Urban, Matthias ; Nguyen, Duc Dat ; Binnig, Carsten
Hrsg.: Bordawekar, Rajesh ; Shmueli, Oded ; Amsterdamer, Yael ; Firmani, Donatella ; Kipf, Andreas (2023)
OmniscientDB: A Large Language Model-Augmented DBMS That Knows What Other DBMSs Do Not Know.
6th International Workshop on Exploiting Artificial Intelligence Techniques for Data Management (aiDM'23). Seattle, USA (18.06.2023)
doi: 10.1145/3593078.3593933
Konferenzveröffentlichung, Bibliographie
Kurzbeschreibung (Abstract)
In this paper, we present our vision of OmniscientDB, a novel database that leverages the implicitly-stored knowledge in large language models to augment datasets for analytical queries or even machine learning tasks. OmiscientDB empowers its users to augment their datasets by means of simple SQL queries and thus has the potential to dramatically reduce the manual overhead associated with data integration. It uses automatic prompt engineering to construct appropriate prompts for given SQL queries and passes them to a large language model like GPT-3 to contribute additional data (i.e., new rows, columns, or entire tables), augmenting the explicitly stored data. Our initial evaluation demonstrates the general feasibility of our vision, explores different prompting techniques in greater detail, and points towards several directions for future research.
| Typ des Eintrags: | Konferenzveröffentlichung |
|---|---|
| Erschienen: | 2023 |
| Herausgeber: | Bordawekar, Rajesh ; Shmueli, Oded ; Amsterdamer, Yael ; Firmani, Donatella ; Kipf, Andreas |
| Autor(en): | Urban, Matthias ; Nguyen, Duc Dat ; Binnig, Carsten |
| Art des Eintrags: | Bibliographie |
| Titel: | OmniscientDB: A Large Language Model-Augmented DBMS That Knows What Other DBMSs Do Not Know |
| Sprache: | Englisch |
| Publikationsjahr: | 20 Juni 2023 |
| Verlag: | ACM |
| Buchtitel: | Proceedings of the Sixth International Workshop on Exploiting Artificial Intelligence Techniques for Data Management |
| Veranstaltungstitel: | 6th International Workshop on Exploiting Artificial Intelligence Techniques for Data Management (aiDM'23) |
| Veranstaltungsort: | Seattle, USA |
| Veranstaltungsdatum: | 18.06.2023 |
| DOI: | 10.1145/3593078.3593933 |
| Kurzbeschreibung (Abstract): | In this paper, we present our vision of OmniscientDB, a novel database that leverages the implicitly-stored knowledge in large language models to augment datasets for analytical queries or even machine learning tasks. OmiscientDB empowers its users to augment their datasets by means of simple SQL queries and thus has the potential to dramatically reduce the manual overhead associated with data integration. It uses automatic prompt engineering to construct appropriate prompts for given SQL queries and passes them to a large language model like GPT-3 to contribute additional data (i.e., new rows, columns, or entire tables), augmenting the explicitly stored data. Our initial evaluation demonstrates the general feasibility of our vision, explores different prompting techniques in greater detail, and points towards several directions for future research. |
| Zusätzliche Informationen: | Art.No.: 4 |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Data and AI Systems |
| Hinterlegungsdatum: | 24 Jul 2023 13:03 |
| Letzte Änderung: | 25 Jul 2023 16:26 |
| PPN: | 50991800X |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum