Latent-space adversarial training with post-aware calibration for defending large language models against jailbreak attacks Article Swipe
Yi Xin
,
Ting Li
,
Linlin Wang
,
Xiaoling Wang
,
Liang He
·
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2501.10639
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2501.10639
Ensuring safety alignment is a critical requirement for large language models (LLMs), particularly given increasing deployment in real-world applications. Despite considerable advancements, LLMs remain susceptible to jailbreak attacks, which exploit system vulnerabilities to circumvent safety measures and elicit harmful or inappropriate outputs. Furthermore, while adversarial training-based defense methods have shown promise, a prevalent issue is the unintended over-defense behavior, wherein models excessively reject benign queries, significantly undermining their practical utility. To address these limitations, we introduce LATPC, a Latent-space Adversarial Training with Post-aware Calibration framework. LATPC dynamically identifies safety-critical latent dimensions by contrasting harmful and benign inputs, enabling the adaptive construction of targeted refusal feature removal attacks. This mechanism allows adversarial training to concentrate on real-world jailbreak tactics that disguise harmful queries as benign ones. During inference, LATPC employs an efficient embedding-level calibration mechanism to minimize over-defense behaviors with negligible computational overhead. Experimental results across five types of disguise-based jailbreak attacks demonstrate that LATPC achieves a superior balance between safety and utility compared to existing defense frameworks. Further analysis demonstrates the effectiveness of leveraging safety-critical dimensions in developing robust defense methods against jailbreak attacks.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2501.10639
- https://arxiv.org/pdf/2501.10639
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4406692476
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4406692476Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2501.10639Digital Object Identifier
- Title
-
Latent-space adversarial training with post-aware calibration for defending large language models against jailbreak attacksWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-01-18Full publication date if available
- Authors
-
Yi Xin, Ting Li, Linlin Wang, Xiaoling Wang, Liang HeList of authors in order
- Landing page
-
https://arxiv.org/abs/2501.10639Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2501.10639Direct 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/2501.10639Direct OA link when available
- Concepts
-
Adversarial system, Training (meteorology), Space (punctuation), Calibration, Computer science, Artificial intelligence, Computer security, Machine learning, Psychology, Mathematics, Statistics, Geography, Meteorology, Operating systemTop 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/W4406692476 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2501.10639 |
| ids.doi | https://doi.org/10.48550/arxiv.2501.10639 |
| ids.openalex | https://openalex.org/W4406692476 |
| fwci | |
| type | preprint |
| title | Latent-space adversarial training with post-aware calibration for defending large language models against jailbreak attacks |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T11689 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.992900013923645 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1702 |
| topics[0].subfield.display_name | Artificial Intelligence |
| topics[0].display_name | Adversarial Robustness in Machine Learning |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C37736160 |
| concepts[0].level | 2 |
| concepts[0].score | 0.8959192037582397 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q1801315 |
| concepts[0].display_name | Adversarial system |
| concepts[1].id | https://openalex.org/C2777211547 |
| concepts[1].level | 2 |
| concepts[1].score | 0.6913646459579468 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q17141490 |
| concepts[1].display_name | Training (meteorology) |
| concepts[2].id | https://openalex.org/C2778572836 |
| concepts[2].level | 2 |
| concepts[2].score | 0.634892463684082 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q380933 |
| concepts[2].display_name | Space (punctuation) |
| concepts[3].id | https://openalex.org/C165838908 |
| concepts[3].level | 2 |
| concepts[3].score | 0.6227689981460571 |
| concepts[3].wikidata | https://www.wikidata.org/wiki/Q736777 |
| concepts[3].display_name | Calibration |
| concepts[4].id | https://openalex.org/C41008148 |
| concepts[4].level | 0 |
| concepts[4].score | 0.5501967072486877 |
| concepts[4].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[4].display_name | Computer science |
| concepts[5].id | https://openalex.org/C154945302 |
| concepts[5].level | 1 |
| concepts[5].score | 0.48440486192703247 |
| concepts[5].wikidata | https://www.wikidata.org/wiki/Q11660 |
| concepts[5].display_name | Artificial intelligence |
| concepts[6].id | https://openalex.org/C38652104 |
| concepts[6].level | 1 |
| concepts[6].score | 0.42635494470596313 |
| concepts[6].wikidata | https://www.wikidata.org/wiki/Q3510521 |
| concepts[6].display_name | Computer security |
| concepts[7].id | https://openalex.org/C119857082 |
| concepts[7].level | 1 |
| concepts[7].score | 0.42510873079299927 |
| concepts[7].wikidata | https://www.wikidata.org/wiki/Q2539 |
| concepts[7].display_name | Machine learning |
| concepts[8].id | https://openalex.org/C15744967 |
| concepts[8].level | 0 |
| concepts[8].score | 0.3353906273841858 |
| concepts[8].wikidata | https://www.wikidata.org/wiki/Q9418 |
| concepts[8].display_name | Psychology |
| concepts[9].id | https://openalex.org/C33923547 |
| concepts[9].level | 0 |
| concepts[9].score | 0.16498076915740967 |
| concepts[9].wikidata | https://www.wikidata.org/wiki/Q395 |
| concepts[9].display_name | Mathematics |
| concepts[10].id | https://openalex.org/C105795698 |
| concepts[10].level | 1 |
| concepts[10].score | 0.16345912218093872 |
| concepts[10].wikidata | https://www.wikidata.org/wiki/Q12483 |
| concepts[10].display_name | Statistics |
| concepts[11].id | https://openalex.org/C205649164 |
| concepts[11].level | 0 |
| concepts[11].score | 0.13358429074287415 |
| concepts[11].wikidata | https://www.wikidata.org/wiki/Q1071 |
| concepts[11].display_name | Geography |
| concepts[12].id | https://openalex.org/C153294291 |
| concepts[12].level | 1 |
| concepts[12].score | 0.0 |
| concepts[12].wikidata | https://www.wikidata.org/wiki/Q25261 |
| concepts[12].display_name | Meteorology |
| concepts[13].id | https://openalex.org/C111919701 |
| concepts[13].level | 1 |
| concepts[13].score | 0.0 |
| concepts[13].wikidata | https://www.wikidata.org/wiki/Q9135 |
| concepts[13].display_name | Operating system |
| keywords[0].id | https://openalex.org/keywords/adversarial-system |
| keywords[0].score | 0.8959192037582397 |
| keywords[0].display_name | Adversarial system |
| keywords[1].id | https://openalex.org/keywords/training |
| keywords[1].score | 0.6913646459579468 |
| keywords[1].display_name | Training (meteorology) |
| keywords[2].id | https://openalex.org/keywords/space |
| keywords[2].score | 0.634892463684082 |
| keywords[2].display_name | Space (punctuation) |
| keywords[3].id | https://openalex.org/keywords/calibration |
| keywords[3].score | 0.6227689981460571 |
| keywords[3].display_name | Calibration |
| keywords[4].id | https://openalex.org/keywords/computer-science |
| keywords[4].score | 0.5501967072486877 |
| keywords[4].display_name | Computer science |
| keywords[5].id | https://openalex.org/keywords/artificial-intelligence |
| keywords[5].score | 0.48440486192703247 |
| keywords[5].display_name | Artificial intelligence |
| keywords[6].id | https://openalex.org/keywords/computer-security |
| keywords[6].score | 0.42635494470596313 |
| keywords[6].display_name | Computer security |
| keywords[7].id | https://openalex.org/keywords/machine-learning |
| keywords[7].score | 0.42510873079299927 |
| keywords[7].display_name | Machine learning |
| keywords[8].id | https://openalex.org/keywords/psychology |
| keywords[8].score | 0.3353906273841858 |
| keywords[8].display_name | Psychology |
| keywords[9].id | https://openalex.org/keywords/mathematics |
| keywords[9].score | 0.16498076915740967 |
| keywords[9].display_name | Mathematics |
| keywords[10].id | https://openalex.org/keywords/statistics |
| keywords[10].score | 0.16345912218093872 |
| keywords[10].display_name | Statistics |
| keywords[11].id | https://openalex.org/keywords/geography |
| keywords[11].score | 0.13358429074287415 |
| keywords[11].display_name | Geography |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2501.10639 |
| 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 | cc-by |
| locations[0].pdf_url | https://arxiv.org/pdf/2501.10639 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | https://openalex.org/licenses/cc-by |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2501.10639 |
| locations[1].id | doi:10.48550/arxiv.2501.10639 |
| 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 | cc-by |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | https://openalex.org/licenses/cc-by |
| 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.2501.10639 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5044754372 |
| authorships[0].author.orcid | https://orcid.org/0000-0002-6988-5121 |
| authorships[0].author.display_name | Yi Xin |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Yi, Xin |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5100416826 |
| authorships[1].author.orcid | https://orcid.org/0000-0001-5145-3024 |
| authorships[1].author.display_name | Ting Li |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Li, Yue |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5100348981 |
| authorships[2].author.orcid | https://orcid.org/0000-0002-2033-9277 |
| authorships[2].author.display_name | Linlin Wang |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Wang, Linlin |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5100344619 |
| authorships[3].author.orcid | https://orcid.org/0000-0001-8049-2057 |
| authorships[3].author.display_name | Xiaoling Wang |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Wang, Xiaoling |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5115597505 |
| authorships[4].author.orcid | |
| authorships[4].author.display_name | Liang He |
| authorships[4].author_position | last |
| authorships[4].raw_author_name | He, Liang |
| authorships[4].is_corresponding | False |
| has_content.pdf | True |
| has_content.grobid_xml | True |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2501.10639 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | Latent-space adversarial training with post-aware calibration for defending large language models against jailbreak attacks |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T11689 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.992900013923645 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1702 |
| primary_topic.subfield.display_name | Artificial Intelligence |
| primary_topic.display_name | Adversarial Robustness in Machine Learning |
| related_works | https://openalex.org/W2502115930, https://openalex.org/W2482350142, https://openalex.org/W4246396837, https://openalex.org/W3126451824, https://openalex.org/W1561927205, https://openalex.org/W3191453585, https://openalex.org/W4297672492, https://openalex.org/W4310988119, https://openalex.org/W4285226279, https://openalex.org/W4288019534 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2501.10639 |
| 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 | cc-by |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2501.10639 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | https://openalex.org/licenses/cc-by |
| 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/2501.10639 |
| primary_location.id | pmh:oai:arXiv.org:2501.10639 |
| 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 | cc-by |
| primary_location.pdf_url | https://arxiv.org/pdf/2501.10639 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | https://openalex.org/licenses/cc-by |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2501.10639 |
| publication_date | 2025-01-18 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 4, 51, 77, 155 |
| abstract_inverted_index.To | 70 |
| abstract_inverted_index.an | 129 |
| abstract_inverted_index.as | 122 |
| abstract_inverted_index.by | 91 |
| abstract_inverted_index.in | 16, 176 |
| abstract_inverted_index.is | 3, 54 |
| abstract_inverted_index.of | 101, 147, 172 |
| abstract_inverted_index.on | 114 |
| abstract_inverted_index.or | 39 |
| abstract_inverted_index.to | 25, 32, 112, 134, 163 |
| abstract_inverted_index.we | 74 |
| abstract_inverted_index.and | 36, 94, 160 |
| abstract_inverted_index.for | 7 |
| abstract_inverted_index.the | 55, 98, 170 |
| abstract_inverted_index.LLMs | 22 |
| abstract_inverted_index.This | 107 |
| abstract_inverted_index.five | 145 |
| abstract_inverted_index.have | 48 |
| abstract_inverted_index.that | 118, 152 |
| abstract_inverted_index.with | 81, 138 |
| abstract_inverted_index.LATPC | 85, 127, 153 |
| abstract_inverted_index.given | 13 |
| abstract_inverted_index.issue | 53 |
| abstract_inverted_index.large | 8 |
| abstract_inverted_index.ones. | 124 |
| abstract_inverted_index.shown | 49 |
| abstract_inverted_index.their | 67 |
| abstract_inverted_index.these | 72 |
| abstract_inverted_index.types | 146 |
| abstract_inverted_index.which | 28 |
| abstract_inverted_index.while | 43 |
| abstract_inverted_index.During | 125 |
| abstract_inverted_index.LATPC, | 76 |
| abstract_inverted_index.across | 144 |
| abstract_inverted_index.allows | 109 |
| abstract_inverted_index.benign | 63, 95, 123 |
| abstract_inverted_index.elicit | 37 |
| abstract_inverted_index.latent | 89 |
| abstract_inverted_index.models | 10, 60 |
| abstract_inverted_index.reject | 62 |
| abstract_inverted_index.remain | 23 |
| abstract_inverted_index.robust | 178 |
| abstract_inverted_index.safety | 1, 34, 159 |
| abstract_inverted_index.system | 30 |
| abstract_inverted_index.(LLMs), | 11 |
| abstract_inverted_index.Despite | 19 |
| abstract_inverted_index.Further | 167 |
| abstract_inverted_index.address | 71 |
| abstract_inverted_index.against | 181 |
| abstract_inverted_index.attacks | 150 |
| abstract_inverted_index.balance | 157 |
| abstract_inverted_index.between | 158 |
| abstract_inverted_index.defense | 46, 165, 179 |
| abstract_inverted_index.employs | 128 |
| abstract_inverted_index.exploit | 29 |
| abstract_inverted_index.feature | 104 |
| abstract_inverted_index.harmful | 38, 93, 120 |
| abstract_inverted_index.inputs, | 96 |
| abstract_inverted_index.methods | 47, 180 |
| abstract_inverted_index.queries | 121 |
| abstract_inverted_index.refusal | 103 |
| abstract_inverted_index.removal | 105 |
| abstract_inverted_index.results | 143 |
| abstract_inverted_index.tactics | 117 |
| abstract_inverted_index.utility | 161 |
| abstract_inverted_index.wherein | 59 |
| abstract_inverted_index.Ensuring | 0 |
| abstract_inverted_index.Training | 80 |
| abstract_inverted_index.achieves | 154 |
| abstract_inverted_index.adaptive | 99 |
| abstract_inverted_index.analysis | 168 |
| abstract_inverted_index.attacks, | 27 |
| abstract_inverted_index.attacks. | 106, 183 |
| abstract_inverted_index.compared | 162 |
| abstract_inverted_index.critical | 5 |
| abstract_inverted_index.disguise | 119 |
| abstract_inverted_index.enabling | 97 |
| abstract_inverted_index.existing | 164 |
| abstract_inverted_index.language | 9 |
| abstract_inverted_index.measures | 35 |
| abstract_inverted_index.minimize | 135 |
| abstract_inverted_index.outputs. | 41 |
| abstract_inverted_index.promise, | 50 |
| abstract_inverted_index.queries, | 64 |
| abstract_inverted_index.superior | 156 |
| abstract_inverted_index.targeted | 102 |
| abstract_inverted_index.training | 111 |
| abstract_inverted_index.utility. | 69 |
| abstract_inverted_index.alignment | 2 |
| abstract_inverted_index.behavior, | 58 |
| abstract_inverted_index.behaviors | 137 |
| abstract_inverted_index.efficient | 130 |
| abstract_inverted_index.introduce | 75 |
| abstract_inverted_index.jailbreak | 26, 116, 149, 182 |
| abstract_inverted_index.mechanism | 108, 133 |
| abstract_inverted_index.overhead. | 141 |
| abstract_inverted_index.practical | 68 |
| abstract_inverted_index.prevalent | 52 |
| abstract_inverted_index.Post-aware | 82 |
| abstract_inverted_index.circumvent | 33 |
| abstract_inverted_index.deployment | 15 |
| abstract_inverted_index.developing | 177 |
| abstract_inverted_index.dimensions | 90, 175 |
| abstract_inverted_index.framework. | 84 |
| abstract_inverted_index.identifies | 87 |
| abstract_inverted_index.increasing | 14 |
| abstract_inverted_index.inference, | 126 |
| abstract_inverted_index.leveraging | 173 |
| abstract_inverted_index.negligible | 139 |
| abstract_inverted_index.real-world | 17, 115 |
| abstract_inverted_index.unintended | 56 |
| abstract_inverted_index.Adversarial | 79 |
| abstract_inverted_index.Calibration | 83 |
| abstract_inverted_index.adversarial | 44, 110 |
| abstract_inverted_index.calibration | 132 |
| abstract_inverted_index.concentrate | 113 |
| abstract_inverted_index.contrasting | 92 |
| abstract_inverted_index.demonstrate | 151 |
| abstract_inverted_index.dynamically | 86 |
| abstract_inverted_index.excessively | 61 |
| abstract_inverted_index.frameworks. | 166 |
| abstract_inverted_index.requirement | 6 |
| abstract_inverted_index.susceptible | 24 |
| abstract_inverted_index.undermining | 66 |
| abstract_inverted_index.Experimental | 142 |
| abstract_inverted_index.Furthermore, | 42 |
| abstract_inverted_index.Latent-space | 78 |
| abstract_inverted_index.considerable | 20 |
| abstract_inverted_index.construction | 100 |
| abstract_inverted_index.demonstrates | 169 |
| abstract_inverted_index.limitations, | 73 |
| abstract_inverted_index.over-defense | 57, 136 |
| abstract_inverted_index.particularly | 12 |
| abstract_inverted_index.advancements, | 21 |
| abstract_inverted_index.applications. | 18 |
| abstract_inverted_index.computational | 140 |
| abstract_inverted_index.effectiveness | 171 |
| abstract_inverted_index.inappropriate | 40 |
| abstract_inverted_index.significantly | 65 |
| abstract_inverted_index.disguise-based | 148 |
| abstract_inverted_index.training-based | 45 |
| abstract_inverted_index.embedding-level | 131 |
| abstract_inverted_index.safety-critical | 88, 174 |
| abstract_inverted_index.vulnerabilities | 31 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 5 |
| citation_normalized_percentile |