diff --git "a/data/retrieval_individual-f62bcad9-19c4-430b-a7ba-aa2bfeba99dc.jsonl" "b/data/retrieval_individual-f62bcad9-19c4-430b-a7ba-aa2bfeba99dc.jsonl" --- "a/data/retrieval_individual-f62bcad9-19c4-430b-a7ba-aa2bfeba99dc.jsonl" +++ "b/data/retrieval_individual-f62bcad9-19c4-430b-a7ba-aa2bfeba99dc.jsonl" @@ -1,214 +1,214 @@ -{"tstamp": 1722404021.5866, "task_type": "retrieval", "type": "chat", "model": "intfloat/e5-mistral-7b-instruct", "gen_params": {}, "start": 1722404020.1107, "finish": 1722404021.5866, "ip": "", "conv_id": "e1b36a45539a4451b6a7d436badf788c", "model_name": "intfloat/e5-mistral-7b-instruct", "prompt": "who scored the most points in a game nba history", "output": [["who scored the most points in a game nba history", "List of NBA single-game scoring leaders\n\nThis is a complete listing of National Basketball Association players who have scored 60 or more points in a regular season game.\nThis feat has been accomplished 89 times in NBA history. Thirty-six players have scored 60 or more points in a game, while just ten players have joined the 70-point club. Only 10 players have scored 60 or more points on more than one occasion: Wilt Chamberlain (32 times); Kobe Bryant (6 times); Damian Lillard (5 times); Michael Jordan and James Harden (4 times); Elgin Baylor (3 times); and Karl-Anthony Towns, Devin Booker, Luka Dončić, and Stephen Curry (2 times). Chamberlain holds the single-game scoring record, having scored 100 points in a game in 1962. The youngest player to score at least 60 points in a regular season game is Devin Booker (70 points; 20 years and 145 days), and the oldest is Bryant (60 points; 37 years and 234 days)."]], "corpus": "wikipedia"} -{"tstamp": 1722404021.5866, "task_type": "retrieval", "type": "chat", "model": "embed-english-v3.0", "gen_params": {}, "start": 1722404020.1107, "finish": 1722404021.5866, "ip": "", "conv_id": "cfd99c56377e4533b7115306e6bf5162", "model_name": "embed-english-v3.0", "prompt": "who scored the most points in a game nba history", "output": [["who scored the most points in a game nba history", "List of highest-scoring NBA games\n\nSummary The highest-scoring regular-season game in NBA history is the triple-overtime game between the Detroit Pistons and the Denver Nuggets on December 13, 1983. The two teams combined to score 370 points, with the Pistons defeating the Nuggets 186–184. An NBA-record four players scored over 40 points in the game, including the Nuggets' Kiki Vandeweghe with a game-high 51. The two teams also set several other NBA records, including the most points scored by one team (186 points), the most points scored by a losing team (184), the most field goals by two teams (142), most field goals by one team (74) and most assists by two teams (93).\nThe highest-scoring regular season game in regulation was between the Golden State Warriors and the Denver Nuggets on November 2, 1990. In that game, Golden State defeated Denver 162–158. The Warriors' Chris Mullin scored a game-high 38 points. The Nuggets were coached by Doug Moe from 1980 to 1990 and Paul Westhead from 1990 to 1992, both of whom employed a run-and-gun offensive strategy, which focuses on attempting a high number of field goals while also conceding a large number of points on defense. In fact, Moe's and Westhead's Nuggets were participants in four of the ten highest-scoring regular season games in NBA history. The Warriors were coached by Don Nelson from 1988 to 1995 and 2006 to 2010. He employed Nellie Ball, a style of run and gun that uses smaller, more athletic players to outrun opponents. Another notable high-scoring regular season game is a March 2, 1962, game between the Philadelphia Warriors and the New York Knicks. In that game, the Warriors' Wilt Chamberlain scored an NBA-record 100 points."]], "corpus": "wikipedia"} +{"tstamp": 1722404021.5866, "task_type": "retrieval", "type": "chat", "model": "intfloat/e5-mistral-7b-instruct", "gen_params": {}, "start": 1722404020.1107, "finish": 1722404021.5866, "ip": "", "conv_id": "e1b36a45539a4451b6a7d436badf788c", "model_name": "intfloat/e5-mistral-7b-instruct", "prompt": "who scored the most points in a game nba history", "output": [["who scored the most points in a game nba history", "List of NBA single-game scoring leaders\n\nThis is a complete listing of National Basketball Association players who have scored 60 or more points in a regular season game.\nThis feat has been accomplished 89 times in NBA history. Thirty-six players have scored 60 or more points in a game, while just ten players have joined the 70-point club. Only 10 players have scored 60 or more points on more than one occasion: Wilt Chamberlain (32 times); Kobe Bryant (6 times); Damian Lillard (5 times); Michael Jordan and James Harden (4 times); Elgin Baylor (3 times); and Karl-Anthony Towns, Devin Booker, Luka Don\u010di\u0107, and Stephen Curry (2 times). Chamberlain holds the single-game scoring record, having scored 100 points in a game in 1962. The youngest player to score at least 60 points in a regular season game is Devin Booker (70 points; 20 years and 145 days), and the oldest is Bryant (60 points; 37 years and 234 days)."]], "corpus": "wikipedia"} +{"tstamp": 1722404021.5866, "task_type": "retrieval", "type": "chat", "model": "embed-english-v3.0", "gen_params": {}, "start": 1722404020.1107, "finish": 1722404021.5866, "ip": "", "conv_id": "cfd99c56377e4533b7115306e6bf5162", "model_name": "embed-english-v3.0", "prompt": "who scored the most points in a game nba history", "output": [["who scored the most points in a game nba history", "List of highest-scoring NBA games\n\nSummary The highest-scoring regular-season game in NBA history is the triple-overtime game between the Detroit Pistons and the Denver Nuggets on December 13, 1983. The two teams combined to score 370 points, with the Pistons defeating the Nuggets 186\u2013184. An NBA-record four players scored over 40 points in the game, including the Nuggets' Kiki Vandeweghe with a game-high 51. The two teams also set several other NBA records, including the most points scored by one team (186 points), the most points scored by a losing team (184), the most field goals by two teams (142), most field goals by one team (74) and most assists by two teams (93).\nThe highest-scoring regular season game in regulation was between the Golden State Warriors and the Denver Nuggets on November 2, 1990. In that game, Golden State defeated Denver 162\u2013158. The Warriors' Chris Mullin scored a game-high 38 points. The Nuggets were coached by Doug Moe from 1980 to 1990 and Paul Westhead from 1990 to 1992, both of whom employed a run-and-gun offensive strategy, which focuses on attempting a high number of field goals while also conceding a large number of points on defense. In fact, Moe's and Westhead's Nuggets were participants in four of the ten highest-scoring regular season games in NBA history. The Warriors were coached by Don Nelson from 1988 to 1995 and 2006 to 2010. He employed Nellie Ball, a style of run and gun that uses smaller, more athletic players to outrun opponents. Another notable high-scoring regular season game is a March 2, 1962, game between the Philadelphia Warriors and the New York Knicks. In that game, the Warriors' Wilt Chamberlain scored an NBA-record 100 points."]], "corpus": "wikipedia"} {"tstamp": 1722404025.3419, "task_type": "retrieval", "type": "chat", "model": "GritLM/GritLM-7B", "gen_params": {}, "start": 1722404024.8194, "finish": 1722404025.3419, "ip": "", "conv_id": "5ef17d307c164fcf9ce84bb32d4ddf44", "model_name": "GritLM/GritLM-7B", "prompt": "will system restore remove malware?", "output": [["will system restore remove malware?", "Superuser Stackexchange\n\nQ: Can system restore remove virus from the computer? Can a system restore to a past time remove the virus from the computer?\n\nA: The short answer is no.\nSystem Restore can help spread a virus though.\nSee here for more info: Removing Viruses with System Restore. Quoting the relevant parts:\n\nWhen System Restore first appeared in the market, computer manufacturers and other companies claimed ... that it could remove viruses and fix other problems with the computer.\nIn reality, System Restore can create copies of the infected files. And some viruses may be capable of infecting the restore volume as well as the actual system files. When a person cleans their computer using an anti-virus, then uses System Restore, they may inadvertently re-infect the computer. Or if they use System Restore as a means of removal, either the restore will fail (if the anti-virus cleans the virus out during the restore process) or the restore will replace the file with an infected version.\n\n"]], "corpus": "stackexchange"} {"tstamp": 1722404025.3419, "task_type": "retrieval", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722404024.8194, "finish": 1722404025.3419, "ip": "", "conv_id": "6a4fe1dc03d94381ade5917f12c7933a", "model_name": "BAAI/bge-large-en-v1.5", "prompt": "will system restore remove malware?", "output": [["will system restore remove malware?", "Superuser Stackexchange\n\nQ: Can system restore remove virus from the computer? Can a system restore to a past time remove the virus from the computer?\n\nA: The short answer is no.\nSystem Restore can help spread a virus though.\nSee here for more info: Removing Viruses with System Restore. Quoting the relevant parts:\n\nWhen System Restore first appeared in the market, computer manufacturers and other companies claimed ... that it could remove viruses and fix other problems with the computer.\nIn reality, System Restore can create copies of the infected files. And some viruses may be capable of infecting the restore volume as well as the actual system files. When a person cleans their computer using an anti-virus, then uses System Restore, they may inadvertently re-infect the computer. Or if they use System Restore as a means of removal, either the restore will fail (if the anti-virus cleans the virus out during the restore process) or the restore will replace the file with an infected version.\n\n"]], "corpus": "stackexchange"} {"tstamp": 1722404070.8053, "task_type": "retrieval", "type": "chat", "model": "mixedbread-ai/mxbai-embed-large-v1", "gen_params": {}, "start": 1722404069.9427, "finish": 1722404070.8053, "ip": "", "conv_id": "4c228e854b09431e85d5b528f5c976b6", "model_name": "mixedbread-ai/mxbai-embed-large-v1", "prompt": "How can I automatically generate code-switching sentences for language models using GANs?", "output": [["How can I automatically generate code-switching sentences for language models using GANs?", "Title: Code-switching Sentence Generation by Generative Adversarial Networks and its Application to Data Augmentation\n\nAbstract: Code-switching is about dealing with alternative languages in speech or text. It is partially speaker-depend and domain-related, so completely explaining the phenomenon by linguistic rules is challenging. Compared to most monolingual tasks, insufficient data is an issue for code-switching. To mitigate the issue without expensive human annotation, we proposed an unsupervised method for code-switching data augmentation. By utilizing a generative adversarial network, we can generate intra-sentential code-switching sentences from monolingual sentences. We applied proposed method on two corpora, and the result shows that the generated code-switching sentences improve the performance of code-switching language models."]], "corpus": "arxiv"} {"tstamp": 1722404070.8053, "task_type": "retrieval", "type": "chat", "model": "text-embedding-3-large", "gen_params": {}, "start": 1722404069.9427, "finish": 1722404070.8053, "ip": "", "conv_id": "187adc143b07496aa26bec9a0fd3ab2e", "model_name": "text-embedding-3-large", "prompt": "How can I automatically generate code-switching sentences for language models using GANs?", "output": [["How can I automatically generate code-switching sentences for language models using GANs?", "Title: Code-switching Sentence Generation by Generative Adversarial Networks and its Application to Data Augmentation\n\nAbstract: Code-switching is about dealing with alternative languages in speech or text. It is partially speaker-depend and domain-related, so completely explaining the phenomenon by linguistic rules is challenging. Compared to most monolingual tasks, insufficient data is an issue for code-switching. To mitigate the issue without expensive human annotation, we proposed an unsupervised method for code-switching data augmentation. By utilizing a generative adversarial network, we can generate intra-sentential code-switching sentences from monolingual sentences. We applied proposed method on two corpora, and the result shows that the generated code-switching sentences improve the performance of code-switching language models."]], "corpus": "arxiv"} {"tstamp": 1722405094.8844, "task_type": "retrieval", "type": "chat", "model": "sentence-transformers/all-MiniLM-L6-v2", "gen_params": {}, "start": 1722405094.3211, "finish": 1722405094.8844, "ip": "", "conv_id": "b036221f1e544dad991207219401409f", "model_name": "sentence-transformers/all-MiniLM-L6-v2", "prompt": "Code with switch statements", "output": [["Code with switch statements", "Stackoverflow Stackexchange\n\nQ: Can a concept satisfaction of an expression, contains both type and the reference? Is there a way to make the following code not so bloated?\nI mean join both type and a reference somehow (|| does not work).\ntemplate\nconcept IntegralVector = std::integral &&\nrequires(T t)\n{\n { t.size() } -> std::convertible_to;\n}\n&& (requires(T t)\n{\n { t[0] } -> std::same_as;\n} || requires(T t)\n{\n { t[0] } -> std::same_as;\n});\n\nA working trick can be:\n{ 0 + t[0] } -> std::integral;\n\nBut I want to stick with typename T::value_type\n\nA: You probably want something like this:\ntemplate \nconcept decays_to = std::same_as, U>;\n\nTo use as:\ntemplate\nconcept IntegralVector =\n std::integral\n && requires (T t) {\n { t.size() } -> std::convertible_to;\n { t[0] } -> decays_to;\n };\n\nThis also catches value_type const& as an option, which I'm not sure was intentionally omitted.\n"]], "corpus": "stackexchange"} {"tstamp": 1722405094.8844, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722405094.3211, "finish": 1722405094.8844, "ip": "", "conv_id": "df83449bd8404578b1bbc7184ed9ddca", "model_name": "voyage-multilingual-2", "prompt": "Code with switch statements", "output": [["Code with switch statements", "Stackoverflow Stackexchange\n\nQ: Switch statement inside a switch statement? I have to evaluate many conditions. In my case, I have to do something like this:\nswitch(id)\n{\n case 5:\n // switch some other cases here\n case 6:\n // set some value\n ...\n }\n\nIs it good practice to have another switch in case 5? If not, what's better? Having if statements there?\n\nA: Switch statements aren't bad practice. Nested switch statements might be messy to look at.\nPerhaps think about embedding the nested switch statement in another method to improve clarity.\n"]], "corpus": "stackexchange"} -{"tstamp": 1722405138.8514, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722405138.4946, "finish": 1722405138.8514, "ip": "", "conv_id": "b033f666e0b646ce836e10684bf70fc8", "model_name": "voyage-multilingual-2", "prompt": "is it done yet meaning?", "output": [["is it done yet meaning?", "English Stackexchange\n\nQ: Any difference between “Are you done?” and “Are you done yet?” I see people in movies saying \n\nAre you done? \n\nand \n\nAre you done yet?\n\nAnd I wonder what that the addition of yet might mean or suggest in the second version which is absent in the first one. \nAre those two versions completely equivalent, or is there some subtle distinction?\n\nA: The use of yet here emphasizes that it has taken a reasonable amount of time or that it has taken too long and implicitly expects an answer in the affirmative. \n“Are you done?” is just a question to find out whether he is done with it, whereas “Are you done yet” says you had better get done soon if you aren’t already.\n"]], "corpus": "stackexchange"} +{"tstamp": 1722405138.8514, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722405138.4946, "finish": 1722405138.8514, "ip": "", "conv_id": "b033f666e0b646ce836e10684bf70fc8", "model_name": "voyage-multilingual-2", "prompt": "is it done yet meaning?", "output": [["is it done yet meaning?", "English Stackexchange\n\nQ: Any difference between \u201cAre you done?\u201d and \u201cAre you done yet?\u201d I see people in movies saying \n\nAre you done? \n\nand \n\nAre you done yet?\n\nAnd I wonder what that the addition of yet might mean or suggest in the second version which is absent in the first one. \nAre those two versions completely equivalent, or is there some subtle distinction?\n\nA: The use of yet here emphasizes that it has taken a reasonable amount of time or that it has taken too long and implicitly expects an answer in the affirmative. \n\u201cAre you done?\u201d is just a question to find out whether he is done with it, whereas \u201cAre you done yet\u201d says you had better get done soon if you aren\u2019t already.\n"]], "corpus": "stackexchange"} {"tstamp": 1722405138.8514, "task_type": "retrieval", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722405138.4946, "finish": 1722405138.8514, "ip": "", "conv_id": "8b5f03d793634e30bceff265d231aa73", "model_name": "BAAI/bge-large-en-v1.5", "prompt": "is it done yet meaning?", "output": [["is it done yet meaning?", "Ell Stackexchange\n\nQ: \"I'm done\" or \"I've done\", which is correct? I want to know if \"I'm almost done\" is correct, or whether it should be \"I've almost done\" as a present perfect tense. I often read this on Facebook news feed. Is it correct? \n\nA: *\n\n*\"To be done\" is idiomatic and means \"to have finished\". For example, \"Have you written that email yet?\" \"Yes, I'm done.\"\n\n*\"To be done with something\" (or \"to be finished with something\") to get rid of something or stop using it for good. \"I'm done with email! It takes up so much of my time.\"\n\n\n*\n\n*But \"To have finished with something\" means to have finished using it for now. \"Can I use the computer now, please?\" \"Sure, I've finished with it.\"\n\n\n*\"To have done\" is the present perfect of \"to do\". \"Have you ever sent an email by mistake when you meant to cancel it?\" \"Yes, I've done it hundreds of times.\"\n"]], "corpus": "stackexchange"} {"tstamp": 1722405177.8483, "task_type": "retrieval", "type": "chat", "model": "Alibaba-NLP/gte-Qwen2-7B-instruct", "gen_params": {}, "start": 1722405177.5671, "finish": 1722405177.8483, "ip": "", "conv_id": "11258040188c42fbb0b15638d9f6dd84", "model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "prompt": "How to make cake?", "output": [["How to make cake?", "Cake\n\nCake is a flour confection made from flour, sugar, and other ingredients and is usually baked. In their oldest forms, cakes were modifications of bread, but cakes now cover a wide range of preparations that can be simple or elaborate and which share features with desserts such as pastries, meringues, custards, and pies.\nThe most common ingredients include flour, sugar, eggs, fat (such as butter, oil, or margarine), a liquid, and a leavening agent, such as baking soda or baking powder. Common additional ingredients include dried, candied, or fresh fruit, nuts, cocoa, and extracts such as vanilla, with numerous substitutions for the primary ingredients. Cakes can also be filled with fruit preserves, nuts, or dessert sauces (like custard, jelly, cooked fruit, whipped cream, or syrups), iced with buttercream or other icings, and decorated with marzipan, piped borders, or candied fruit.\nCake is often served as a celebratory dish on ceremonial occasions, such as weddings, anniversaries, and birthdays. There are countless cake recipes; some are bread-like, some are rich and elaborate, and many are centuries old. Cake making is no longer a complicated procedure; while at one time considerable labor went into cake making (particularly the whisking of egg foams), baking equipment and directions have been simplified so that even the most amateur of cooks may bake a cake."]], "corpus": "wikipedia"} {"tstamp": 1722405177.8483, "task_type": "retrieval", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722405177.5671, "finish": 1722405177.8483, "ip": "", "conv_id": "b198fea204d3473292df16a104e1ed81", "model_name": "BAAI/bge-large-en-v1.5", "prompt": "How to make cake?", "output": [["How to make cake?", "Cake\n\nCake is a flour confection made from flour, sugar, and other ingredients and is usually baked. In their oldest forms, cakes were modifications of bread, but cakes now cover a wide range of preparations that can be simple or elaborate and which share features with desserts such as pastries, meringues, custards, and pies.\nThe most common ingredients include flour, sugar, eggs, fat (such as butter, oil, or margarine), a liquid, and a leavening agent, such as baking soda or baking powder. Common additional ingredients include dried, candied, or fresh fruit, nuts, cocoa, and extracts such as vanilla, with numerous substitutions for the primary ingredients. Cakes can also be filled with fruit preserves, nuts, or dessert sauces (like custard, jelly, cooked fruit, whipped cream, or syrups), iced with buttercream or other icings, and decorated with marzipan, piped borders, or candied fruit.\nCake is often served as a celebratory dish on ceremonial occasions, such as weddings, anniversaries, and birthdays. There are countless cake recipes; some are bread-like, some are rich and elaborate, and many are centuries old. Cake making is no longer a complicated procedure; while at one time considerable labor went into cake making (particularly the whisking of egg foams), baking equipment and directions have been simplified so that even the most amateur of cooks may bake a cake."]], "corpus": "wikipedia"} {"tstamp": 1722405226.4848, "task_type": "retrieval", "type": "chat", "model": "Alibaba-NLP/gte-Qwen2-7B-instruct", "gen_params": {}, "start": 1722405226.254, "finish": 1722405226.4848, "ip": "", "conv_id": "5eb5421a209846b8aab73cddb8a8c999", "model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "prompt": "Scene detection from video", "output": [["Scene detection from video", "Title: Dealing with sequences in the RGBDT space\n\nAbstract: Most of the current research in computer vision is focused on working with single images without taking in account temporal information. We present a probabilistic non-parametric model that mixes multiple information cues from devices to segment regions that contain moving objects in image sequences. We prepared an experimental setup to show the importance of using previous information for obtaining an accurate segmentation result, using a novel dataset that provides sequences in the RGBDT space. We label the detected regions ts with a state-of-the-art human detector. Each one of the detected regions is at least marked as human once."]], "corpus": "arxiv"} {"tstamp": 1722405226.4848, "task_type": "retrieval", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722405226.254, "finish": 1722405226.4848, "ip": "", "conv_id": "39376a868a284ddebf65e0e4719b5d4c", "model_name": "BAAI/bge-large-en-v1.5", "prompt": "Scene detection from video", "output": [["Scene detection from video", "Title: Learnable Optimal Sequential Grouping for Video Scene Detection\n\nAbstract: Video scene detection is the task of dividing videos into temporal semantic chapters. This is an important preliminary step before attempting to analyze heterogeneous video content. Recently, Optimal Sequential Grouping (OSG) was proposed as a powerful unsupervised solution to solve a formulation of the video scene detection problem. In this work, we extend the capabilities of OSG to the learning regime. By giving the capability to both learn from examples and leverage a robust optimization formulation, we can boost performance and enhance the versatility of the technology. We present a comprehensive analysis of incorporating OSG into deep learning neural networks under various configurations. These configurations include learning an embedding in a straight-forward manner, a tailored loss designed to guide the solution of OSG, and an integrated model where the learning is performed through the OSG pipeline. With thorough evaluation and analysis, we assess the benefits and behavior of the various configurations, and show that our learnable OSG approach exhibits desirable behavior and enhanced performance compared to the state of the art."]], "corpus": "arxiv"} -{"tstamp": 1722406120.4774, "task_type": "retrieval", "type": "chat", "model": "GritLM/GritLM-7B", "gen_params": {}, "start": 1722406120.2293, "finish": 1722406120.4774, "ip": "", "conv_id": "f003c46cffad4cc8ae52ae4cdc5a8b1b", "model_name": "GritLM/GritLM-7B", "prompt": "Merhaba Türkçe biliyor musun", "output": [["Merhaba Türkçe biliyor musun", "Old Anatolian Turkish\n\nExamples Following texts are excerpts of the Qabus-nama taken from Turan Fikret's Old Anatolian Turkish: Syntactic Structure (1996):\n: \"Our proofs.\"\ndevletlü gişiler: \"Fortunate ones.\"\nzinhār zinhār: \"Never.\"\np��rlikde yigitlenmek rüsvāylıqdur: \"It is a shame to act like a young man in old age.\"\nbulardan artanı beytü’l mālda qoyalar: \"They should put in public treasury that which remained from them.\"\nbirgün bu ilçiyile oturur iken Qısri Büzürcmihre sorar: \"One day when he was sitting with this [foreign] ambassador, Chosroes asked Büzürcmihr.\"\nKelām-ı mecīd: \"The word of the most glorious (God), the Qoran.\"\ndar'ül-harb: \"Countries outside of the domain of Islam.\"\nTaŋrı aŋa raḥmet itmez: \"God does not forgive him.\"\naġırlaŋ aṭaŋuzı anaŋuzı egerçi kāfirse daqı: \"Respect your parents even if they are unbelievers.\"\nNe qul kim alam āzād olsun: \"Every slave that I would buy should be freed\"\nve cāhil gişileri gişi sanma ve hünersüzleri bilür sayma: \"And do not consider the ignorant ones the [real] men, and [do not consider] the untalented ones the knowledgeable ones.\"\nzinhār işüŋi ṭanışmaqdan ʿārlanma: \"Beware, and never be ashamed of learning your job.\"\nsen yalan söyleyesi gişi degülsin: \"You are not someone who would lie.\"\nartuq zaḥmet çeküp artuq ṭamaʿ eyleme: \"Do not work hard to satisfy your greed.\""]], "corpus": "wikipedia"} +{"tstamp": 1722406120.4774, "task_type": "retrieval", "type": "chat", "model": "GritLM/GritLM-7B", "gen_params": {}, "start": 1722406120.2293, "finish": 1722406120.4774, "ip": "", "conv_id": "f003c46cffad4cc8ae52ae4cdc5a8b1b", "model_name": "GritLM/GritLM-7B", "prompt": "Merhaba T\u00fcrk\u00e7e biliyor musun", "output": [["Merhaba T\u00fcrk\u00e7e biliyor musun", "Old Anatolian Turkish\n\nExamples Following texts are excerpts of the Qabus-nama taken from Turan Fikret's Old Anatolian Turkish: Syntactic Structure (1996):\n: \"Our proofs.\"\ndevletl\u00fc gi\u015filer: \"Fortunate ones.\"\nzinh\u0101r zinh\u0101r: \"Never.\"\np\u012brlikde yigitlenmek r\u00fcsv\u0101yl\u0131qdur: \"It is a shame to act like a young man in old age.\"\nbulardan artan\u0131 beyt\u00fc\u2019l m\u0101lda qoyalar: \"They should put in public treasury that which remained from them.\"\nbirg\u00fcn bu il\u00e7iyile oturur iken Q\u0131sri B\u00fcz\u00fcrcmihre sorar: \"One day when he was sitting with this [foreign] ambassador, Chosroes asked B\u00fcz\u00fcrcmihr.\"\nKel\u0101m-\u0131 mec\u012bd: \"The word of the most glorious (God), the Qoran.\"\ndar'\u00fcl-harb: \"Countries outside of the domain of Islam.\"\nTa\u014br\u0131 a\u014ba ra\u1e25met itmez: \"God does not forgive him.\"\na\u0121\u0131rla\u014b a\u1e6da\u014buz\u0131 ana\u014buz\u0131 eger\u00e7i k\u0101firse daq\u0131: \"Respect your parents even if they are unbelievers.\"\nNe qul kim alam \u0101z\u0101d olsun: \"Every slave that I would buy should be freed\"\nve c\u0101hil gi\u015fileri gi\u015fi sanma ve h\u00fcners\u00fczleri bil\u00fcr sayma: \"And do not consider the ignorant ones the [real] men, and [do not consider] the untalented ones the knowledgeable ones.\"\nzinh\u0101r i\u015f\u00fc\u014bi \u1e6dan\u0131\u015fmaqdan \u02bf\u0101rlanma: \"Beware, and never be ashamed of learning your job.\"\nsen yalan s\u00f6yleyesi gi\u015fi deg\u00fclsin: \"You are not someone who would lie.\"\nartuq za\u1e25met \u00e7ek\u00fcp artuq \u1e6dama\u02bf eyleme: \"Do not work hard to satisfy your greed.\""]], "corpus": "wikipedia"} {"tstamp": 1722406128.4958, "task_type": "retrieval", "type": "chat", "model": "Alibaba-NLP/gte-Qwen2-7B-instruct", "gen_params": {}, "start": 1722406128.1407, "finish": 1722406128.4958, "ip": "", "conv_id": "fe4e46bfcf3e4e68bc1cc80ad40bba78", "model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "prompt": "semantic scenes", "output": [["semantic scenes", "Title: SceneNet: Understanding Real World Indoor Scenes With Synthetic Data\n\nAbstract: Scene understanding is a prerequisite to many high level tasks for any automated intelligent machine operating in real world environments. Recent attempts with supervised learning have shown promise in this direction but also highlighted the need for enormous quantity of supervised data --- performance increases in proportion to the amount of data used. However, this quickly becomes prohibitive when considering the manual labour needed to collect such data. In this work, we focus our attention on depth based semantic per-pixel labelling as a scene understanding problem and show the potential of computer graphics to generate virtually unlimited labelled data from synthetic 3D scenes. By carefully synthesizing training data with appropriate noise models we show comparable performance to state-of-the-art RGBD systems on NYUv2 dataset despite using only depth data as input and set a benchmark on depth-based segmentation on SUN RGB-D dataset. Additionally, we offer a route to generating synthesized frame or video data, and understanding of different factors influencing performance gains."]], "corpus": "arxiv"} {"tstamp": 1722406128.4958, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722406128.1407, "finish": 1722406128.4958, "ip": "", "conv_id": "0d885f93765d4b04aa68a2008bdeedef", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "semantic scenes", "output": [["semantic scenes", "Title: Semantic scene descriptions as an objective of human vision\n\nAbstract: Interpreting the meaning of a visual scene requires not only identification of its constituent objects, but also a rich semantic characterization of object interrelations. Here, we study the neural mechanisms underlying visuo-semantic transformations by applying modern computational techniques to a large-scale 7T fMRI dataset of human brain responses elicited by complex natural scenes. Using semantic embeddings obtained by applying linguistic deep learning models to human-generated scene descriptions, we identify a widely distributed network of brain regions that encode semantic scene descriptions. Importantly, these semantic embeddings better explain activity in these regions than traditional object category labels. In addition, they are effective predictors of activity despite the fact that the participants did not actively engage in a semantic task, suggesting that visuo-semantic transformations are a default mode of vision. In support of this view, we then show that highly accurate reconstructions of scene captions can be directly linearly decoded from patterns of brain activity. Finally, a recurrent convolutional neural network trained on semantic embeddings further outperforms semantic embeddings in predicting brain activity, providing a mechanistic model of the brain's visuo-semantic transformations. Together, these experimental and computational results suggest that transforming visual input into rich semantic scene descriptions may be a central objective of the visual system, and that focusing efforts on this new objective may lead to improved models of visual information processing in the human brain."]], "corpus": "arxiv"} {"tstamp": 1722406203.6626, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722406203.4447, "finish": 1722406203.6626, "ip": "", "conv_id": "1c804bb1dffc4db9afad3ff24837c3b1", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "scene detection", "output": [["scene detection", "Title: Deep learning for scene recognition from visual data: a survey\n\nAbstract: The use of deep learning techniques has exploded during the last few years, resulting in a direct contribution to the field of artificial intelligence. This work aims to be a review of the state-of-the-art in scene recognition with deep learning models from visual data. Scene recognition is still an emerging field in computer vision, which has been addressed from a single image and dynamic image perspective. We first give an overview of available datasets for image and video scene recognition. Later, we describe ensemble techniques introduced by research papers in the field. Finally, we give some remarks on our findings and discuss what we consider challenges in the field and future lines of research. This paper aims to be a future guide for model selection for the task of scene recognition."]], "corpus": "arxiv"} {"tstamp": 1722406203.6626, "task_type": "retrieval", "type": "chat", "model": "BM25", "gen_params": {}, "start": 1722406203.4447, "finish": 1722406203.6626, "ip": "", "conv_id": "392a2ac2390e4e8a9b05c0122d6ad2c2", "model_name": "BM25", "prompt": "scene detection", "output": [["scene detection", "Title: Open-Vocabulary Object Detection via Scene Graph Discovery\n\nAbstract: In recent years, open-vocabulary (OV) object detection has attracted increasing research attention. Unlike traditional detection, which only recognizes fixed-category objects, OV detection aims to detect objects in an open category set. Previous works often leverage vision-language (VL) training data (e.g., referring grounding data) to recognize OV objects. However, they only use pairs of nouns and individual objects in VL data, while these data usually contain much more information, such as scene graphs, which are also crucial for OV detection. In this paper, we propose a novel Scene-Graph-Based Discovery Network (SGDN) that exploits scene graph cues for OV detection. Firstly, a scene-graph-based decoder (SGDecoder) including sparse scene-graph-guided attention (SSGA) is presented. It captures scene graphs and leverages them to discover OV objects. Secondly, we propose scene-graph-based prediction (SGPred), where we build a scene-graph-based offset regression (SGOR) mechanism to enable mutual enhancement between scene graph extraction and object localization. Thirdly, we design a cross-modal learning mechanism in SGPred. It takes scene graphs as bridges to improve the consistency between cross-modal embeddings for OV object classification. Experiments on COCO and LVIS demonstrate the effectiveness of our approach. Moreover, we show the ability of our model for OV scene graph detection, while previous OV scene graph generation methods cannot tackle this task."]], "corpus": "arxiv"} -{"tstamp": 1722406341.6245, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722406341.1226, "finish": 1722406341.6245, "ip": "", "conv_id": "7a058e06c2b049c4a60e4ac7ab04d00f", "model_name": "voyage-multilingual-2", "prompt": "巴黎奥运会举办时间", "output": [["巴黎奥运会举办时间", "2024 Summer Olympics\n\nThe 2024 Summer Olympics (), officially the Games of the XXXIII Olympiad () and officially branded as Paris 2024, is an upcoming international multi-sport event scheduled to take place from 26 July (the date of the opening ceremony) to 11 August 2024 in France, with some competitions starting on 24 July. Paris is the main host city, with events held at 16 other cities spread across Metropolitan France, plus one subsite in Tahiti—an island within the French overseas country and overseas collectivity of French Polynesia."]], "corpus": "wikipedia"} -{"tstamp": 1722406341.6245, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722406341.1226, "finish": 1722406341.6245, "ip": "", "conv_id": "1276c96374c14ff9a875f0f17fa86b76", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "巴黎奥运会举办时间", "output": [["巴黎奥运会举办时间", "Paris bid for the 2024 Summer Olympics\n\nParis 2024 is the successful bid to bring the Games of the XXXIII Olympiad and the XVII Paralympic Games, to the French capital city. Paris formally announced its intention to bid on 23 June 2015 – the date on which Olympic Day is globally celebrated. Following withdrawals in the 2024 Summer Olympics bidding process that led to just two candidate cities (Los Angeles and Paris), the IOC announced that the 2028 Summer Olympics would be awarded at the same time as the 2024 Games. After Los Angeles agreed on 31 July 2017 to host the 2028 Games. It was officially announced at the IOC Session in Lima, Peru.\nParis previously hosted the 1900 Summer Olympics and the 1924 Summer Olympics. Paris will be the second city (after London) to host the Olympic Games three times. Of note, 2024 marks the 100th anniversary of Paris' 1924 Summer Olympics, as well as the first Olympic Winter Games in Chamonix. On July 31, 2017, it was announced that rival bidder Los Angeles would host in 2028, effectively giving Paris the 2024 games.\nDates\nThe Olympic Games will be held from 26 July 2024 to 11 August 2024; while the Paralympic Games will be from 28 August 2024 to 8 September 2024. Paris will be the second city to host the games three times after London (1908, 1948 and 2012). These would be the sixth hosted games in France, and the third in summer. Los Angeles, which was announced as the 2028 Games host, will be the third city to host three times (1932 and 1984)."]], "corpus": "wikipedia"} -{"tstamp": 1722406459.7523, "task_type": "retrieval", "type": "chat", "model": "text-embedding-3-large", "gen_params": {}, "start": 1722406459.1814, "finish": 1722406459.7523, "ip": "", "conv_id": "1efdece5969d41a4b26d3106c5a5441b", "model_name": "text-embedding-3-large", "prompt": "RAG 的流程", "output": [["RAG 的流程", "V(D)J recombination\n\nProcess V(D)J recombination begins when V(D)J recombinase (through the activity of RAG1) binds a RSS flanking a coding gene segment (V, D, or J) and creates a single-strand nick in the DNA between the first base of the RSS (just before the heptamer) and the coding segment. This is essentially energetically neutral (no need for ATP hydrolysis) and results in the formation of a free 3' hydroxyl group and a 5' phosphate group on the same strand. The reactive hydroxyl group is positioned by the recombinase to attack the phosphodiester bond of opposite strand, forming two DNA ends: a hairpin (stem-loop) on the coding segment and a blunt end on the signal segment. The current model is that DNA nicking and hairpin formation occurs on both strands simultaneously (or nearly so) in a complex known as a recombination center.\nThe blunt signal ends are flush ligated together to form a circular piece of DNA containing all of the intervening sequences between the coding segments known as a signal joint (although circular in nature, this is not to be confused with a plasmid). While originally thought to be lost during successive cell divisions, there is evidence that signal joints may re-enter the genome and lead to pathologies by activating oncogenes or interrupting tumor suppressor gene function(s)[Ref]."]], "corpus": "wikipedia"} -{"tstamp": 1722406459.7523, "task_type": "retrieval", "type": "chat", "model": "intfloat/e5-mistral-7b-instruct", "gen_params": {}, "start": 1722406459.1814, "finish": 1722406459.7523, "ip": "", "conv_id": "43354e26646248698331bb5de5c1e838", "model_name": "intfloat/e5-mistral-7b-instruct", "prompt": "RAG 的流程", "output": [["RAG 的流程", "Raft (algorithm)\n\nRaft is a consensus algorithm designed as an alternative to the Paxos family of algorithms. It was meant to be more understandable than Paxos by means of separation of logic, but it is also formally proven safe and offers some additional features. Raft offers a generic way to distribute a state machine across a cluster of computing systems, ensuring that each node in the cluster agrees upon the same series of state transitions. It has a number of open-source reference implementations, with full-specification implementations in Go, C++, Java, and Scala. It is named after Reliable, Replicated, Redundant, And Fault-Tolerant.\nRaft is not a Byzantine fault tolerant (BFT) algorithm; the nodes trust the elected leader."]], "corpus": "wikipedia"} +{"tstamp": 1722406341.6245, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722406341.1226, "finish": 1722406341.6245, "ip": "", "conv_id": "7a058e06c2b049c4a60e4ac7ab04d00f", "model_name": "voyage-multilingual-2", "prompt": "\u5df4\u9ece\u5965\u8fd0\u4f1a\u4e3e\u529e\u65f6\u95f4", "output": [["\u5df4\u9ece\u5965\u8fd0\u4f1a\u4e3e\u529e\u65f6\u95f4", "2024 Summer Olympics\n\nThe 2024 Summer Olympics (), officially the Games of the XXXIII Olympiad () and officially branded as Paris 2024, is an upcoming international multi-sport event scheduled to take place from 26\u00a0July (the date of the opening ceremony) to 11\u00a0August 2024 in France, with some competitions starting on 24\u00a0July. Paris is the main host city, with events held at 16 other cities spread across Metropolitan France, plus one subsite in Tahiti\u2014an island within the French overseas country and overseas collectivity of French Polynesia."]], "corpus": "wikipedia"} +{"tstamp": 1722406341.6245, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722406341.1226, "finish": 1722406341.6245, "ip": "", "conv_id": "1276c96374c14ff9a875f0f17fa86b76", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "\u5df4\u9ece\u5965\u8fd0\u4f1a\u4e3e\u529e\u65f6\u95f4", "output": [["\u5df4\u9ece\u5965\u8fd0\u4f1a\u4e3e\u529e\u65f6\u95f4", "Paris bid for the 2024 Summer Olympics\n\nParis 2024 is the successful bid to bring the Games of the XXXIII Olympiad and the XVII Paralympic Games, to the French capital city. Paris formally announced its intention to bid on 23 June 2015 \u2013 the date on which Olympic Day is globally celebrated. Following withdrawals in the 2024 Summer Olympics bidding process that led to just two candidate cities (Los Angeles and Paris), the IOC announced that the 2028 Summer Olympics would be awarded at the same time as the 2024 Games. After Los Angeles agreed on 31 July 2017 to host the 2028 Games. It was officially announced at the IOC Session in Lima, Peru.\nParis previously hosted the 1900 Summer Olympics and the 1924 Summer Olympics. Paris will be the second city (after London) to host the Olympic Games three times. Of note, 2024 marks the 100th anniversary of Paris' 1924 Summer Olympics, as well as the first Olympic Winter Games in Chamonix. On July 31, 2017, it was announced that rival bidder Los Angeles would host in 2028, effectively giving Paris the 2024 games.\nDates\nThe Olympic Games will be held from 26 July 2024 to 11 August 2024; while the Paralympic Games will be from 28 August 2024 to 8 September 2024. Paris will be the second city to host the games three times after London (1908, 1948 and 2012). These would be the sixth hosted games in France, and the third in summer. Los Angeles, which was announced as the 2028 Games host, will be the third city to host three times (1932 and 1984)."]], "corpus": "wikipedia"} +{"tstamp": 1722406459.7523, "task_type": "retrieval", "type": "chat", "model": "text-embedding-3-large", "gen_params": {}, "start": 1722406459.1814, "finish": 1722406459.7523, "ip": "", "conv_id": "1efdece5969d41a4b26d3106c5a5441b", "model_name": "text-embedding-3-large", "prompt": "RAG \u7684\u6d41\u7a0b", "output": [["RAG \u7684\u6d41\u7a0b", "V(D)J recombination\n\nProcess V(D)J recombination begins when V(D)J recombinase (through the activity of RAG1) binds a RSS flanking a coding gene segment (V, D, or J) and creates a single-strand nick in the DNA between the first base of the RSS (just before the heptamer) and the coding segment. This is essentially energetically neutral (no need for ATP hydrolysis) and results in the formation of a free 3' hydroxyl group and a 5' phosphate group on the same strand. The reactive hydroxyl group is positioned by the recombinase to attack the phosphodiester bond of opposite strand, forming two DNA ends: a hairpin (stem-loop) on the coding segment and a blunt end on the signal segment. The current model is that DNA nicking and hairpin formation occurs on both strands simultaneously (or nearly so) in a complex known as a recombination center.\nThe blunt signal ends are flush ligated together to form a circular piece of DNA containing all of the intervening sequences between the coding segments known as a signal joint (although circular in nature, this is not to be confused with a plasmid). While originally thought to be lost during successive cell divisions, there is evidence that signal joints may re-enter the genome and lead to pathologies by activating oncogenes or interrupting tumor suppressor gene function(s)[Ref]."]], "corpus": "wikipedia"} +{"tstamp": 1722406459.7523, "task_type": "retrieval", "type": "chat", "model": "intfloat/e5-mistral-7b-instruct", "gen_params": {}, "start": 1722406459.1814, "finish": 1722406459.7523, "ip": "", "conv_id": "43354e26646248698331bb5de5c1e838", "model_name": "intfloat/e5-mistral-7b-instruct", "prompt": "RAG \u7684\u6d41\u7a0b", "output": [["RAG \u7684\u6d41\u7a0b", "Raft (algorithm)\n\nRaft is a consensus algorithm designed as an alternative to the Paxos family of algorithms. It was meant to be more understandable than Paxos by means of separation of logic, but it is also formally proven safe and offers some additional features. Raft offers a generic way to distribute a state machine across a cluster of computing systems, ensuring that each node in the cluster agrees upon the same series of state transitions. It has a number of open-source reference implementations, with full-specification implementations in Go, C++, Java, and Scala. It is named after Reliable, Replicated, Redundant, And Fault-Tolerant.\nRaft is not a Byzantine fault tolerant (BFT) algorithm; the nodes trust the elected leader."]], "corpus": "wikipedia"} {"tstamp": 1722407368.8114, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722407368.4256, "finish": 1722407368.8114, "ip": "", "conv_id": "749ebe2a52df465dbc75e3cee25a46f4", "model_name": "voyage-multilingual-2", "prompt": "Paper discussing the risks of benchmark leakage in evaluating large language models", "output": [["Paper discussing the risks of benchmark leakage in evaluating large language models", "Title: Don't Make Your LLM an Evaluation Benchmark Cheater\n\nAbstract: Large language models~(LLMs) have greatly advanced the frontiers of artificial intelligence, attaining remarkable improvement in model capacity. To assess the model performance, a typical approach is to construct evaluation benchmarks for measuring the ability level of LLMs in different aspects. Despite that a number of high-quality benchmarks have been released, the concerns about the appropriate use of these benchmarks and the fair comparison of different models are increasingly growing. Considering these concerns, in this paper, we discuss the potential risk and impact of inappropriately using evaluation benchmarks and misleadingly interpreting the evaluation results. Specially, we focus on a special issue that would lead to inappropriate evaluation, \\ie \\emph{benchmark leakage}, referring that the data related to evaluation sets is occasionally used for model training. This phenomenon now becomes more common since pre-training data is often prepared ahead of model test. We conduct extensive experiments to study the effect of benchmark leverage, and find that it can dramatically boost the evaluation results, which would finally lead to an unreliable assessment of model performance. To improve the use of existing evaluation benchmarks, we finally present several guidelines for both LLM developers and benchmark maintainers. We hope this work can draw attention to appropriate training and evaluation of LLMs."]], "corpus": "arxiv"} {"tstamp": 1722407368.8114, "task_type": "retrieval", "type": "chat", "model": "embed-english-v3.0", "gen_params": {}, "start": 1722407368.4256, "finish": 1722407368.8114, "ip": "", "conv_id": "0606e1958b1742f0a9ac8ebac7a38347", "model_name": "embed-english-v3.0", "prompt": "Paper discussing the risks of benchmark leakage in evaluating large language models", "output": [["Paper discussing the risks of benchmark leakage in evaluating large language models", "Title: Benchmarking Benchmark Leakage in Large Language Models\n\nAbstract: Amid the expanding use of pre-training data, the phenomenon of benchmark dataset leakage has become increasingly prominent, exacerbated by opaque training processes and the often undisclosed inclusion of supervised data in contemporary Large Language Models (LLMs). This issue skews benchmark effectiveness and fosters potentially unfair comparisons, impeding the field's healthy development. To address this, we introduce a detection pipeline utilizing Perplexity and N-gram accuracy, two simple and scalable metrics that gauge a model's prediction precision on benchmark, to identify potential data leakages. By analyzing 31 LLMs under the context of mathematical reasoning, we reveal substantial instances of training even test set misuse, resulting in potentially unfair comparisons. These findings prompt us to offer several recommendations regarding model documentation, benchmark setup, and future evaluations. Notably, we propose the \"Benchmark Transparency Card\" to encourage clear documentation of benchmark utilization, promoting transparency and healthy developments of LLMs. we have made our leaderboard, pipeline implementation, and model predictions publicly available, fostering future research."]], "corpus": "arxiv"} -{"tstamp": 1722407474.9642, "task_type": "retrieval", "type": "chat", "model": "mixedbread-ai/mxbai-embed-large-v1", "gen_params": {}, "start": 1722407474.7684, "finish": 1722407474.9642, "ip": "", "conv_id": "3a6107aa4ef346eeb4450c8dbcd1170e", "model_name": "mixedbread-ai/mxbai-embed-large-v1", "prompt": "姚明在nba总共拿了多少分", "output": [["姚明在nba总共拿了多少分", "American Basketball Association (2000–present)\n\nJavon McCrea Jeremy McNeil\nPete Mickeal\nAnthony Miller\nOliver Miller\nPercy Miller\nWillie Mitchell\nJamario Moon\nChris Morris\nIsaiah Morris\nLawrence Moten\nByron Mouton\nEric Murdock\nTyrone Nesby\nTyler Newton\nEd O'Bannon\nDoug Overton\nJosh Pace\nGerald Paddio\nJannero Pargo\nRoyce Parran\nAnthony Pelle\nMike Penberthy\nDarren Phillip\nChris Porter\nRashaad Powell\nJames Reaves\nKhalid Reeves\nKareem Reid\nEric Riley\nJohn Roberson\nLawrence Roberts\nStanley Roberts\nRyan Robertson\nJames Robinson\nMike Robinson\nDennis Rodman *\nRené Rougeau\nTrevor Ruffin\nJaRon Rush\nKareem Rush\nBryon Russell\nMark Sanford\nJason Sasser\nAkeem Scott\nDeRonn Scott\nShea Seals\nClayton Shields\nPaul Shirley\nTroy Simons\nDuane Simpkins\nLazarus Sims\nReggie Slater\nDoug Smith\nEddie Smith\nTony Smith\nIsaac Spencer\nCurtis Staples\nPerry Stevenson\nAnthony Steward\nJohn Strickland\nJayceon Taylor\nDoug Thomas\nJamel Thomas\nTorey Thomas\nScotty Thurman\nClay Tucker\nJoah Tucker\nNick VanderLaan\nDavid Vanterpool\nFred Vinson\nJermaine Walker\nMatt Walsh\nRex Walters\nJerod Ward\nReginald Warren\nJameel Watkins\nSylvania Watkins\nC. J. Webster\nTony Weeden\nDominick Welch\nBubba Wells\nRobert Whaley\nDeJuan Wheat\nTyson Wheeler\nDavin White\nLou White\nDonald Whiteside\nBrandon Williams\nJason Williams\nJerry Williams\nLarry Williams\nRichie Williams\nTim Winn\nTerrence Woodyard\nDamian Woolfolk\nMetta World Peace\nGalen Young\nAsia:\nMatt Freije\nChen Hsin-an\nMark Magsumbol\nSun Mingming\nGuy Parselany\nYoav Saffar"]], "corpus": "wikipedia"} -{"tstamp": 1722407474.9642, "task_type": "retrieval", "type": "chat", "model": "BM25", "gen_params": {}, "start": 1722407474.7684, "finish": 1722407474.9642, "ip": "", "conv_id": "4ac35af1927c4e64b90ce469dcb143d3", "model_name": "BM25", "prompt": "姚明在nba总共拿了多少分", "output": [["姚明在nba总共拿了多少分", "List of Desert Island Discs episodes\n\nThe BBC Radio 4 programme Desert Island Discs invites castaways to choose eight pieces of music, a book (in addition to the Bible – or a religious text appropriate to that person's beliefs – and the Complete Works of Shakespeare) and a luxury item that they would take to an imaginary desert island, where they will be marooned indefinitely. The rules state that the chosen luxury item must not be anything animate or indeed anything that enables the castaway to escape from the island, for instance a radio set, sailing yacht or aeroplane. The choices of book and luxury can sometimes give insight into the guest's life, and the choices of guests are listed here."]], "corpus": "wikipedia"} -{"tstamp": 1722407542.9225, "task_type": "retrieval", "type": "chat", "model": "text-embedding-3-large", "gen_params": {}, "start": 1722407542.2185, "finish": 1722407542.9225, "ip": "", "conv_id": "37e30a5661ae4eb5a9ac1959e2a853c7", "model_name": "text-embedding-3-large", "prompt": "how many points yaoming score in NBA", "output": [["how many points yaoming score in NBA", "Yao Ming\n\n\"I was just really happy to make that shot\", Yao said after the Americans' 101–70 victory. \"It was the first score in our Olympic campaign here at home and I'll always remember it. It represents that we can keep our heads up in the face of really tough odds.\"\nFollowing an overtime defeat to Spain, Yao scored 30 points in a win over Angola, and 25 points in a three-point win against Germany, which clinched China's place in the quarterfinals. However, China lost to Lithuania in the quarterfinals by 26 points, eliminating them from the tournament. Yao's 19 points a game were the second-highest in the Olympics, and his averages of 8.2 rebounds and 1.5 blocks per game were third overall.\nCareer statistics\nCBA statistics\nNBA statistics\nRegular season\nPlayoffs\nAwards and achievements\nNaismith Memorial Basketball Hall of Fame: Class of 2016\nFIBA Hall of Fame: Class of 2023\n8× NBA All-Star: 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2011\n5× All-NBA Team:\nSecond Team: 2007, 2009\nThird Team: 2004, 2006, 2008\nNBA All-Rookie First Team: 2003\nNBA Rookie All-Star Game: 2004\nGold medal winner with Team China at the 2001, 2003, and 2005 FIBA Asia Cups\nMVP of the 2001, 2003, 2005 FIBA Asia Cups\nFIBA Diamond Ball Top Scorer: 2004\nAll-Tournament Team, FIBA World Cup: 2002\nChinese Basketball Association Champion: 2001–02"]], "corpus": "wikipedia"} +{"tstamp": 1722407474.9642, "task_type": "retrieval", "type": "chat", "model": "mixedbread-ai/mxbai-embed-large-v1", "gen_params": {}, "start": 1722407474.7684, "finish": 1722407474.9642, "ip": "", "conv_id": "3a6107aa4ef346eeb4450c8dbcd1170e", "model_name": "mixedbread-ai/mxbai-embed-large-v1", "prompt": "\u59da\u660e\u5728nba\u603b\u5171\u62ff\u4e86\u591a\u5c11\u5206", "output": [["\u59da\u660e\u5728nba\u603b\u5171\u62ff\u4e86\u591a\u5c11\u5206", "American Basketball Association (2000\u2013present)\n\nJavon McCrea Jeremy McNeil\nPete Mickeal\nAnthony Miller\nOliver Miller\nPercy Miller\nWillie Mitchell\nJamario Moon\nChris Morris\nIsaiah Morris\nLawrence Moten\nByron Mouton\nEric Murdock\nTyrone Nesby\nTyler Newton\nEd O'Bannon\nDoug Overton\nJosh Pace\nGerald Paddio\nJannero Pargo\nRoyce Parran\nAnthony Pelle\nMike Penberthy\nDarren Phillip\nChris Porter\nRashaad Powell\nJames Reaves\nKhalid Reeves\nKareem Reid\nEric Riley\nJohn Roberson\nLawrence Roberts\nStanley Roberts\nRyan Robertson\nJames Robinson\nMike Robinson\nDennis Rodman *\nRen\u00e9 Rougeau\nTrevor Ruffin\nJaRon Rush\nKareem Rush\nBryon Russell\nMark Sanford\nJason Sasser\nAkeem Scott\nDeRonn Scott\nShea Seals\nClayton Shields\nPaul Shirley\nTroy Simons\nDuane Simpkins\nLazarus Sims\nReggie Slater\nDoug Smith\nEddie Smith\nTony Smith\nIsaac Spencer\nCurtis Staples\nPerry Stevenson\nAnthony Steward\nJohn Strickland\nJayceon Taylor\nDoug Thomas\nJamel Thomas\nTorey Thomas\nScotty Thurman\nClay Tucker\nJoah Tucker\nNick VanderLaan\nDavid Vanterpool\nFred Vinson\nJermaine Walker\nMatt Walsh\nRex Walters\nJerod Ward\nReginald Warren\nJameel Watkins\nSylvania Watkins\nC. J. Webster\nTony Weeden\nDominick Welch\nBubba Wells\nRobert Whaley\nDeJuan Wheat\nTyson Wheeler\nDavin White\nLou White\nDonald Whiteside\nBrandon Williams\nJason Williams\nJerry Williams\nLarry Williams\nRichie Williams\nTim Winn\nTerrence Woodyard\nDamian Woolfolk\nMetta World Peace\nGalen Young\nAsia:\nMatt Freije\nChen Hsin-an\nMark Magsumbol\nSun Mingming\nGuy Parselany\nYoav Saffar"]], "corpus": "wikipedia"} +{"tstamp": 1722407474.9642, "task_type": "retrieval", "type": "chat", "model": "BM25", "gen_params": {}, "start": 1722407474.7684, "finish": 1722407474.9642, "ip": "", "conv_id": "4ac35af1927c4e64b90ce469dcb143d3", "model_name": "BM25", "prompt": "\u59da\u660e\u5728nba\u603b\u5171\u62ff\u4e86\u591a\u5c11\u5206", "output": [["\u59da\u660e\u5728nba\u603b\u5171\u62ff\u4e86\u591a\u5c11\u5206", "List of Desert Island Discs episodes\n\nThe BBC Radio 4 programme Desert Island Discs invites castaways to choose eight pieces of music, a book (in addition to the Bible \u2013 or a religious text appropriate to that person's beliefs \u2013 and the Complete Works of Shakespeare) and a luxury item that they would take to an imaginary desert island, where they will be marooned indefinitely. The rules state that the chosen luxury item must not be anything animate or indeed anything that enables the castaway to escape from the island, for instance a radio set, sailing yacht or aeroplane. The choices of book and luxury can sometimes give insight into the guest's life, and the choices of guests are listed here."]], "corpus": "wikipedia"} +{"tstamp": 1722407542.9225, "task_type": "retrieval", "type": "chat", "model": "text-embedding-3-large", "gen_params": {}, "start": 1722407542.2185, "finish": 1722407542.9225, "ip": "", "conv_id": "37e30a5661ae4eb5a9ac1959e2a853c7", "model_name": "text-embedding-3-large", "prompt": "how many points yaoming score in NBA", "output": [["how many points yaoming score in NBA", "Yao Ming\n\n\"I was just really happy to make that shot\", Yao said after the Americans' 101\u201370 victory. \"It was the first score in our Olympic campaign here at home and I'll always remember it. It represents that we can keep our heads up in the face of really tough odds.\"\nFollowing an overtime defeat to Spain, Yao scored 30 points in a win over Angola, and 25 points in a three-point win against Germany, which clinched China's place in the quarterfinals. However, China lost to Lithuania in the quarterfinals by 26 points, eliminating them from the tournament. Yao's 19 points a game were the second-highest in the Olympics, and his averages of 8.2 rebounds and 1.5 blocks per game were third overall.\nCareer statistics\nCBA statistics\nNBA statistics\nRegular season\nPlayoffs\nAwards and achievements\nNaismith Memorial Basketball Hall of Fame: Class of 2016\nFIBA Hall of Fame: Class of 2023\n8\u00d7 NBA All-Star: 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2011\n5\u00d7 All-NBA Team:\nSecond Team: 2007, 2009\nThird Team: 2004, 2006, 2008\nNBA All-Rookie First Team: 2003\nNBA Rookie All-Star Game: 2004\nGold medal winner with Team China at the 2001, 2003, and 2005 FIBA Asia Cups\nMVP of the 2001, 2003, 2005 FIBA Asia Cups\nFIBA Diamond Ball Top Scorer: 2004\nAll-Tournament Team, FIBA World Cup: 2002\nChinese Basketball Association Champion: 2001\u201302"]], "corpus": "wikipedia"} {"tstamp": 1722407542.9225, "task_type": "retrieval", "type": "chat", "model": "jinaai/jina-embeddings-v2-base-en", "gen_params": {}, "start": 1722407542.2185, "finish": 1722407542.9225, "ip": "", "conv_id": "8b0e27cb6f5b49f3a284afe24d7376a5", "model_name": "jinaai/jina-embeddings-v2-base-en", "prompt": "how many points yaoming score in NBA", "output": [["how many points yaoming score in NBA", "Points per game\n\nPoints per game, often abbreviated PPG, is the average number of points scored by a player per game played in a sport, over the course of a series of games, a whole season, or a career. It is calculated by dividing the total number of points by number of games. The terminology is often used in basketball and ice hockey. For description of sports points see points for ice hockey or points for basketball. In games divided into fixed time periods, especially those in which a player may exit and re-enter the game multiple or an unlimited number of times, a player may receive the same credit (in this context, a liability) for participation in a game regardless of how long (i.e., for what portion of the game clock's elapsing) they were actually on the field or court. For this reason, the points-per-game statistic may understate the contribution of players who are highly effective but used only in certain specific \"pinch\" or \"clutch\" scenarios, such that a points-per-unit-time figure (e.g., \"points per 48 minutes\" in the case of professional basketball) may better represent their effectiveness within the context in which a coach or manager plays them. Although the points-per-game statistic has the advantage of factoring in the breadth of scenarios in which the player is effective, in that a player effective in many different scenarios will play more minutes per game and therefore contribute more to the team's overall performance, it still fails to distinguish between an ineffective player, an effective \"pinch\"/\"clutch\" offensive player, and a player assuming a primarily defensive role in a position whose title does not necessarily make the nature of their role obvious (e.g., basketball forward and star rebounder Dennis Rodman).\nPPG has also been used as an alternative method for ranking association football teams, particularly during the COVID-19 pandemic, as a way to better compare performance when there is a differential in matches played (and thus traditional point-scoring is unsuitable). Major League Soccer used it to decide the standings for the 2020 season, as some teams had played as little as 15 of their planned 23 regular season matches."]], "corpus": "wikipedia"} {"tstamp": 1722407598.8028, "task_type": "retrieval", "type": "chat", "model": "intfloat/e5-mistral-7b-instruct", "gen_params": {}, "start": 1722407598.394, "finish": 1722407598.8028, "ip": "", "conv_id": "fb0e83c097454b778b24089a2dc2ffdd", "model_name": "intfloat/e5-mistral-7b-instruct", "prompt": "how many points jordan score in NBA", "output": [["how many points jordan score in NBA", "Michael Jordan\n\nLegacy Jordan's talent was clear from his first NBA season; by November 1984, he was being compared to Julius Erving. Larry Bird said that rookie Jordan was the best player he ever saw, and that Jordan was \"one of a kind\", and comparable to Wayne Gretzky as an athlete. In his first game in Madison Square Garden against the New York Knicks, Jordan received a near minute-long standing ovation. After Jordan established the single game playoff record of 63 points against the Boston Celtics on April 20, 1986, Bird described him as \"God disguised as Michael Jordan\".\nJordan led the NBA in scoring in 10 seasons (NBA record) and tied Wilt Chamberlain's record of seven consecutive scoring titles. Jordan was a fixture of the NBA All-Defensive First Team, making the roster nine times (NBA record shared with Gary Payton, Kevin Garnett, and Kobe Bryant). He also holds the top career regular season and playoff scoring averages of 30.1 and 33.4 ppg, respectively. By 1998, the season of his Finals-winning shot against the Jazz, he was well known throughout the league as a clutch performer. In the regular season, Jordan was the Bulls' primary threat in the final seconds of a close game and in the playoffs; he would always ask for the ball at crunch time. Jordan's total of 5,987 points in the playoffs is the second-highest among NBA career playoff scoring leaders. He scored 32,292 points in the regular season, placing him fifth on the NBA all-time scoring list behind LeBron James, Kareem Abdul-Jabbar, Karl Malone, and Bryant."]], "corpus": "wikipedia"} {"tstamp": 1722407598.8028, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722407598.394, "finish": 1722407598.8028, "ip": "", "conv_id": "20b2626082524424be46612f00bedc57", "model_name": "voyage-multilingual-2", "prompt": "how many points jordan score in NBA", "output": [["how many points jordan score in NBA", "List of career achievements by Michael Jordan\n\nHe also scored 49 points on and 63 on in back-to-back games against the Boston Celtics.\nGames scoring 45 or more points, career: 23\nGames scoring 40 or more points, career: 38\nGames scoring 30 or more points, career: 109\nGames scoring 20 or more points, career: 173\nJordan failed to score 20 points only six times in 179 playoff games.\nConsecutive games scoring 20 or more points: 60, to\nAlso holds fourth (see below)\nConsecutive games scoring 10 or more points: 179, to\nJordan's entire playoff career\nConsecutive points scored in a game: 17, during second half (from 73 to 90 points), at New York Knicks,\nBroken by Ray Allen on\nPoints, one postseason: 759 (1992)\nScoring 35 or more points in all games, any playoff series: Twice\n5 games, vs. Cleveland Cavaliers, 1988 First Round\n5 games, vs. Philadelphia 76ers, 1990 Conference Semifinals\nJoined by Jerry West (6 games, Los Angeles Lakers vs. Baltimore Bullets, 1965) and Bernard King (5 games, New York Knicks vs. Detroit Pistons, 1984)\nScoring 30 or more points in all games, any playoff series: Seven times\n3 games, vs. Boston Celtics, 1987 First Round\n5 games, vs. Cleveland Cavaliers, 1988 First Round\n5 games, vs. Cleveland Cavaliers, 1989 First Round"]], "corpus": "wikipedia"} -{"tstamp": 1722408546.1189, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722408545.8764, "finish": 1722408546.1189, "ip": "", "conv_id": "acfe862748ef4696bf6bb2049519ecd5", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "What's the purpose of `if __name__ == '__main__'`?", "output": [["What's the purpose of `if __name__ == '__main__'`?", "Stackoverflow Stackexchange\n\nQ: What does `if name == \"__main__\"` mean in Python? \nPossible Duplicate:\nWhat does do? \n\nI have wrote scripts in Python for quite a while now and I study more of Python as I need it. When reading other people's code I meet if name == \"__main__\": construct quite often. \nWhat is it good for?\n\nA: Basically, \nThere's a distinction between the \"main\" script file and external files which were imported or referenced in another way. If the script is the \"main\" script then the special variable __name__ will equal \"__main__\".\nYou can use this to protect the \"execution\" code from the classes and variables the script has. This enables you to import a script and use classes and variables without actually running that script's main code if it has any.\nSee also: What does if name == “main”: do?\n"]], "corpus": "stackexchange"} +{"tstamp": 1722408546.1189, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722408545.8764, "finish": 1722408546.1189, "ip": "", "conv_id": "acfe862748ef4696bf6bb2049519ecd5", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "What's the purpose of `if __name__ == '__main__'`?", "output": [["What's the purpose of `if __name__ == '__main__'`?", "Stackoverflow Stackexchange\n\nQ: What does `if name == \"__main__\"` mean in Python? \nPossible Duplicate:\nWhat does do? \n\nI have wrote scripts in Python for quite a while now and I study more of Python as I need it. When reading other people's code I meet if name == \"__main__\": construct quite often. \nWhat is it good for?\n\nA: Basically, \nThere's a distinction between the \"main\" script file and external files which were imported or referenced in another way. If the script is the \"main\" script then the special variable __name__ will equal \"__main__\".\nYou can use this to protect the \"execution\" code from the classes and variables the script has. This enables you to import a script and use classes and variables without actually running that script's main code if it has any.\nSee also: What does if name == \u201cmain\u201d: do?\n"]], "corpus": "stackexchange"} {"tstamp": 1722408546.1189, "task_type": "retrieval", "type": "chat", "model": "jinaai/jina-embeddings-v2-base-en", "gen_params": {}, "start": 1722408545.8764, "finish": 1722408546.1189, "ip": "", "conv_id": "a971d6d6aca3462bb6054b2089e3d4d7", "model_name": "jinaai/jina-embeddings-v2-base-en", "prompt": "What's the purpose of `if __name__ == '__main__'`?", "output": [["What's the purpose of `if __name__ == '__main__'`?", "Stackoverflow Stackexchange\n\nQ: html dangerous tags to avoid while developing a chat application I am developing a chat application using PHP and jQuery... all messages sent by the chat pair are appended to a
. While doing this, I found that tags such as