abstract_inverted_index.contexts | 18 |
| abstract_inverted_index.continue | 5 |
| abstract_inverted_index.existing | 62 |
| abstract_inverted_index.growing. | 23 |
| abstract_inverted_index.identify | 161 |
| abstract_inverted_index.improves | 221 |
| abstract_inverted_index.language | 2 |
| abstract_inverted_index.overcome | 128 |
| abstract_inverted_index.overhead | 47 |
| abstract_inverted_index.overlook | 99 |
| abstract_inverted_index.parallel | 195 |
| abstract_inverted_index.reducing | 39 |
| abstract_inverted_index.sparsity | 101, 181 |
| abstract_inverted_index.Extensive | 204 |
| abstract_inverted_index.LongBench | 231 |
| abstract_inverted_index.achieving | 132 |
| abstract_inverted_index.attention | 92, 150 |
| abstract_inverted_index.designing | 73 |
| abstract_inverted_index.different | 105, 188 |
| abstract_inverted_index.effective | 115 |
| abstract_inverted_index.efficient | 52 |
| abstract_inverted_index.encoding. | 96 |
| abstract_inverted_index.framework | 176 |
| abstract_inverted_index.generated | 33 |
| abstract_inverted_index.head-wise | 194 |
| abstract_inverted_index.implement | 192 |
| abstract_inverted_index.important | 70, 84 |
| abstract_inverted_index.influence | 89 |
| abstract_inverted_index.introduce | 143 |
| abstract_inverted_index.mechanism | 200 |
| abstract_inverted_index.preserves | 238 |
| abstract_inverted_index.redundant | 40 |
| abstract_inverted_index.retrieval | 240 |
| abstract_inverted_index.strategy, | 168 |
| abstract_inverted_index.approaches | 78 |
| abstract_inverted_index.attention. | 60 |
| abstract_inverted_index.increasing | 59 |
| abstract_inverted_index.inference. | 43 |
| abstract_inverted_index.positional | 95 |
| abstract_inverted_index.preserving | 112 |
| abstract_inverted_index.previously | 32 |
| abstract_inverted_index.processing | 15 |
| abstract_inverted_index.redundancy | 103, 183 |
| abstract_inverted_index.zero-class | 199 |
| abstract_inverted_index.accumulated | 91, 149 |
| abstract_inverted_index.compression | 53, 65, 74, 136, 139, 167, 196, 213 |
| abstract_inverted_index.demonstrate | 206 |
| abstract_inverted_index.effectively | 38 |
| abstract_inverted_index.efficiency. | 203 |
| abstract_inverted_index.experiments | 205 |
| abstract_inverted_index.identifying | 69 |
| abstract_inverted_index.importance. | 164 |
| abstract_inverted_index.information | 116 |
| abstract_inverted_index.performance | 209 |
| abstract_inverted_index.perplexity, | 220 |
| abstract_inverted_index.significant | 50 |
| abstract_inverted_index.strategies. | 75 |
| abstract_inverted_index.Furthermore, | 97 |
| abstract_inverted_index.Global-Local | 145 |
| abstract_inverted_index.applications | 21 |
| abstract_inverted_index.computations | 41 |
| abstract_inverted_index.consistently | 216 |
| abstract_inverted_index.difficulties | 110 |
| abstract_inverted_index.limitations, | 130 |
| abstract_inverted_index.Additionally, | 190 |
| abstract_inverted_index.Specifically, | 141 |
| abstract_inverted_index.distributions | 82 |
| abstract_inverted_index.perspectives: | 68 |
| abstract_inverted_index.Evict-then-Merge | 175 |
| abstract_inverted_index.Needle-in-a-Haystack | 255 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 7 |
| citation_normalized_percentile |