{"id":40975,"date":"2025-10-01T15:53:44","date_gmt":"2025-10-01T13:53:44","guid":{"rendered":"https:\/\/www.l3s.de\/?page_id=40975"},"modified":"2025-11-20T10:04:28","modified_gmt":"2025-11-20T09:04:28","slug":"thirdwave","status":"publish","type":"page","link":"https:\/\/www.l3s.de\/de\/research-at-l3s\/all-projects\/thirdwave\/","title":{"rendered":"THIRDWAVE"},"content":{"rendered":"
\n \n
\n \n \n
\n \n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n
\n \n \n \"\"\n
<\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n \n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n \n
\n

Start:<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n \n
\n

01.11.2025<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n \n
\n

Ende:<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n \n
\n

31.10.2029<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n
\n \n
\n
\n \n \n
\n \n
\n \n \n
\n

Webseite:<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n
\n \n \n
\n \n
\n \n
thirdwave.l3s.uni-hannover.de<\/span><\/a>\n <\/span>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n
\n
\n \n \n
\n \n
\n \n <\/div>\n
\n \n
\n \n
\n
\n \n \n
\n \n
\n \n
\n \n \n

THIRDWAVE<\/h2>\n <\/div>\n <\/div>\n
\n \n <\/div>\n
\n \n \n
\n

THIRDWAVE baut ein internationales, interdisziplin\u00e4res Netzwerk auf, um neurosymbolische Grundlagenmodelle voranzubringen. Durch die Integration von Large Language Models (LLMs), Knowledge Graphs (KGs) und symbolischem Schlussfolgern erm\u00f6glicht THIRDWAVE vertrauensw\u00fcrdige, erkl\u00e4rbare und dom\u00e4nenspezifische KI-Systeme. Das Projekt befasst sich mit Herausforderungen in den Bereichen Medizin, Lebensmittelsysteme, Geodaten und Bildung durch Entsendungen und gemeinsame Forschung. Die vier Hauptziele dieses Projekts sind:<\/p>\n <\/div>\n <\/div>\n

\n \n \n
\n

\n
\n <\/p>\n

LLMs verstehen:<\/strong><\/p>\n

Analyse von F\u00e4higkeiten und Grenzen zur Verbesserung von Leistung, Benutzerfreundlichkeit und Vertrauensw\u00fcrdigkeit.<\/p>\n

\n
\n <\/p>\n

LLMs verbessern:<\/strong><\/p>\n

Verbesserung von Fairness, sachlicher Genauigkeit und Robustheit durch externe Wissensquellen und\nmenschlicher Zusammenarbeit.<\/p>\n

\n
\n <\/p>\n

Weiterentwicklung der LLM-gesteuerten neurosymbolischen KI:<\/strong><\/p>\n

Entwicklung hybrider Systeme, die LLMs mit symbolischem Denken f\u00fcr strukturierte Wissensdarstellung und bessere Entscheidungsunterst\u00fctzung kombinieren.<\/p>\n

\n
\n <\/p>\n

Anwendungsf\u00e4lle und Bewertung:<\/strong><\/p>\n

Anwendung von LLM-gesteuerter neurosymbolischer KI in den Bereichen Gesundheitswesen, Bildung, Geodaten und Lebensmittelinformationstechnik, Validierung der Skalierbarkeit und gesellschaftlichen Auswirkungen.\n
\n <\/p>\n <\/div>\n <\/div>\n

\n \n <\/div>\n
\n \n
\n \n \n F\u00f6rderprogramm<\/span>\n
<\/path><\/svg><\/span><\/div>\n
<\/path><\/svg><\/span><\/div>\n <\/a>\n
\n
\n \n
\n \n \n
\n

Horizon Europe \u2014 Marie Sk\u0142odowska-Curie Actions (MSCA) Staff Exchanges fund<\/p>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n <\/div>\n

\n \n \n Projektpartner<\/span>\n
<\/path><\/svg><\/span><\/div>\n
<\/path><\/svg><\/span><\/div>\n <\/a>\n
\n
\n \n
\n \n \n
\n