Method and device for generating adversarial text
The invention discloses an adversarial text generation method and device, and belongs to the technical field of artificial intelligence. The method comprises the steps of obtaining a to-be-disturbed text; determining a to-be-disturbed text segment of the to-be-disturbed text, wherein the to-be-distu...
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Patent |
| Sprache: | chi ; eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | DUAN HAIXIN ZHU ANNAN ZHANG JIA DENG CHUYUN WAN RONGFEI LIU MINGXUAN |
| description | The invention discloses an adversarial text generation method and device, and belongs to the technical field of artificial intelligence. The method comprises the steps of obtaining a to-be-disturbed text; determining a to-be-disturbed text segment of the to-be-disturbed text, wherein the to-be-disturbed text segment comprises a plurality of words; executing perturbation operation on part or all of the words in the multiple words of the to-be-perturbed text segment to obtain a perturbed text segment of the to-be-perturbed text segment; and obtaining an adversarial text of the to-be-disturbed text based on the disturbing text According to the method and the device, one or more words in the to-be-disturbed text segment can be transformed, and the text segments with various lengths can be uniformly considered to execute the disturbance operation, so that the diversity of disturbance modes is ensured, and abundant and diversified adversarial texts can be generated according to the disturbance modes.
本申请公开了一种对抗文本的生 |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN117252156A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN117252156A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN117252156A3</originalsourceid><addsrcrecordid>eNrjZDD0TS3JyE9RSMxLUUhJLctMTlVIyy9SSE_NSy1KLMnMS1dITClLLSpOLMpMzFEoSa0o4WFgTUvMKU7lhdLcDIpuriHOHrqpBfnxqcUFiclAvSXxzn6GhuZGpkaGpmaOxsSoAQCEAiuW</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Method and device for generating adversarial text</title><source>esp@cenet</source><creator>DUAN HAIXIN ; ZHU ANNAN ; ZHANG JIA ; DENG CHUYUN ; WAN RONGFEI ; LIU MINGXUAN</creator><creatorcontrib>DUAN HAIXIN ; ZHU ANNAN ; ZHANG JIA ; DENG CHUYUN ; WAN RONGFEI ; LIU MINGXUAN</creatorcontrib><description>The invention discloses an adversarial text generation method and device, and belongs to the technical field of artificial intelligence. The method comprises the steps of obtaining a to-be-disturbed text; determining a to-be-disturbed text segment of the to-be-disturbed text, wherein the to-be-disturbed text segment comprises a plurality of words; executing perturbation operation on part or all of the words in the multiple words of the to-be-perturbed text segment to obtain a perturbed text segment of the to-be-perturbed text segment; and obtaining an adversarial text of the to-be-disturbed text based on the disturbing text According to the method and the device, one or more words in the to-be-disturbed text segment can be transformed, and the text segments with various lengths can be uniformly considered to execute the disturbance operation, so that the diversity of disturbance modes is ensured, and abundant and diversified adversarial texts can be generated according to the disturbance modes.
本申请公开了一种对抗文本的生</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTING ; COUNTING ; ELECTRIC DIGITAL DATA PROCESSING ; PHYSICS</subject><creationdate>2023</creationdate><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20231219&DB=EPODOC&CC=CN&NR=117252156A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,777,882,25545,76296</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20231219&DB=EPODOC&CC=CN&NR=117252156A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>DUAN HAIXIN</creatorcontrib><creatorcontrib>ZHU ANNAN</creatorcontrib><creatorcontrib>ZHANG JIA</creatorcontrib><creatorcontrib>DENG CHUYUN</creatorcontrib><creatorcontrib>WAN RONGFEI</creatorcontrib><creatorcontrib>LIU MINGXUAN</creatorcontrib><title>Method and device for generating adversarial text</title><description>The invention discloses an adversarial text generation method and device, and belongs to the technical field of artificial intelligence. The method comprises the steps of obtaining a to-be-disturbed text; determining a to-be-disturbed text segment of the to-be-disturbed text, wherein the to-be-disturbed text segment comprises a plurality of words; executing perturbation operation on part or all of the words in the multiple words of the to-be-perturbed text segment to obtain a perturbed text segment of the to-be-perturbed text segment; and obtaining an adversarial text of the to-be-disturbed text based on the disturbing text According to the method and the device, one or more words in the to-be-disturbed text segment can be transformed, and the text segments with various lengths can be uniformly considered to execute the disturbance operation, so that the diversity of disturbance modes is ensured, and abundant and diversified adversarial texts can be generated according to the disturbance modes.
本申请公开了一种对抗文本的生</description><subject>CALCULATING</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>ELECTRIC DIGITAL DATA PROCESSING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2023</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZDD0TS3JyE9RSMxLUUhJLctMTlVIyy9SSE_NSy1KLMnMS1dITClLLSpOLMpMzFEoSa0o4WFgTUvMKU7lhdLcDIpuriHOHrqpBfnxqcUFiclAvSXxzn6GhuZGpkaGpmaOxsSoAQCEAiuW</recordid><startdate>20231219</startdate><enddate>20231219</enddate><creator>DUAN HAIXIN</creator><creator>ZHU ANNAN</creator><creator>ZHANG JIA</creator><creator>DENG CHUYUN</creator><creator>WAN RONGFEI</creator><creator>LIU MINGXUAN</creator><scope>EVB</scope></search><sort><creationdate>20231219</creationdate><title>Method and device for generating adversarial text</title><author>DUAN HAIXIN ; ZHU ANNAN ; ZHANG JIA ; DENG CHUYUN ; WAN RONGFEI ; LIU MINGXUAN</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN117252156A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2023</creationdate><topic>CALCULATING</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>ELECTRIC DIGITAL DATA PROCESSING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>DUAN HAIXIN</creatorcontrib><creatorcontrib>ZHU ANNAN</creatorcontrib><creatorcontrib>ZHANG JIA</creatorcontrib><creatorcontrib>DENG CHUYUN</creatorcontrib><creatorcontrib>WAN RONGFEI</creatorcontrib><creatorcontrib>LIU MINGXUAN</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>DUAN HAIXIN</au><au>ZHU ANNAN</au><au>ZHANG JIA</au><au>DENG CHUYUN</au><au>WAN RONGFEI</au><au>LIU MINGXUAN</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Method and device for generating adversarial text</title><date>2023-12-19</date><risdate>2023</risdate><abstract>The invention discloses an adversarial text generation method and device, and belongs to the technical field of artificial intelligence. The method comprises the steps of obtaining a to-be-disturbed text; determining a to-be-disturbed text segment of the to-be-disturbed text, wherein the to-be-disturbed text segment comprises a plurality of words; executing perturbation operation on part or all of the words in the multiple words of the to-be-perturbed text segment to obtain a perturbed text segment of the to-be-perturbed text segment; and obtaining an adversarial text of the to-be-disturbed text based on the disturbing text According to the method and the device, one or more words in the to-be-disturbed text segment can be transformed, and the text segments with various lengths can be uniformly considered to execute the disturbance operation, so that the diversity of disturbance modes is ensured, and abundant and diversified adversarial texts can be generated according to the disturbance modes.
本申请公开了一种对抗文本的生</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN117252156A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTING COUNTING ELECTRIC DIGITAL DATA PROCESSING PHYSICS |
| title | Method and device for generating adversarial text |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-20T11%3A58%3A13IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=DUAN%20HAIXIN&rft.date=2023-12-19&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN117252156A%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |