Question Answering On Webquestions

Metrics

Results

Performance results of various models on this benchmark

Model Name	EM	Paper Title	Repository
PaLM-540B (Zero-Shot)	10.6	PaLM: Scaling Language Modeling with Pathways
GPT-3-175B (Few-Shot)	41.5	Language Models are Few-Shot Learners
PaLM 2-S (one-shot)	21.8	PaLM 2 Technical Report
ToT	26.3	Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models
ToT	26.3	Tree of Thoughts: Deliberate Problem Solving with Large Language Models
React	38.3	ReAct: Synergizing Reasoning and Acting in Language Models
DSP	59.4	Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models
REALM	40.7	REALM: Retrieval-Augmented Language Model Pre-Training
CoT	42.5	Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
GPT-3-175B (Zero-Shot)	14.4	Language Models are Few-Shot Learners
CoT	42.5	Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models
Memory Networks (ensemble)	-	Large-scale Simple Question Answering with Memory Networks
PaLM 2-L (one-shot)	28.2	PaLM 2 Technical Report
T5.1.1-XXL+SSM	42.8	Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
PaLM-540B (One-Shot)	22.6	PaLM: Scaling Language Modeling with Pathways
GPT-3-175B (One-Shot)	25.3	Language Models are Few-Shot Learners
FiE+PAQ	56.3	FiE: Building a Global Probability Space by Leveraging Early Fusion in Encoder for Open-Domain Question Answering	-
DPR	42.4	Dense Passage Retrieval for Open-Domain Question Answering
Self-Ask	31.1	Measuring and Narrowing the Compositionality Gap in Language Models
Subgraph embeddings	-	Question Answering with Subgraph Embeddings

0 of 37 row(s) selected.