Scheduled Commit

data/retrieval_battle-8abaf4d0-63ed-4c7b-8187-1a073a1031a0.jsonl

@@ -10,3 +10,4 @@
 {"tstamp": 1725636873.093, "task_type": "retrieval", "type": "leftvote", "models": ["", ""], "ip": "", "0_conv_id": "b0853e75f2de498b81c77029ff1e5a76", "0_model_name": "text-embedding-3-large", "0_prompt": "Explain to a high-schooler what is self-attention mechanism?", "0_output": [["Explain to a high-schooler what is self-attention mechanism?", "Attention (machine learning)\n\nPredecessors Predecessors of the mechanism were used in recurrent neural networks which, however, calculated \"soft\" weights sequentially and, at each step, considered the current word and other words within the context window. They were known as multiplicative modules, sigma pi units, and hyper-networks. They have been used in long short-term memory (LSTM) networks, multi-sensory data processing (sound, images, video, and text) in perceivers, fast weight controller's memory, reasoning tasks in differentiable neural computers, and neural Turing machines.\nCore calculations\nThe attention network was designed to identify the highest correlations amongst words within a sentence, assuming that it has learned those patterns from the training corpus.  This correlation is captured in neuronal weights through backpropagation, either from self-supervised pretraining or supervised fine-tuning.\nThe example below (an encoder-only QKV variant of an attention network) shows how correlations are identified once a network has been trained and has the right weights.  When looking at the word \"that\" in the sentence \"see that girl run\", the network should be able to identify \"girl\" as a highly correlated word.  For simplicity this example focuses on the word \"that\", but in reality all words receive this treatment in parallel and the resulting soft-weights and context vectors are stacked into matrices for further task-specific use."]], "0_corpus": "wikipedia", "1_conv_id": "078f99e6df2c4bc5ac74bb4a521f7c89", "1_model_name": "nomic-ai/nomic-embed-text-v1.5", "1_prompt": "Explain to a high-schooler what is self-attention mechanism?", "1_output": [["Explain to a high-schooler what is self-attention mechanism?", "Attention\n\nAttention in social contexts Social attention is one special form of attention that involves the allocation of limited processing resources in a social context. Previous studies on social attention often regard how attention is directed toward socially relevant stimuli such as faces and gaze directions of other individuals. In contrast to attending-to-others, a different line of researches has shown that self-related information such as own face and name automatically captures attention and is preferentially processed comparing to other-related information. These contrasting effects between attending-to-others and attending-to-self prompt a synthetic view in a recent Opinion article proposing that social attention operates at two polarizing states: In one extreme, individual tends to attend to the self and prioritize self-related information over others', and, in the other extreme, attention is allocated to other individuals to infer their intentions and desires. Attending-to-self and attending-to-others mark the two ends of an otherwise continuum spectrum of social attention. For a given behavioral context, the mechanisms underlying these two polarities might interact and compete with each other in order to determine a saliency map of social attention that guides our behaviors. An imbalanced competition between these two behavioral and cognitive processes will cause cognitive disorders and neurological symptoms such as autism spectrum disorders and Williams syndrome."]], "1_corpus": "wikipedia"}
 {"tstamp": 1725640458.5335, "task_type": "retrieval", "type": "bothbadvote", "models": ["", ""], "ip": "", "0_conv_id": "4674012a4897421b8e8a88c5ac5e922f", "0_model_name": "mixedbread-ai/mxbai-embed-large-v1", "0_prompt": "In English when to use –, — vs -?", "0_output": [["In English when to use –, — vs -?", "English Stackexchange\n\nQ: When to use -, – and —? \nPossible Duplicate:\nWhen should I use an em-dash, an en-dash, and a hyphen? \n\nThis is about hyphens (-), en-dashes (–) and em-dashes (—).\nWhen to use which one? To be honest, I always use em-dashes unless I join words with a hyphen, but I never use an en-dash.\n\nA: Hyphens are used in compound modifiers.\n\n\n*\n\n*a well-trained professional\n\n*over-the-counter drugs\n\n\nEn dashes are used in place of \"to\" to connect numbers or words.\n\n\n*\n\n*The London–Amsterdam flight is delayed.\n\n*The meeting is from 3:00–4:30 p.m.\n\n\nEm dashes help set off amplifying or explanatory statements.\n\n\n*\n\n*The vampire—it had been awake half the night—flew out of its coffin.\n\n"]], "0_corpus": "stackexchange", "1_conv_id": "a9fa621b838044cd97b0988c812787be", "1_model_name": "intfloat/e5-mistral-7b-instruct", "1_prompt": "In English when to use –, — vs -?", "1_output": [["In English when to use –, — vs -?", "English Stackexchange\n\nQ: When to use -, – and —? \nPossible Duplicate:\nWhen should I use an em-dash, an en-dash, and a hyphen? \n\nThis is about hyphens (-), en-dashes (–) and em-dashes (—).\nWhen to use which one? To be honest, I always use em-dashes unless I join words with a hyphen, but I never use an en-dash.\n\nA: Hyphens are used in compound modifiers.\n\n\n*\n\n*a well-trained professional\n\n*over-the-counter drugs\n\n\nEn dashes are used in place of \"to\" to connect numbers or words.\n\n\n*\n\n*The London–Amsterdam flight is delayed.\n\n*The meeting is from 3:00–4:30 p.m.\n\n\nEm dashes help set off amplifying or explanatory statements.\n\n\n*\n\n*The vampire—it had been awake half the night—flew out of its coffin.\n\n"]], "1_corpus": "stackexchange"}
 {"tstamp": 1725643405.3301, "task_type": "retrieval", "type": "tievote", "models": ["", ""], "ip": "", "0_conv_id": "b7c6b2f9d1a846a992258ca2e0c6e352", "0_model_name": "voyage-multilingual-2", "0_prompt": "how to use re-ranking to improve the quality of output of a RAG system?", "0_output": [["how to use re-ranking to improve the quality of output of a RAG system?", "Title: Stochastic Rank Aggregation\n\nAbstract: This paper addresses the problem of rank aggregation, which aims to find a consensus ranking among multiple ranking inputs. Traditional rank aggregation methods are deterministic, and can be categorized into explicit and implicit methods depending on whether rank information is explicitly or implicitly utilized. Surprisingly, experimental results on real data sets show that explicit rank aggregation methods would not work as well as implicit methods, although rank information is critical for the task. Our analysis indicates that the major reason might be the unreliable rank information from incomplete ranking inputs. To solve this problem, we propose to incorporate uncertainty into rank aggregation and tackle the problem in both unsupervised and supervised scenario. We call this novel framework {stochastic rank aggregation} (St.Agg for short). Specifically, we introduce a prior distribution on ranks, and transform the ranking functions or objectives in traditional explicit methods to their expectations over this distribution. Our experiments on benchmark data sets show that the proposed St.Agg outperforms the baselines in both unsupervised and supervised scenarios."]], "0_corpus": "arxiv", "1_conv_id": "e2577bdeb6df4e859f219a9208898703", "1_model_name": "nomic-ai/nomic-embed-text-v1.5", "1_prompt": "how to use re-ranking to improve the quality of output of a RAG system?", "1_output": [["how to use re-ranking to improve the quality of output of a RAG system?", "Title: Don't Forget to Connect! Improving RAG with Graph-based Reranking\n\nAbstract: Retrieval Augmented Generation (RAG) has greatly improved the performance of Large Language Model (LLM) responses by grounding generation with context from existing documents. These systems work well when documents are clearly relevant to a question context. But what about when a document has partial information, or less obvious connections to the context? And how should we reason about connections between documents? In this work, we seek to answer these two core questions about RAG generation. We introduce G-RAG, a reranker based on graph neural networks (GNNs) between the retriever and reader in RAG. Our method combines both connections between documents and semantic information (via Abstract Meaning Representation graphs) to provide a context-informed ranker for RAG. G-RAG outperforms state-of-the-art approaches while having smaller computational footprint. Additionally, we assess the performance of PaLM 2 as a reranker and find it to significantly underperform G-RAG. This result emphasizes the importance of reranking for RAG even when using Large Language Models."]], "1_corpus": "arxiv"}
+{"tstamp": 1725646069.3566, "task_type": "retrieval", "type": "leftvote", "models": ["", ""], "ip": "", "0_conv_id": "560af02f403644c9a3145b3cba22fb16", "0_model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "0_prompt": "who is a marketing manager in a company", "0_output": [["who is a marketing manager in a company", "Chief marketing officer\n\nA chief marketing officer (CMO), also called a global marketing officer or marketing director, or chief brand officer, is a corporate executive responsible for managing marketing activities in an organization. Whilst historically these titles may have signified a legal responsibility, for example at Companies House in the UK, the titles are less strict/formal in the 21st Century and allow companies to acknowledge the evolving and increasingly significant role that marketers can play in an organisation, not least because of the inherent character of successful marketers. The CMO leads brand management, marketing communications (including advertising, promotions and public relations), market research, product marketing, distribution channel management, pricing, customer success, and customer service.\nThe CMO is a member of the C-suite and typically reports to the chief executive officer. A number of senior vice presidents, vice presidents, directors, and other senior marketing managers responsible for various parts of the marketing strategy may report directly to the CMO.\nA study from consulting firm Spencer Stuart in 2021 showed that women made up 47% of CMO positions in 2020, an increase from the 43% reported in 2019. 13% of CMOs had racially or ethnically diverse background in 2020, down from 14% in 2019.\nResponsibilities\nThe day-to-day tasks and responsibilities are often categorically different from one another, due to the fluid nature of the CMO's skill set: language is needed to stitch together all aspects of the company. Thus, in a given day the CMO completes tasks that fall into many different categories:\nAnalytical tasks, such as pricing and market research"]], "0_corpus": "wikipedia", "1_conv_id": "75c6d4cd28944c33a1f5fc2565143070", "1_model_name": "embed-english-v3.0", "1_prompt": "who is a marketing manager in a company", "1_output": [["who is a marketing manager in a company", "Marketing management\n\nMarketing management therefore often makes use of various organizational control systems, such as sales forecasts, and sales force and reseller incentive programs, sales force management systems, and customer relationship management tools (CRM). Some software vendors have begun using the term customer data platform or marketing resource management to describe systems that facilitate an integrated approach for controlling marketing resources.  In some cases, these efforts may be linked to various supply chain management systems, such as enterprise resource planning (ERP), material requirements planning (MRP), efficient consumer response (ECR), and inventory management systems.\nInternational marketing management\nGlobalization has led some firms to market beyond the borders of their home countries, making international marketing a part of those firms' marketing strategy. Marketing managers are often responsible for influencing the level, timing, and composition of customer demand. In part, this is because the role of a marketing manager (or sometimes called managing marketer in small- and medium-sized enterprises) can vary significantly based on a business's size, corporate culture, and industry context.\nFor example, in a small- and medium-sized enterprises, the managing marketer may contribute in both managerial and marketing operations roles for the company brands. In a large consumer products company, the marketing manager may act as the overall general manager of his or her assigned product.\nTo create an effective, cost-efficient marketing management strategy, firms must possess a detailed, objective understanding of their own business and the market in which they operate.  In analyzing these issues, the discipline of marketing management often overlaps with the related discipline of strategic planning."]], "1_corpus": "wikipedia"}

data/retrieval_individual-8abaf4d0-63ed-4c7b-8187-1a073a1031a0.jsonl

@@ -37,3 +37,5 @@
 {"tstamp": 1725640501.9295, "task_type": "retrieval", "type": "chat", "model": "Alibaba-NLP/gte-Qwen2-7B-instruct", "gen_params": {}, "start": 1725640501.6009, "finish": 1725640501.9295, "ip": "", "conv_id": "1f1f60432b414525b2dfc25f4f8ae2f1", "model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "prompt": "Секс", "output": [["Секс", "Sex (disambiguation)\n\nSex is the biological distinction of an organism between male and female.\nSex or SEX may also refer to:\nBiology and behaviour\nAnimal sexual behaviour\nCopulation (zoology)\nGroup sex\nHuman female sexuality\nHuman male sexuality\nHuman sexual activity\nNon-penetrative sex, or sexual outercourse\nSex drive, a person's overall sexual drive or desire for sexual activity\nSexual intercourse, also called copulation or coitus\nTransgender sexuality\nGender, the distinction between male and female or masculinity and femininity within an individual's gender identity\nSex–gender distinction\nHuman sexuality\nMating types, a distinction of gametes, whether in anisogamous or isogamous species\nSexing, the act of discerning the sex of an animal\nSexual reproduction, a process of combining and mixing genetic traits, associated with the generation of new individuals, by means of meiosis and fertilization\nGenetic recombination, the process of mixing genetic traits solely, occurring both in organisms with sexual or asexual reproduction\nArt and entertainment\nFilm and television\nSex (1920 film), a film by Fred Niblo\nSex (2024 film), a Norwegian film by Dag Johan Haugerud\n\"Sex\" (Brass Eye), a 1997 television episode\n\"Sex\" (Kath & Kim episode)\nSex (TV series), an Australian television series\nSex: The Annabel Chong Story, a 1999 documentary film\nLiterature\nSex (book), a 1992 book by Madonna and Steven Meisel\nSex (play), a 1926 play by Mae West\nMusic\nSEX, pseudonym of American rapper Young Thug"]], "corpus": "wikipedia"}
 {"tstamp": 1725643374.2974, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1725643373.8849, "finish": 1725643374.2974, "ip": "", "conv_id": "b7c6b2f9d1a846a992258ca2e0c6e352", "model_name": "voyage-multilingual-2", "prompt": "how to use re-ranking to improve the quality of output of a RAG system?", "output": [["how to use re-ranking to improve the quality of output of a RAG system?", "Title: Stochastic Rank Aggregation\n\nAbstract: This paper addresses the problem of rank aggregation, which aims to find a consensus ranking among multiple ranking inputs. Traditional rank aggregation methods are deterministic, and can be categorized into explicit and implicit methods depending on whether rank information is explicitly or implicitly utilized. Surprisingly, experimental results on real data sets show that explicit rank aggregation methods would not work as well as implicit methods, although rank information is critical for the task. Our analysis indicates that the major reason might be the unreliable rank information from incomplete ranking inputs. To solve this problem, we propose to incorporate uncertainty into rank aggregation and tackle the problem in both unsupervised and supervised scenario. We call this novel framework {stochastic rank aggregation} (St.Agg for short). Specifically, we introduce a prior distribution on ranks, and transform the ranking functions or objectives in traditional explicit methods to their expectations over this distribution. Our experiments on benchmark data sets show that the proposed St.Agg outperforms the baselines in both unsupervised and supervised scenarios."]], "corpus": "arxiv"}
 {"tstamp": 1725643374.2974, "task_type": "retrieval", "type": "chat", "model": "nomic-ai/nomic-embed-text-v1.5", "gen_params": {}, "start": 1725643373.8849, "finish": 1725643374.2974, "ip": "", "conv_id": "e2577bdeb6df4e859f219a9208898703", "model_name": "nomic-ai/nomic-embed-text-v1.5", "prompt": "how to use re-ranking to improve the quality of output of a RAG system?", "output": [["how to use re-ranking to improve the quality of output of a RAG system?", "Title: Don't Forget to Connect! Improving RAG with Graph-based Reranking\n\nAbstract: Retrieval Augmented Generation (RAG) has greatly improved the performance of Large Language Model (LLM) responses by grounding generation with context from existing documents. These systems work well when documents are clearly relevant to a question context. But what about when a document has partial information, or less obvious connections to the context? And how should we reason about connections between documents? In this work, we seek to answer these two core questions about RAG generation. We introduce G-RAG, a reranker based on graph neural networks (GNNs) between the retriever and reader in RAG. Our method combines both connections between documents and semantic information (via Abstract Meaning Representation graphs) to provide a context-informed ranker for RAG. G-RAG outperforms state-of-the-art approaches while having smaller computational footprint. Additionally, we assess the performance of PaLM 2 as a reranker and find it to significantly underperform G-RAG. This result emphasizes the importance of reranking for RAG even when using Large Language Models."]], "corpus": "arxiv"}
+{"tstamp": 1725646049.0605, "task_type": "retrieval", "type": "chat", "model": "Alibaba-NLP/gte-Qwen2-7B-instruct", "gen_params": {}, "start": 1725646048.7121, "finish": 1725646049.0605, "ip": "", "conv_id": "560af02f403644c9a3145b3cba22fb16", "model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "prompt": "who is a marketing manager in a company", "output": [["who is a marketing manager in a company", "Chief marketing officer\n\nA chief marketing officer (CMO), also called a global marketing officer or marketing director, or chief brand officer, is a corporate executive responsible for managing marketing activities in an organization. Whilst historically these titles may have signified a legal responsibility, for example at Companies House in the UK, the titles are less strict/formal in the 21st Century and allow companies to acknowledge the evolving and increasingly significant role that marketers can play in an organisation, not least because of the inherent character of successful marketers. The CMO leads brand management, marketing communications (including advertising, promotions and public relations), market research, product marketing, distribution channel management, pricing, customer success, and customer service.\nThe CMO is a member of the C-suite and typically reports to the chief executive officer. A number of senior vice presidents, vice presidents, directors, and other senior marketing managers responsible for various parts of the marketing strategy may report directly to the CMO.\nA study from consulting firm Spencer Stuart in 2021 showed that women made up 47% of CMO positions in 2020, an increase from the 43% reported in 2019. 13% of CMOs had racially or ethnically diverse background in 2020, down from 14% in 2019.\nResponsibilities\nThe day-to-day tasks and responsibilities are often categorically different from one another, due to the fluid nature of the CMO's skill set: language is needed to stitch together all aspects of the company. Thus, in a given day the CMO completes tasks that fall into many different categories:\nAnalytical tasks, such as pricing and market research"]], "corpus": "wikipedia"}
+{"tstamp": 1725646049.0605, "task_type": "retrieval", "type": "chat", "model": "embed-english-v3.0", "gen_params": {}, "start": 1725646048.7121, "finish": 1725646049.0605, "ip": "", "conv_id": "75c6d4cd28944c33a1f5fc2565143070", "model_name": "embed-english-v3.0", "prompt": "who is a marketing manager in a company", "output": [["who is a marketing manager in a company", "Marketing management\n\nMarketing management therefore often makes use of various organizational control systems, such as sales forecasts, and sales force and reseller incentive programs, sales force management systems, and customer relationship management tools (CRM). Some software vendors have begun using the term customer data platform or marketing resource management to describe systems that facilitate an integrated approach for controlling marketing resources.  In some cases, these efforts may be linked to various supply chain management systems, such as enterprise resource planning (ERP), material requirements planning (MRP), efficient consumer response (ECR), and inventory management systems.\nInternational marketing management\nGlobalization has led some firms to market beyond the borders of their home countries, making international marketing a part of those firms' marketing strategy. Marketing managers are often responsible for influencing the level, timing, and composition of customer demand. In part, this is because the role of a marketing manager (or sometimes called managing marketer in small- and medium-sized enterprises) can vary significantly based on a business's size, corporate culture, and industry context.\nFor example, in a small- and medium-sized enterprises, the managing marketer may contribute in both managerial and marketing operations roles for the company brands. In a large consumer products company, the marketing manager may act as the overall general manager of his or her assigned product.\nTo create an effective, cost-efficient marketing management strategy, firms must possess a detailed, objective understanding of their own business and the market in which they operate.  In analyzing these issues, the discipline of marketing management often overlaps with the related discipline of strategic planning."]], "corpus": "wikipedia"}