SIB-200: A Large-Scale Multilingual Dataset for Topic Classification in 200+ Languages and Dialects
다양한 언어 및 방양에 대한 주제 분류를 위한 SIB-200의 중요성
PEARL: A Review-driven Persona-Knowledge grounded Conversational Recommendation Dataset
Conversational recommender systems benefit from PEARL's persona and knowledge augmentation, enhancing user preferences and recommendations.
KIWI: A Dataset of Knowledge-Intensive Writing Instructions for Answering Research Questions
Current large language models struggle to effectively follow user instructions for revising long-form answers, as shown by the KIWI dataset analysis.
Multimodal ArXiv: Enhancing LVLMs Scientific Comprehension with ArXivCap and ArXivQA
Multimodal ArXiv introduces ArXivCap and ArXivQA to improve LVLMs' understanding of scientific figures, enhancing mathematical reasoning capabilities.
JMultiWOZ: Japanese Multi-Domain Task-Oriented Dialogue Dataset
JMultiWOZ is the first large-scale Japanese multi-domain task-oriented dialogue dataset, providing benchmarks for dialogue state tracking and response generation.
EgoExoLearn: Bridging Asynchronous Procedural Activities in Real World
人間の行動を異なる視点から結びつけるための重要なリソースであるEgoExoLearnデータセットの紹介とその潜在的な応用に焦点を当てる。
EgoExoLearn: Bridging Asynchronous Procedural Activities in Real World
EgoExoLearn dataset explores human ability to bridge asynchronous procedural actions from different viewpoints.
MuSHRoom: Multi-Sensor Hybrid Room Dataset for Joint 3D Reconstruction and Novel View Synthesis
提案されたMuSHRoomデータセットは、消費者向けデバイスで収集された入力を使用して、3D再構築の精度と新しい視点合成の品質を評価するためのベンチマークを提供します。
Entity6K: A Comprehensive Dataset for Real-World Entity Recognition
Entity6K introduces a diverse dataset for real-world entity recognition, addressing the lack of suitable evaluation datasets in open-domain settings.
Entity6K: A Comprehensive Dataset for Real-World Entity Recognition
Entity6K introduces a comprehensive dataset for real-world entity recognition, addressing the lack of suitable evaluation datasets in this field.