Completion Tableau Taches avec les données des Benchs

biblio/Bench.csv

@@ -10,7 +10,7 @@ Bright,https://arxiv.org/abs/2407.12883v4,"https://github.com/xlang-ai/BRIGHT
 
 https://huggingface.co/datasets/xlangai/BRIGHT","- 7 BDD StackExchange fait main
 - TheoremQA + AoPS (Pb de maths + solutions récoltés)
-- 2 BDD codage",,Tâche reasoning intensive -> pas ce qu'on veut
+- 2 BDD codage : pony, LeetCode",,Tâche reasoning intensive -> pas ce qu'on veut
 BEIR,https://arxiv.org/abs/2104.08663v4,"https://github.com/beir-cellar/beir
 
 https://huggingface.co/BeIR
@@ -27,7 +27,11 @@ https://huggingface.co/BeIR
 -
 
 ",,Zero-shot learning en IR
-MS-MARCO,https://arxiv.org/abs/1611.09268,https://huggingface.co/datasets/microsoft/ms_marco,,"1M questions 
+MS-MARCO,https://arxiv.org/abs/1611.09268,https://huggingface.co/datasets/microsoft/ms_marco,"Dataset construit à partir de :
+- log de recherche BING- de documents sélectionné automatiquement par BING
+- d'extraits de ces documents obtenu par annotateur
+- de réponse rédigée par annotateur
+- classification de la question en fonction de la nature de la réponse","1M questions 
 8,841,823 passages issus de 3,563,535 pages web
 182,669 human generated answers
 

biblio/Taches.csv

 Intitulé,Benchs,Datasets,Archis/ Framework,Publis,Descriptions
-Open Domain Question Answering,,,DPR,"[Lewis et al., 2020]"
-Question Answering,KILT,"Natural Question, TriviaQA, HotpotQA","PaperQA, SearChain",,Trouver les extraits attendus contenant la réponse
-"Multi-hop question answering",,"HotpotQA, Musique, WikiMultiHopQA, StartegyQA",SearChain,"terme issu de [Xu et al., 2024] - SearChain", 
+Open Domain Question Answering,,,DPR,"[Lewis et al., 2020]",Trouver l'extrait contennant la réponse dans une base de connaissance (pas qu'un seul document) [def de BEIR]
+Question Answering,"KILT, BEIR","Natural Question, TriviaQA, HotpotQA, FiQA-2018","PaperQA, SearChain",,Trouver les extraits attendus contenant la réponse
+Multi-hop question answering,,"HotpotQA, Musique, WikiMultiHopQA, StartegyQA",SearChain,"terme issu de [Xu et al., 2024] - SearChain"," "
+Long-form QA,,ELI5,SearChain
 Summarization,,,PaperQA
 Détection de contradiction,,,PaperQA
 Jeopardy Question Generation,,,,"[Lewis et al.,2020]"
-Fact Checking,KILT,FEVER,"ATLAS, SearChain"
+Fact Checking,"KILT, BEIR","FEVER, Climate-FEVER, SciFact","ATLAS, SearChain",,"""verifies a claim against a big collection of evidence"" (BEIR)"
 Slot Filling,KILT,"Zero shot RE (zsRE), T-REx","ATLAS, SearChain"
-Long-form QA,,ELI5,SearChain
-Entity Linking,KILT,AIDA CoNLL-YAGO,ATLAS
+Entity Linking,KILT,"AIDA CoNLL-YAGO, WNED-WIKI, WNED-CWEB",ATLAS
+Entity Retrieval,BEIR,DBPedia,,,"""retrieving unique Wikipedia pages to entities mentioned in the query"" (BEIR) /!\ similaire"
 Dialogue,KILT,Wizards of wikipedia,ATLAS
+Citation Prediction,BEIR,SCIDOCS,,,"""retrieve cited papers (output) for a given paper title as input"" (BEIR)"
+Information Retrieval,"BIRCO, MS-MARCO","DORIS-MAE (Reseach), WhatsThatBook (Book description), Clinical-Trial (clinical trial description), RECLIC (litterature citation filling) ",,,Fournir les docuements conrrespondant à la query
+Duplicate Question Retrieval,BEIR,"Quora, CQADupStack",,,"""identifying duplicate questions asked in community question answering (cQA) forums"" (BEIR)
+input : question
+ouput : duplicate question in the forum​
+"
+Argument Retrieval,"BEIR, BIRCO","Touche-2020, ArgAna",,,"""ranking argumentative texts in a collection of focused arguments
+(output) in order of their relevance to a textual query (input) on different topics"" (BEIR)"
+News Retrieval,BEIR,"TREC-NEWS, Robust04",,,"""Given a news headline, we retrieve relevant news articles that provide important context or background information."" (BEIR)"
+Tweet Retrieval,BEIR,Signal-1M,,,"""retrieves relevant tweets for a given news article title"" (BEIR)"
+Bio-medical Information Retrieval,BEIR,"TREC-COVID, BioASQ, NFCorpus",,,"""searching relevant scientific documents such as research papers or blogs for a given scientific query in the biomedical domain"" (BEIR)"
+reasoning-intensive retrieval,Bright,"StackExchange, Pony (code), LeetCode, TheoremQA, AoPS",,,"Tache proche du OD-QA mais ""positive documents only provide useful information to help users derive an answer"" (Bright)"
\ No newline at end of file