Update config_sentence_transformers.json

config_sentence_transformers.json

@@ -1,241 +1,8 @@
 {
-  "prompts": {
-    "STS": "Retrieve semantically similar text.",
-    "BitextMining": "Retrieve translations of the following text.",
-    "Classification": "Classify the topic of the given text.",
-    "MultilabelClassification": "Classify the topic of the given text.",
-    "Clustering": "Classify the topic of the given text.",
-    "PairClassification": "Retrieve semantically entailed text.",
-    "query": "Given a question, retrieve passages that can help answer the question.",
-    "document": "",
-    "BulgarianStoreReviewSentimentClassfication": "Classify the sentiment of the given text.",
-    "CzechProductReviewSentimentClassification": "Classify the sentiment of the given text.",
-    "FinancialPhrasebankClassification": "Classify the sentiment of the given text.",
-    "PoemSentimentClassification": "Classify the sentiment of the given text.",
-    "EstonianValenceClassification": "Classify the sentiment of the given text.",
-    "FilipinoShopeeReviewsClassification": "Classify the sentiment of the given text.",
-    "SentimentAnalysisHindi": "Classify the sentiment of the given text.",
-    "KurdishSentimentClassification": "Classify the sentiment of the given text.",
-    "MacedonianTweetSentimentClassification": "Classify the sentiment of the given text.",
-    "AfriSentiClassification": "Classify the sentiment of the given text.",
-    "CataloniaTweetClassification": "Classify the sentiment of the given text.",
-    "MultiHateClassification": "Classify the sentiment of the given text.",
-    "NusaParagraphEmotionClassification": "Classify the sentiment of the given text.",
-    "NusaX-senti": "Classify the sentiment of the given text.",
-    "SwissJudgementClassification": "Classify the sentiment of the given text.",
-    "PolEmo2.0-OUT": "Classify the sentiment of the given text.",
-    "CSFDSKMovieReviewSentimentClassification": "Classify the sentiment of the given text.",
-    "SlovakMovieReviewSentimentClassification": "Classify the sentiment of the given text.",
-    "CyrillicTurkicLangClassification": "Classify the text into its language.",
-    "IndicLangClassification": "Classify the text into its language.",
-    "NordicLangClassification": "Classify the text into its language.",
-    "ScalaClassification": "Classify the given sentence as linguistically acceptable or not acceptable.",
-    "DalajClassification": "Classify the given sentence as linguistically acceptable or not acceptable.",
-    "ToxicConversationsClassification": "Classify the given comments as either toxic or not toxic.",
-    "IndonesianIdClickbaitClassification": "Classify the given text as either clickbait or not clickbait.",
-    "KorSarcasmClassification": "Classify the given text as either sarcasm or not sarcasm.",
-    "AmazonCounterfactualClassification": "Classify a given Amazon customer review text as either counterfactual or not counterfactual.",
-    "MassiveIntentClassification": "Given a user utterance as query, find the user intents.",
-    "PAC": "Classify the given clause as either abusive or not abusive.",
-    "Banking77Classification": "Given an online banking query, find the corresponding intents.",
-    "ImdbClassification": "Classify the sentiment expressed in the given movie review text from the IMDB dataset.",
-    "MTOPDomainClassification": "Classify the intent domain of the given utterance in task-oriented conversation.",
-    "MassiveScenarioClassification": "Given a user utterance as query, find the user scenarios.",
-    "TweetSentimentExtractionClassification": "Classify the sentiment of a given tweet as either positive, negative, or neutral",
-    "WikiCitiesClustering": "Classify the following city description by country.",
-    "ArXivHierarchicalClusteringP2P": "Identify the main and secondary category of arXiv papers based on the titles and abstracts.",
-    "ArXivHierarchicalClusteringS2S": "Identify the main and secondary category of arXiv papers based on the titles.",
-    "BiorxivClusteringP2P.v2": "Identify the main category of bioRxiv papers based on the titles and abstracts.",
-    "MedrxivClusteringP2P.v2": "Identify the main category of medRxiv papers based on the titles and abstracts.",
-    "StackExchangeClustering.v2": "Identify the topic or theme of StackExchange posts based on the titles.",
-    "StackExchangeClusteringP2P.v2": "Identify the topic or theme of StackExchange posts based on the given paragraphs.",
-    "MedrxivClusteringS2S.v2": "Identify the main category of medRxiv papers based on the titles.",
-    "TwentyNewsgroupsClustering.v2": "Identify the topic or theme of the given news articles.",
-    "SprintDuplicateQuestions": "Retrieve duplicate questions from Sprint forum.",
-    "TwitterURLCorpus": "Retrieve tweets that are semantically similar to the given tweet.",
-    "ArmenianParaphrasePC": "Retrieve paraphrases of the given sentence.",
-    "OpusparcusPC": "Retrieve paraphrases of the given sentence.",
-    "PawsXPairClassification": "Retrieve paraphrases of the given sentence.",
-    "PpcPC": "Retrieve paraphrases of the given sentence.",
-    "TwitterSemEval2015": "Retrieve tweets that are semantically similar to the given tweet.",
-    "KorHateSpeechMLClassification": "Classify the sentiment of the given text.",
-    "CEDRClassification": "Classify the emotion expressed in the given comment into: joy, sadness, surprise, fear, and anger.",
-    "SummEvalSummarization.v2": "Given a news summary, retrieve other semantically similar summaries.",
-    "AppsRetrieval": "Retrieve the most relevant code snippet for the given query.",
-    "CodeEditSearchRetrieval": "Retrieve the most relevant code edit.",
-    "CodeFeedbackMT": "Retrieve the most relevant response for the given query.",
-    "CodeFeedbackST": "Retrieve the most relevant response for the given query.",
-    "CodeSearchNetCCRetrieval": "Retrieve the most relevant code snippet for the given code snippet.",
-    "CodeSearchNetRetrieval": "Retrieve the most relevant code snippet for the given query.",
-    "CodeTransOceanContest": "Retrieve similar code to the given source code.",
-    "CodeTransOceanDL": "Retrieve similar code to the given source code.",
-    "CosQA": "Retrieve the most relevant code snippet for the given query.",
-    "COIRCodeSearchNetRetrieval": "Retrieve the most relevant code summary for the given code snippet.",
-    "StackOverflowQA": "Retrieve the most relevant response for the given query.",
-    "SyntheticText2SQL": "Retrieve the most relevant sql code snippet for the given query.",
-    "AILAStatutes": "Identify the most relevant statutes for the given situation.",
-    "ArguAna": "Given a claim, find documents that refute the claim.",
-    "LegalBenchCorporateLobbying": "Given a bill title, retrieve the corresponding bill summary.",
-    "SCIDOCS": "Given a scientific paper title, retrieve paper abstracts that are cited by the given paper.",
-    "TRECCOVID": "Given a query on COVID-19, retrieve documents that answer the query.",
-    "CovidRetrieval": "Given a query on COVID-19, retrieve documents that answer the query.",
-    "CQADupstackGamingRetrieval": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackUnixRetrieval": "Given a question, retrieve questions that are semantically equivalent.",
-    "ClimateFEVERHardNegatives": "Given a claim about climate change, retrieve documents that support or refute the claim.",
-    "FEVERHardNegatives": "Given a claim, retrieve documents that support or refute the claim.",
-    "FiQA2018": "Given a financial question, retrieve passages that answer the question.",
-    "HotpotQAHardNegatives": "Given a multi-hop question, retrieve passages that answer the question.",
-    "Touche2020Retrieval.v3": "Given a question, retrieve passages that answer the question.",
-    "WebLINXCandidatesReranking": "Given a web navigation step, retrieve relevant elements.",
-    "AskUbuntuDupQuestions": "Retrieve duplicate questions from AskUbuntu forum.",
-    "MindSmallReranking": "Retrieve relevant news articles based on user browsing history.",
-    "NFCorpus": "Given a question, retrieve passages that answer the question.",
-    "TRECCOVID-PL": "Given a query on COVID-19, retrieve documents that answer the query.",
-    "SciFact": "Given a scientific claim, retrieve passages that support or refute the claim.",
-    "SciFact-PL": "Given a scientific claim, retrieve passages that support or refute the claim.",
-    "CmedqaRetrieval": "Given a question, retrieve passages that answer the question.",
-    "CMedQAv2-reranking": "Given a question, retrieve passages that answer the question.",
-    "AngryTweetsClassification": "Classify the sentiment of the given text.",
-    "DanishPoliticalCommentsClassification": "Classify the sentiment of the given text.",
-    "DKHateClassification": "Classify the given comments as either offensive or not offensive.",
-    "LccSentimentClassification": "Classify the sentiment of the given text.",
-    "NoRecClassification": "Classify the sentiment of the given text.",
-    "NorwegianParliamentClassification": "Classify the sentiment of the given text.",
-    "SwedishSentimentClassification": "Classify the sentiment of the given text.",
-    "SweRecClassification": "Classify the sentiment of the given text.",
-    "DanFeverRetrieval": "Given a claim, retrieve documents that support or refute the claim.",
-    "SNLRetrieval": "Given a summary, retrieve the original article.",
-    "SwednRetrieval": "Given a summary, retrieve the original article.",
-    "TV2Nordretrieval": "Given a news summary, retrieve the original article.",
-    "BengaliSentimentAnalysis": "Classify the sentiment of the given text.",
-    "HindiDiscourseClassification": "Classify the given text into one of the five discourse modes: argumentative, narrative, descriptive, dialogic, and informative.",
-    "MTOPIntentClassification": "Classify the intent of the given utterance in task-oriented conversation.",
-    "TweetSentimentClassification": "Classify the sentiment of the given text.",
-    "UrduRomanSentimentClassification": "Classify the sentiment of the given text as either positive, negative, or neutral.",
-    "AmazonReviewsClassification": "Classify the given Amazon review into its appropriate rating category.",
-    "BlurbsClusteringP2P": "Classify the given book title and blurb into its genre.",
-    "BlurbsClusteringS2S": "Classify the given book title into its genre.",
-    "FalseFriendsGermanEnglish": "Retrieve translations of the following text.",
-    "XMarket": "Given a product name search, retrieve the corresponding product description.",
-    "GerDaLIR": "Retrieve documents that are referenced by the given text.",
-    "MLSUMClusteringP2P": "Classify the topic of the given news article.",
-    "SummEvalFr": "Given a news summary, retrieve other semantically similar summaries.",
-    "AllegroReviews": "Classify the sentiment of the given text.",
-    "CBD": "Classify the given text as either cyberbullying or not.",
-    "PolEmo2.0-IN": "Classify the sentiment of the given text.",
-    "PSC": "Retrieve semantically similar text.",
-    "EcomRetrieval": "Given a product name query, retrieve the corresponding product description.",
-    "MedicalRetrieval": "Retrieve the most relevant response for the given query.",
-    "VideoRetrieval": "Given a video search query, retrieve the titles of relevant videos.",
-    "CMedQAv1-reranking": "Retrieve the most relevant response for the given query.",
-    "Waimai": "Classify the sentiment of the given review as either positive or negative.",
-    "OnlineShopping": "Classify the sentiment of the given review as either positive or negative.",
-    "JDReview": "Classify the sentiment of the given review as either positive or negative.",
-    "MultilingualSentiment": "Classify the sentiment of the given review as either positive, negative, or neutral.",
-    "ToxicChatClassification": "Classify the given text as either toxic or not toxic.",
-    "JapaneseSentimentClassification": "Classify the sentiment of the given text.",
-    "WRIMEClassification": "Classify the sentiment of the given text.",
-    "NLPJournalTitleAbsRetrieval.V2": "Given a paper's title, retrieve the corresponding abstract.",
-    "NLPJournalTitleIntroRetrieval.V2": "Given a paper's title, retrieve the corresponding introduction.",
-    "NLPJournalAbsIntroRetrieval.V2": "Given a paper's abstract, retrieve the corresponding introduction.",
-    "NLPJournalAbsArticleRetrieval.V2": "Given a paper's abstract, retrieve the corresponding paper.",
-    "ESCIReranking": "Given a product name query, retrieve the corresponding product description.",
-    "DutchBookReviewSentimentClassification.v2": "Classify the sentiment of the given text.",
-    "VaccinChatNLClassification": "Classify the intent of the given utterance.",
-    "DutchColaClassification": "Classify the given sentence as linguistically acceptable or not acceptable.",
-    "DutchGovernmentBiasClassification": "Classify the given government document as biased or unbiased.",
-    "DutchSarcasticHeadlinesClassification": "Classify the given newspaper headline as sarcastic or not sarcastic.",
-    "XLWICNLPairClassification": "Retrieve semantically similar text.",
-    "CovidDisinformationNLMultiLabelClassification": "Classify the given social media post related to COVID-19 into its misinformation category.",
-    "VABBClusteringS2S": "Identify the main category of the given paper based on the title.",
-    "VABBClusteringP2P": "Identify the main category of the given paper based on the title and abstract.",
-    "ArguAna-NL.v2": "Given a claim, find documents that refute the claim.",
-    "SCIDOCS-NL.v2": "Given a scientific paper title, retrieve paper abstracts that are cited by the given paper.",
-    "SciFact-NL.v2": "Given a scientific claim, retrieve passages that support or refute the claim.",
-    "DutchNewsArticlesRetrieval": "Given a news title, retrieve the original article.",
-    "OpenTenderRetrieval": "Given a title, retrieve the corresponding article.",
-    "VABBRetrieval": "Given a paper's title, retrieve the corresponding abstract.",
-    "GeoreviewClassification": "Classify the given review into its appropriate rating category.",
-    "InappropriatenessClassification": "Classify the given message as either sensitive topic or not.",
-    "KinopoiskClassification": "Classify the sentiment of the given movie review.",
-    "RuReviewsClassification": "Classify the sentiment of the given review as either positive, negative, or neutral.",
-    "RuSciBenchGRNTIClassification": "Identify the main category of the given paper based on the title and abstract.",
-    "RuSciBenchOECDClassification": "Identify the main category of the given paper based on the title and abstract.",
-    "GeoreviewClusteringP2P": "Identify the organization category based on the given review.",
-    "RuSciBenchGRNTIClusteringP2P": "Identify the main category of the given paper based on the title and abstract.",
-    "RuSciBenchOECDClusteringP2P": "Identify the main category of the given paper based on the title and abstract.",
-    "SensitiveTopicsClassification": "Classify the given text into sensitive topics.",
-    "RiaNewsRetrievalHardNegatives.v2": "Given a news title, retrieve the original article.",
-    "PersianFoodSentimentClassification": "Classify the sentiment of the given text as either positive or negative.",
-    "SynPerChatbotConvSAClassification": "Classify the sentiment of the given text.",
-    "SynPerChatbotConvSAToneChatbotClassification": "Classify the sentiment of the given text.",
-    "SynPerChatbotConvSAToneUserClassification": "Classify the sentiment of the given text.",
-    "SynPerChatbotSatisfactionLevelClassification": "Classify the satisfaction level of the given text.",
-    "SynPerTextToneClassification.v3": "Classify the tone of the given text.",
-    "DeepSentiPers.v2": "Classify the sentiment of the given text.",
-    "PersianTextEmotion.v2": "Classify the emotion expressed in the given text into: joy, sadness, surprise, fear, anger, and love.",
-    "StyleClassification": "Classify the style of the given text as either formal or informal.",
-    "PerShopDomainClassification": "Classify the domain of the given utterance in shopping dialogue.",
-    "PerShopIntentClassification": "Classify the intent of the given utterance in shopping dialogue.",
-    "SynPerChatbotRAGFAQPC": "Retrieve the most relevant response for the given query.",
-    "FarsiParaphraseDetection": "Retrieve semantically similar text.",
-    "SynPerTextKeywordsPC": "Identify keywords in the given text.",
-    "SynPerQAPC": "Retrieve the most relevant response for the given query.",
-    "ParsinluQueryParaphPC": "Retrieve semantically similar text.",
-    "SynPerChatbotRAGFAQRetrieval": "Retrieve the most relevant response for the given query.",
-    "HotpotQA-FaHardNegatives": "Given a multi-hop question, retrieve passages that answer the question.",
-    "ArguAna-Fa.v2": "Given a claim, find documents that refute the claim.",
-    "QuoraRetrieval-Fa.v2": "Retrieve questions that are semantically equivalent to the given one.",
-    "SCIDOCS-Fa.v2": "Given a scientific paper title, retrieve paper abstracts that are cited by the given paper.",
-    "SciFact-Fa.v2": "Given a scientific claim, retrieve passages that support or refute the claim.",
-    "TRECCOVID-Fa.v2": "Given a query on COVID-19, retrieve documents that answer the query.",
-    "FEVER-FaHardNegatives": "Given a claim, retrieve documents that support or refute the claim.",
-    "SAMSumFa": "Retrieve the most relevant summary for the given conversation.",
-    "SynPerChatbotSumSRetrieval": "Retrieve the most relevant summary for the given conversation.",
-    "SynPerChatbotRAGSumSRetrieval": "Retrieve the most relevant summary for the given conversation.",
-    "ArguAna-VN": "Given a claim, find documents that refute the claim.",
-    "SciFact-VN": "Given a scientific claim, retrieve passages that support or refute the claim.",
-    "ClimateFEVER-VN": "Given a claim about climate change, retrieve documents that support or refute the claim.",
-    "FEVER-VN": "Given a claim, retrieve documents that support or refute the claim.",
-    "DBPedia-VN": "Given a query, retrieve relevant entity descriptions.",
-    "HotpotQA-VN": "Given a multi-hop question, retrieve passages that answer the question.",
-    "TRECCOVID-VN": "Given a query on COVID-19, retrieve documents that answer the query.",
-    "SCIDOCS-VN": "Given a scientific paper title, retrieve paper abstracts that are cited by the given paper.",
-    "Quora-VN": "Retrieve questions that are semantically equivalent to the given one.",
-    "CQADupstackAndroid-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackGis-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackMathematica-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackPhysics-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackProgrammers-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackStats-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackTex-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackUnix-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackWebmasters-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "CQADupstackWordpress-VN": "Given a question, retrieve questions that are semantically equivalent.",
-    "Banking77VNClassification": "Given an online banking query, find the corresponding intents.",
-    "EmotionVNClassification": "Classify the emotion expressed in the given Twitter message into one of the six emotions: anger, fear, joy, love, sadness, and surprise.",
-    "AmazonCounterfactualVNClassification": "Classify a given Amazon customer review text as either counterfactual or not counterfactual.",
-    "MTOPDomainVNClassification": "Classify the intent domain of the given utterance in task-oriented conversation.",
-    "TweetSentimentExtractionVNClassification": "Classify the sentiment of a given tweet as either positive, negative, or neutral",
-    "ToxicConversationsVNClassification": "Classify the given comments as either toxic or not toxic.",
-    "ImdbVNClassification": "Classify the sentiment expressed in the given movie review text from the IMDB dataset.",
-    "MTOPIntentVNClassification": "Classify the intent of the given utterance in task-oriented conversation.",
-    "MassiveScenarioVNClassification": "Given a user utterance as query, find the user scenarios.",
-    "MassiveIntentVNClassification": "Given a user utterance as query, find the user intents.",
-    "AmazonReviewsVNClassification": "Classify the given Amazon review into its appropriate rating category.",
-    "AmazonPolarityVNClassification": "Classify the given Amazon review as either positive or negative.",
-    "SprintDuplicateQuestions-VN": "Retrieve duplicate questions from Sprint forum.",
-    "TwitterSemEval2015-VN": "Retrieve tweets that are semantically similar to the given tweet.",
-    "TwitterURLCorpus-VN": "Retrieve tweets that are semantically similar to the given tweet.",
-    "TwentyNewsgroupsClustering-VN": "Identify the topic or theme of the given news articles.",
-    "RedditClusteringP2P-VN": "Identify the topic or theme of Reddit posts based on the titles and posts.",
-    "StackExchangeClusteringP2P-VN": "Identify the topic or theme of StackExchange posts based on the given titles and paragraphs.",
-    "StackExchangeClustering-VN": "Identify the topic or theme of StackExchange posts based on the titles.",
-    "RedditClustering-VN": "Identify the topic or theme of Reddit posts based on the titles.",
-    "SciDocsRR-VN": "Given a title of a scientific paper, retrieve the titles of other relevant papers.",
-    "AskUbuntuDupQuestions-VN": "Retrieve duplicate questions from AskUbuntu forum.",
-    "StackOverflowDupQuestions-VN": "Retrieve duplicate questions from StackOverflow forum."
-  },
-  "default_prompt_name": null,
-  "similarity_fn_name": "cosine"
-}
+    "prompts": {
+      "query": "Instruct: Given a question, retrieve passages that can help answer the question.\nQuery: ",
+      "document": ""
+    },
+    "default_prompt_name": null,
+    "similarity_fn_name": "cosine"
+}