Paper Reading AI Learner

Multilingual ColBERT-X

2022-09-03 06:02:52
Dawn Lawrie, Eugene Yang, Douglas W. Oard, James Mayfield

Abstract

ColBERT-X is a dense retrieval model for Cross Language Information Retrieval (CLIR). In CLIR, documents are written in one natural language, while the queries are expressed in another. A related task is multilingual IR (MLIR) where the system creates a single ranked list of documents written in many languages. Given that ColBERT-X relies on a pretrained multilingual neural language model to rank documents, a multilingual training procedure can enable a version of ColBERT-X well-suited for MLIR. This paper describes that training procedure. An important factor for good MLIR ranking is fine-tuning XLM-R using mixed-language batches, where the same query is matched with documents in different languages in the same batch. Neural machine translations of MS MARCO passages are used to fine-tune the model.

Abstract (translated)

URL

https://arxiv.org/abs/2209.01335

PDF

https://arxiv.org/pdf/2209.01335.pdf


Tags
3D Action Action_Localization Action_Recognition Activity Adversarial Agent Attention Autonomous Bert Boundary_Detection Caption Chat Classification CNN Compressive_Sensing Contour Contrastive_Learning Deep_Learning Denoising Detection Dialog Diffusion Drone Dynamic_Memory_Network Edge_Detection Embedding Embodied Emotion Enhancement Face Face_Detection Face_Recognition Facial_Landmark Few-Shot Gait_Recognition GAN Gaze_Estimation Gesture Gradient_Descent Handwriting Human_Parsing Image_Caption Image_Classification Image_Compression Image_Enhancement Image_Generation Image_Matting Image_Retrieval Inference Inpainting Intelligent_Chip Knowledge Knowledge_Graph Language_Model Matching Medical Memory_Networks Multi_Modal Multi_Task NAS NMT Object_Detection Object_Tracking OCR Ontology Optical_Character Optical_Flow Optimization Person_Re-identification Point_Cloud Portrait_Generation Pose Pose_Estimation Prediction QA Quantitative Quantitative_Finance Quantization Re-identification Recognition Recommendation Reconstruction Regularization Reinforcement_Learning Relation Relation_Extraction Represenation Represenation_Learning Restoration Review RNN Salient Scene_Classification Scene_Generation Scene_Parsing Scene_Text Segmentation Self-Supervised Semantic_Instance_Segmentation Semantic_Segmentation Semi_Global Semi_Supervised Sence_graph Sentiment Sentiment_Classification Sketch SLAM Sparse Speech Speech_Recognition Style_Transfer Summarization Super_Resolution Surveillance Survey Text_Classification Text_Generation Tracking Transfer_Learning Transformer Unsupervised Video_Caption Video_Classification Video_Indexing Video_Prediction Video_Retrieval Visual_Relation VQA Weakly_Supervised Zero-Shot