SOTAVerified

Fact Selection

A task where an agent should select at most two sentences from the paper as argumentative facts.

Papers

Showing 1–14 of 14 papers

TitleStatusHype
FaiRR: Faithful and Robust Deductive Reasoning over Natural LanguageCode1
COVID-VTS: Fact Extraction and Verification on Short Video PlatformsCode1
Simple Question Answering with Subgraph Ranking and Joint-Scoring—0
The combination of context information to enhance simple question answering—0
FaiRR: Faithful and Robust Deductive Reasoning over Natural Language—0
CausalR: Causal Reasoning over Natural Language Rulebases—0
Proto-Gen: An end-to-end neural generator for persona and knowledge grounded response generation—0
Pattern-revising Enhanced Simple Question Answering over Knowledge Bases—0
Simple Question Answering by Attentive Convolutional Neural Network—0
UFO: Unified Fact Obtaining for Commonsense Question AnsweringCode0
Expository Text Generation: Imitate, Retrieve, ParaphraseCode0
Integrating Randomness in Large Language Models: A Linear Congruential Generator Approach for Generating Clinically Relevant ContentCode0
KG-CRuSE: Recurrent Walks over Knowledge Graph for Explainable Conversation Reasoning using Semantic EmbeddingsCode0
ArgSciChat: A Dataset for Argumentative Dialogues on Scientific PapersCode0
Show:102550

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1TF-IDFFact-F116.22—Unverified
2S-BERTFact-F113.65—Unverified
3LED(Q,P)Fact-F110.58—Unverified
4LED(Q,P,H)Fact-F18.5—Unverified