Large Multimodal Model Compression via Iterative Efficient Pruning and Distillation Article Swipe
Maolin Wang
,
Yao Zhao
,
Jiajia Liu
,
Jingdong Chen
,
Chenyi Zhuang
,
Jinjie Gu
,
Ruocheng Guo
,
Xiangyu Zhao
·
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.1145/3589335.3648321
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.1145/3589335.3648321
The deployment of Large Multimodal Models (LMMs) within Ant Group has significantly advanced multimodal tasks in payment, security, and advertising, notably enhancing advertisement audition tasks in Alipay. However, the deployment of such sizable models introduces challenges, particularly in increased latency and carbon emissions, which are antithetical to the ideals of Green AI. This paper introduces a novel multi-stage compression strategy for our proprietary LLM, AntGMM. Our methodology pivots on three main aspects: employing small training sample sizes, addressing multi-level redundancy through multi-stage pruning, and introducing an advanced distillation loss design. In our research, we constructed a dataset, the Multimodal Advertisement Audition Dataset (MAAD), from real-world scenarios within Alipay, and conducted experiments to validate the reliability of our proposed strategy. Furthermore, the effectiveness of our strategy is evident in its operational success in Alipay's real-world multimodal advertisement audition for three months from September 2023. Notably, our approach achieved a substantial reduction in latency, decreasing it from 700ms to 90ms, while maintaining online performance with only a slight performance decrease. Moreover, our compressed model is estimated to reduce electricity consumption by approximately 75 million kWh annually compared to the direct deployment of AntGMM, demonstrating our commitment to green AI initiatives.
Related Topics
Concepts
Metadata
- Type
- article
- Language
- en
- Landing Page
- https://doi.org/10.1145/3589335.3648321
- https://dl.acm.org/doi/pdf/10.1145/3589335.3648321
- OA Status
- gold
- Cited By
- 7
- References
- 12
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4396844175
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4396844175Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.1145/3589335.3648321Digital Object Identifier
- Title
-
Large Multimodal Model Compression via Iterative Efficient Pruning and DistillationWork title
- Type
-
articleOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2024Year of publication
- Publication date
-
2024-05-12Full publication date if available
- Authors
-
Maolin Wang, Yao Zhao, Jiajia Liu, Jingdong Chen, Chenyi Zhuang, Jinjie Gu, Ruocheng Guo, Xiangyu ZhaoList of authors in order
- Landing page
-
https://doi.org/10.1145/3589335.3648321Publisher landing page
- PDF URL
-
https://dl.acm.org/doi/pdf/10.1145/3589335.3648321Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
goldOpen access status per OpenAlex
- OA URL
-
https://dl.acm.org/doi/pdf/10.1145/3589335.3648321Direct OA link when available
- Concepts
-
Computer science, Pruning, Compression (physics), Distillation, Data compression, Iterative method, Artificial intelligence, Algorithm, Biology, Chemistry, Agronomy, Composite material, Materials science, Organic chemistryTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
7Total citation count in OpenAlex
- Citations by year (recent)
-
2025: 5, 2024: 2Per-year citation counts (last 5 years)
- References (count)
-
12Number of works referenced by this work
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4396844175 |
|---|---|
| doi | https://doi.org/10.1145/3589335.3648321 |
| ids.doi | https://doi.org/10.1145/3589335.3648321 |
| ids.openalex | https://openalex.org/W4396844175 |
| fwci | 3.71110292 |
| type | article |
| title | Large Multimodal Model Compression via Iterative Efficient Pruning and Distillation |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | 244 |
| biblio.first_page | 235 |
| topics[0].id | https://openalex.org/T10901 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9965000152587891 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1707 |
| topics[0].subfield.display_name | Computer Vision and Pattern Recognition |
| topics[0].display_name | Advanced Data Compression Techniques |
| topics[1].id | https://openalex.org/T10201 |
| topics[1].field.id | https://openalex.org/fields/17 |
| topics[1].field.display_name | Computer Science |
| topics[1].score | 0.9807000160217285 |
| topics[1].domain.id | https://openalex.org/domains/3 |
| topics[1].domain.display_name | Physical Sciences |
| topics[1].subfield.id | https://openalex.org/subfields/1702 |
| topics[1].subfield.display_name | Artificial Intelligence |
| topics[1].display_name | Speech Recognition and Synthesis |
| topics[2].id | https://openalex.org/T10860 |
| topics[2].field.id | https://openalex.org/fields/17 |
| topics[2].field.display_name | Computer Science |
| topics[2].score | 0.9621000289916992 |
| topics[2].domain.id | https://openalex.org/domains/3 |
| topics[2].domain.display_name | Physical Sciences |
| topics[2].subfield.id | https://openalex.org/subfields/1711 |
| topics[2].subfield.display_name | Signal Processing |
| topics[2].display_name | Speech and Audio Processing |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C41008148 |
| concepts[0].level | 0 |
| concepts[0].score | 0.7255657911300659 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[0].display_name | Computer science |
| concepts[1].id | https://openalex.org/C108010975 |
| concepts[1].level | 2 |
| concepts[1].score | 0.6904696226119995 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q500094 |
| concepts[1].display_name | Pruning |
| concepts[2].id | https://openalex.org/C180016635 |
| concepts[2].level | 2 |
| concepts[2].score | 0.6413254141807556 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q2712821 |
| concepts[2].display_name | Compression (physics) |
| concepts[3].id | https://openalex.org/C204030448 |
| concepts[3].level | 2 |
| concepts[3].score | 0.598303496837616 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q101017 |
| concepts[3].display_name | Distillation |
| concepts[4].id | https://openalex.org/C78548338 |
| concepts[4].level | 2 |
| concepts[4].score | 0.5105815529823303 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q2493 |
| concepts[4].display_name | Data compression |
| concepts[5].id | https://openalex.org/C159694833 |
| concepts[5].level | 2 |
| concepts[5].score | 0.41666197776794434 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q2321565 |
| concepts[5].display_name | Iterative method |
| concepts[6].id | https://openalex.org/C154945302 |
| concepts[6].level | 1 |
| concepts[6].score | 0.38467487692832947 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[6].display_name | Artificial intelligence |
| concepts[7].id | https://openalex.org/C11413529 |
| concepts[7].level | 1 |
| concepts[7].score | 0.2371421456336975 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q8366 |
| concepts[7].display_name | Algorithm |
| concepts[8].id | https://openalex.org/C86803240 |
| concepts[8].level | 0 |
| concepts[8].score | 0.0 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q420 |
| concepts[8].display_name | Biology |
| concepts[9].id | https://openalex.org/C185592680 |
| concepts[9].level | 0 |
| concepts[9].score | 0.0 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q2329 |
| concepts[9].display_name | Chemistry |
| concepts[10].id | https://openalex.org/C6557445 |
| concepts[10].level | 1 |
| concepts[10].score | 0.0 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q173113 |
| concepts[10].display_name | Agronomy |
| concepts[11].id | https://openalex.org/C159985019 |
| concepts[11].level | 1 |
| concepts[11].score | 0.0 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q181790 |
| concepts[11].display_name | Composite material |
| concepts[12].id | https://openalex.org/C192562407 |
| concepts[12].level | 0 |
| concepts[12].score | 0.0 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q228736 |
| concepts[12].display_name | Materials science |
| concepts[13].id | https://openalex.org/C178790620 |
| concepts[13].level | 1 |
| concepts[13].score | 0.0 |
| concepts[13].wikidata | https://www.wikidata.org/wiki/Q11351 |
| concepts[13].display_name | Organic chemistry |
| keywords[0].id | https://openalex.org/keywords/computer-science |
| keywords[0].score | 0.7255657911300659 |
| keywords[0].display_name | Computer science |
| keywords[1].id | https://openalex.org/keywords/pruning |
| keywords[1].score | 0.6904696226119995 |
| keywords[1].display_name | Pruning |
| keywords[2].id | https://openalex.org/keywords/compression |
| keywords[2].score | 0.6413254141807556 |
| keywords[2].display_name | Compression (physics) |
| keywords[3].id | https://openalex.org/keywords/distillation |
| keywords[3].score | 0.598303496837616 |
| keywords[3].display_name | Distillation |
| keywords[4].id | https://openalex.org/keywords/data-compression |
| keywords[4].score | 0.5105815529823303 |
| keywords[4].display_name | Data compression |
| keywords[5].id | https://openalex.org/keywords/iterative-method |
| keywords[5].score | 0.41666197776794434 |
| keywords[5].display_name | Iterative method |
| keywords[6].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[6].score | 0.38467487692832947 |
| keywords[6].display_name | Artificial intelligence |
| keywords[7].id | https://openalex.org/keywords/algorithm |
| keywords[7].score | 0.2371421456336975 |
| keywords[7].display_name | Algorithm |
| language | en |
| locations[0].id | doi:10.1145/3589335.3648321 |
| locations[0].is_oa | True |
| locations[0].source | |
| locations[0].license | |
| locations[0].pdf_url | https://dl.acm.org/doi/pdf/10.1145/3589335.3648321 |
| locations[0].version | publishedVersion |
| locations[0].raw_type | proceedings-article |
| locations[0].license_id | |
| locations[0].is_accepted | True |
| locations[0].is_published | True |
| locations[0].raw_source_name | Companion Proceedings of the ACM Web Conference 2024 |
| locations[0].landing_page_url | https://doi.org/10.1145/3589335.3648321 |
| indexed_in | crossref |
| authorships[0].author.id | https://openalex.org/A5021037797 |
| authorships[0].author.orcid | https://orcid.org/0000-0002-0073-0172 |
| authorships[0].author.display_name | Maolin Wang |
| authorships[0].countries | HK |
| authorships[0].affiliations[0].institution_ids | https://openalex.org/I168719708 |
| authorships[0].affiliations[0].raw_affiliation_string | City University of Hong Kong, Hong Kong, China |
| authorships[0].institutions[0].id | https://openalex.org/I168719708 |
| authorships[0].institutions[0].ror | https://ror.org/03q8dnn23 |
| authorships[0].institutions[0].type | education |
| authorships[0].institutions[0].lineage | https://openalex.org/I168719708 |
| authorships[0].institutions[0].country_code | HK |
| authorships[0].institutions[0].display_name | City University of Hong Kong |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Maolin Wang |
| authorships[0].is_corresponding | False |
| authorships[0].raw_affiliation_strings | City University of Hong Kong, Hong Kong, China |
| authorships[1].author.id | https://openalex.org/A5019108113 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-0807-929X |
| authorships[1].author.display_name | Yao Zhao |
| authorships[1].affiliations[0].raw_affiliation_string | AntGroup, Hangzhou, China |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Yao Zhao |
| authorships[1].is_corresponding | False |
| authorships[1].raw_affiliation_strings | AntGroup, Hangzhou, China |
| authorships[2].author.id | https://openalex.org/A5060310537 |
| authorships[2].author.orcid | https://orcid.org/0000-0003-3020-2677 |
| authorships[2].author.display_name | Jiajia Liu |
| authorships[2].affiliations[0].raw_affiliation_string | AntGroup, Hangzhou, China |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Jiajia Liu |
| authorships[2].is_corresponding | False |
| authorships[2].raw_affiliation_strings | AntGroup, Hangzhou, China |
| authorships[3].author.id | https://openalex.org/A5014922345 |
| authorships[3].author.orcid | https://orcid.org/0000-0002-1872-2592 |
| authorships[3].author.display_name | Jingdong Chen |
| authorships[3].affiliations[0].raw_affiliation_string | AntGroup, Hangzhou, China |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Jingdong Chen |
| authorships[3].is_corresponding | False |
| authorships[3].raw_affiliation_strings | AntGroup, Hangzhou, China |
| authorships[4].author.id | https://openalex.org/A5061810614 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-1891-5666 |
| authorships[4].author.display_name | Chenyi Zhuang |
| authorships[4].affiliations[0].raw_affiliation_string | AntGroup, Hangzhou, China |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Chenyi Zhuang |
| authorships[4].is_corresponding | False |
| authorships[4].raw_affiliation_strings | AntGroup, Hangzhou, China |
| authorships[5].author.id | https://openalex.org/A5053242349 |
| authorships[5].author.orcid | https://orcid.org/0000-0001-7596-4945 |
| authorships[5].author.display_name | Jinjie Gu |
| authorships[5].affiliations[0].raw_affiliation_string | AntGroup, Hangzhou, China |
| authorships[5].author_position | middle |
| authorships[5].raw_author_name | Jinjie Gu |
| authorships[5].is_corresponding | False |
| authorships[5].raw_affiliation_strings | AntGroup, Hangzhou, China |
| authorships[6].author.id | https://openalex.org/A5054719216 |
| authorships[6].author.orcid | https://orcid.org/0000-0002-8522-6142 |
| authorships[6].author.display_name | Ruocheng Guo |
| authorships[6].affiliations[0].raw_affiliation_string | ByteDance Research, London, United Kingdom |
| authorships[6].author_position | middle |
| authorships[6].raw_author_name | Ruocheng Guo |
| authorships[6].is_corresponding | False |
| authorships[6].raw_affiliation_strings | ByteDance Research, London, United Kingdom |
| authorships[7].author.id | https://openalex.org/A5100645854 |
| authorships[7].author.orcid | https://orcid.org/0000-0003-2926-4416 |
| authorships[7].author.display_name | Xiangyu Zhao |
| authorships[7].countries | HK |
| authorships[7].affiliations[0].institution_ids | https://openalex.org/I168719708 |
| authorships[7].affiliations[0].raw_affiliation_string | City University of Hong Kong, Hong Kong, China |
| authorships[7].institutions[0].id | https://openalex.org/I168719708 |
| authorships[7].institutions[0].ror | https://ror.org/03q8dnn23 |
| authorships[7].institutions[0].type | education |
| authorships[7].institutions[0].lineage | https://openalex.org/I168719708 |
| authorships[7].institutions[0].country_code | HK |
| authorships[7].institutions[0].display_name | City University of Hong Kong |
| authorships[7].author_position | last |
| authorships[7].raw_author_name | Xiangyu Zhao |
| authorships[7].is_corresponding | False |
| authorships[7].raw_affiliation_strings | City University of Hong Kong, Hong Kong, China |
| has_content.pdf | True |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://dl.acm.org/doi/pdf/10.1145/3589335.3648321 |
| open_access.oa_status | gold |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Large Multimodal Model Compression via Iterative Efficient Pruning and Distillation |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T03:46:38.306776 |
| primary_topic.id | https://openalex.org/T10901 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9965000152587891 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1707 |
| primary_topic.subfield.display_name | Computer Vision and Pattern Recognition |
| primary_topic.display_name | Advanced Data Compression Techniques |
| related_works | https://openalex.org/W2373300491, https://openalex.org/W2378744544, https://openalex.org/W2594301978, https://openalex.org/W2379704676, https://openalex.org/W1998810860, https://openalex.org/W4206442282, https://openalex.org/W2384505857, https://openalex.org/W2355171581, https://openalex.org/W2612632602, https://openalex.org/W2321805087 |
| cited_by_count | 7 |
| counts_by_year[0].year | 2025 |
| counts_by_year[0].cited_by_count | 5 |
| counts_by_year[1].year | 2024 |
| counts_by_year[1].cited_by_count | 2 |
| locations_count | 1 |
| best_oa_location.id | doi:10.1145/3589335.3648321 |
| best_oa_location.is_oa | True |
| best_oa_location.source | |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://dl.acm.org/doi/pdf/10.1145/3589335.3648321 |
| best_oa_location.version | publishedVersion |
| best_oa_location.raw_type | proceedings-article |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | True |
| best_oa_location.is_published | True |
| best_oa_location.raw_source_name | Companion Proceedings of the ACM Web Conference 2024 |
| best_oa_location.landing_page_url | https://doi.org/10.1145/3589335.3648321 |
| primary_location.id | doi:10.1145/3589335.3648321 |
| primary_location.is_oa | True |
| primary_location.source | |
| primary_location.license | |
| primary_location.pdf_url | https://dl.acm.org/doi/pdf/10.1145/3589335.3648321 |
| primary_location.version | publishedVersion |
| primary_location.raw_type | proceedings-article |
| primary_location.license_id | |
| primary_location.is_accepted | True |
| primary_location.is_published | True |
| primary_location.raw_source_name | Companion Proceedings of the ACM Web Conference 2024 |
| primary_location.landing_page_url | https://doi.org/10.1145/3589335.3648321 |
| publication_date | 2024-05-12 |
| publication_year | 2024 |
| referenced_works | https://openalex.org/W2276892413, https://openalex.org/W6775706467, https://openalex.org/W3034288893, https://openalex.org/W3174657338, https://openalex.org/W2965862774, https://openalex.org/W2998183051, https://openalex.org/W4385571831, https://openalex.org/W4382463788, https://openalex.org/W4386076668, https://openalex.org/W2996913633, https://openalex.org/W3015609966, https://openalex.org/W2990714382 |
| referenced_works_count | 12 |
| abstract_inverted_index.a | 55, 95, 147, 164 |
| abstract_inverted_index.75 | 180 |
| abstract_inverted_index.AI | 196 |
| abstract_inverted_index.In | 90 |
| abstract_inverted_index.an | 85 |
| abstract_inverted_index.by | 178 |
| abstract_inverted_index.in | 15, 25, 37, 127, 131, 150 |
| abstract_inverted_index.is | 125, 172 |
| abstract_inverted_index.it | 153 |
| abstract_inverted_index.of | 2, 30, 49, 115, 122, 189 |
| abstract_inverted_index.on | 68 |
| abstract_inverted_index.to | 46, 111, 156, 174, 185, 194 |
| abstract_inverted_index.we | 93 |
| abstract_inverted_index.AI. | 51 |
| abstract_inverted_index.Ant | 8 |
| abstract_inverted_index.Our | 65 |
| abstract_inverted_index.The | 0 |
| abstract_inverted_index.and | 18, 40, 83, 108 |
| abstract_inverted_index.are | 44 |
| abstract_inverted_index.for | 60, 137 |
| abstract_inverted_index.has | 10 |
| abstract_inverted_index.its | 128 |
| abstract_inverted_index.kWh | 182 |
| abstract_inverted_index.our | 61, 91, 116, 123, 144, 169, 192 |
| abstract_inverted_index.the | 28, 47, 97, 113, 120, 186 |
| abstract_inverted_index.LLM, | 63 |
| abstract_inverted_index.This | 52 |
| abstract_inverted_index.from | 103, 140, 154 |
| abstract_inverted_index.loss | 88 |
| abstract_inverted_index.main | 70 |
| abstract_inverted_index.only | 163 |
| abstract_inverted_index.such | 31 |
| abstract_inverted_index.with | 162 |
| abstract_inverted_index.2023. | 142 |
| abstract_inverted_index.700ms | 155 |
| abstract_inverted_index.90ms, | 157 |
| abstract_inverted_index.Green | 50 |
| abstract_inverted_index.Group | 9 |
| abstract_inverted_index.Large | 3 |
| abstract_inverted_index.green | 195 |
| abstract_inverted_index.model | 171 |
| abstract_inverted_index.novel | 56 |
| abstract_inverted_index.paper | 53 |
| abstract_inverted_index.small | 73 |
| abstract_inverted_index.tasks | 14, 24 |
| abstract_inverted_index.three | 69, 138 |
| abstract_inverted_index.which | 43 |
| abstract_inverted_index.while | 158 |
| abstract_inverted_index.(LMMs) | 6 |
| abstract_inverted_index.Models | 5 |
| abstract_inverted_index.carbon | 41 |
| abstract_inverted_index.direct | 187 |
| abstract_inverted_index.ideals | 48 |
| abstract_inverted_index.models | 33 |
| abstract_inverted_index.months | 139 |
| abstract_inverted_index.online | 160 |
| abstract_inverted_index.pivots | 67 |
| abstract_inverted_index.reduce | 175 |
| abstract_inverted_index.sample | 75 |
| abstract_inverted_index.sizes, | 76 |
| abstract_inverted_index.slight | 165 |
| abstract_inverted_index.within | 7, 106 |
| abstract_inverted_index.(MAAD), | 102 |
| abstract_inverted_index.Alipay, | 107 |
| abstract_inverted_index.Alipay. | 26 |
| abstract_inverted_index.AntGMM, | 190 |
| abstract_inverted_index.AntGMM. | 64 |
| abstract_inverted_index.Dataset | 101 |
| abstract_inverted_index.design. | 89 |
| abstract_inverted_index.evident | 126 |
| abstract_inverted_index.latency | 39 |
| abstract_inverted_index.million | 181 |
| abstract_inverted_index.notably | 20 |
| abstract_inverted_index.sizable | 32 |
| abstract_inverted_index.success | 130 |
| abstract_inverted_index.through | 80 |
| abstract_inverted_index.Alipay's | 132 |
| abstract_inverted_index.Audition | 100 |
| abstract_inverted_index.However, | 27 |
| abstract_inverted_index.Notably, | 143 |
| abstract_inverted_index.achieved | 146 |
| abstract_inverted_index.advanced | 12, 86 |
| abstract_inverted_index.annually | 183 |
| abstract_inverted_index.approach | 145 |
| abstract_inverted_index.aspects: | 71 |
| abstract_inverted_index.audition | 23, 136 |
| abstract_inverted_index.compared | 184 |
| abstract_inverted_index.dataset, | 96 |
| abstract_inverted_index.latency, | 151 |
| abstract_inverted_index.payment, | 16 |
| abstract_inverted_index.proposed | 117 |
| abstract_inverted_index.pruning, | 82 |
| abstract_inverted_index.strategy | 59, 124 |
| abstract_inverted_index.training | 74 |
| abstract_inverted_index.validate | 112 |
| abstract_inverted_index.Moreover, | 168 |
| abstract_inverted_index.September | 141 |
| abstract_inverted_index.conducted | 109 |
| abstract_inverted_index.decrease. | 167 |
| abstract_inverted_index.employing | 72 |
| abstract_inverted_index.enhancing | 21 |
| abstract_inverted_index.estimated | 173 |
| abstract_inverted_index.increased | 38 |
| abstract_inverted_index.reduction | 149 |
| abstract_inverted_index.research, | 92 |
| abstract_inverted_index.scenarios | 105 |
| abstract_inverted_index.security, | 17 |
| abstract_inverted_index.strategy. | 118 |
| abstract_inverted_index.Multimodal | 4, 98 |
| abstract_inverted_index.addressing | 77 |
| abstract_inverted_index.commitment | 193 |
| abstract_inverted_index.compressed | 170 |
| abstract_inverted_index.decreasing | 152 |
| abstract_inverted_index.deployment | 1, 29, 188 |
| abstract_inverted_index.emissions, | 42 |
| abstract_inverted_index.introduces | 34, 54 |
| abstract_inverted_index.multimodal | 13, 134 |
| abstract_inverted_index.real-world | 104, 133 |
| abstract_inverted_index.redundancy | 79 |
| abstract_inverted_index.challenges, | 35 |
| abstract_inverted_index.compression | 58 |
| abstract_inverted_index.constructed | 94 |
| abstract_inverted_index.consumption | 177 |
| abstract_inverted_index.electricity | 176 |
| abstract_inverted_index.experiments | 110 |
| abstract_inverted_index.introducing | 84 |
| abstract_inverted_index.maintaining | 159 |
| abstract_inverted_index.methodology | 66 |
| abstract_inverted_index.multi-level | 78 |
| abstract_inverted_index.multi-stage | 57, 81 |
| abstract_inverted_index.operational | 129 |
| abstract_inverted_index.performance | 161, 166 |
| abstract_inverted_index.proprietary | 62 |
| abstract_inverted_index.reliability | 114 |
| abstract_inverted_index.substantial | 148 |
| abstract_inverted_index.Furthermore, | 119 |
| abstract_inverted_index.advertising, | 19 |
| abstract_inverted_index.antithetical | 45 |
| abstract_inverted_index.distillation | 87 |
| abstract_inverted_index.initiatives. | 197 |
| abstract_inverted_index.particularly | 36 |
| abstract_inverted_index.Advertisement | 99 |
| abstract_inverted_index.advertisement | 22, 135 |
| abstract_inverted_index.approximately | 179 |
| abstract_inverted_index.demonstrating | 191 |
| abstract_inverted_index.effectiveness | 121 |
| abstract_inverted_index.significantly | 11 |
| cited_by_percentile_year.max | 98 |
| cited_by_percentile_year.min | 94 |
| countries_distinct_count | 1 |
| institutions_distinct_count | 8 |
| citation_normalized_percentile.value | 0.89790187 |
| citation_normalized_percentile.is_in_top_1_percent | False |
| citation_normalized_percentile.is_in_top_10_percent | True |