Semantic Parsing On Spider

Metriken

Accuracy

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

		Paper Title
RESDSQL-3B + NatSQL	84.1	RESDSQL: Decoupling Schema Linking and Skeleton Parsing for Text-to-SQL
code-davinci-002 175B (LEVER)	81.9	LEVER: Learning to Verify Language-to-Code Generation with Execution
RASAT+PICARD	75.5	RASAT: Integrating Relational Structures into Pretrained Seq2Seq Model for Text-to-SQL
Graphix-3B + PICARD	74.0	Graphix-T5: Mixing Pre-Trained Transformers with Graph-Aware Layers for Text-to-SQL Parsing
T5-3B + PICARD	71.9	PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models
SADGA + GAP	70.1	SADGA: Structure-Aware Dual Graph Aggregation Network for Text-to-SQL
RATSQL + GAP	69.7	Learning Contextual Representations for Semantic Parsing with Generation-Augmented Pre-Training
RATSQL + Grammar-Augmented Pre-Training	69.6	GraPPa: Grammar-Augmented Pre-Training for Table Semantic Parsing
RATSQL + BERT	65.6	RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers
Exact Set Matching	19.7	Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task

0 of 10 row(s) selected.