This paper presents contributions in two directions: first we propose a new system for Frame Identification (FI), based on pre-trained text encoders trained discriminatively and graphs embedding, producing state of the art performance and, second, we take in consideration all the extremely different procedures used to evaluate systems for this task performing a complete evaluation over two benchmarks and all possible splits and cleaning procedures used in the FI literature.
Combining ELECTRA and Adaptive Graph Encoding for Frame Identification / Tamburini Fabio. - ELETTRONICO. - (2022), pp. 1671-1679. (Intervento presentato al convegno Thirteenth Language Resources and Evaluation Conference tenutosi a Marseille, France nel 20-25 June 2022).
Combining ELECTRA and Adaptive Graph Encoding for Frame Identification
Tamburini Fabio
2022
Abstract
This paper presents contributions in two directions: first we propose a new system for Frame Identification (FI), based on pre-trained text encoders trained discriminatively and graphs embedding, producing state of the art performance and, second, we take in consideration all the extremely different procedures used to evaluate systems for this task performing a complete evaluation over two benchmarks and all possible splits and cleaning procedures used in the FI literature.File | Dimensione | Formato | |
---|---|---|---|
2022.lrec-1.178.pdf
accesso aperto
Descrizione: Contributo in Atti di Convegno
Tipo:
Versione (PDF) editoriale
Licenza:
Licenza per Accesso Aperto. Creative Commons Attribuzione - Non commerciale (CCBYNC)
Dimensione
787.41 kB
Formato
Adobe PDF
|
787.41 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.