Scheduled Commit

data/retrieval_battle-5b78c89f-244c-4bc4-9693-2c3c7acb6977.jsonl

@@ -28,3 +28,5 @@
 {"tstamp": 1723094744.5351, "task_type": "retrieval", "type": "tievote", "models": ["", ""], "ip": "", "0_conv_id": "9943bf1b8ea34a4ea2826a1621c8ef09", "0_model_name": "text-embedding-004", "0_prompt": "what are line codes", "0_output": [["what are line codes", "Line code\n\nIn telecommunication, a line code is a pattern of voltage, current, or photons used to represent digital data transmitted down a communication channel or written to a storage medium. This repertoire of signals is usually called a constrained code in data storage systems.\nSome signals are more prone to error than others as the physics of the communication channel or storage medium constrains the repertoire of signals that can be used reliably.\nCommon line encodings are unipolar, polar, bipolar, and Manchester code."]], "0_corpus": "wikipedia", "1_conv_id": "2aabc28c73954e7a905ada9f9fb32fc1", "1_model_name": "mixedbread-ai/mxbai-embed-large-v1", "1_prompt": "what are line codes", "1_output": [["what are line codes", "Line code\n\nIn telecommunication, a line code is a pattern of voltage, current, or photons used to represent digital data transmitted down a communication channel or written to a storage medium. This repertoire of signals is usually called a constrained code in data storage systems.\nSome signals are more prone to error than others as the physics of the communication channel or storage medium constrains the repertoire of signals that can be used reliably.\nCommon line encodings are unipolar, polar, bipolar, and Manchester code."]], "1_corpus": "wikipedia"}
 {"tstamp": 1723094787.1293, "task_type": "retrieval", "type": "rightvote", "models": ["", ""], "ip": "", "0_conv_id": "c9d70ad9cbbb43ce92b90b1575a7fdc6", "0_model_name": "BAAI/bge-large-en-v1.5", "0_prompt": "what type of coding scheme is used to prepare data for transmission or storage on a physical medium", "0_output": [["what type of coding scheme is used to prepare data for transmission or storage on a physical medium", "Physical layer\n\nThe physical layer performs bit-by-bit or symbol-by-symbol data delivery over a physical transmission medium. It provides a standardized interface to the transmission medium, including a mechanical specification of electrical connectors and cables, for example maximum cable length, an electrical specification of transmission line signal level and impedance. The physical layer is responsible for electromagnetic compatibility including electromagnetic spectrum frequency allocation and specification of signal strength, analog bandwidth, etc. The transmission medium may be electrical or optical  over optical fiber or a wireless communication link such as free-space optical communication or radio.\nLine coding is used to convert data into a pattern of electrical fluctuations which may be modulated onto a carrier wave or infrared light. The flow of data is managed with bit synchronization in synchronous serial communication or start-stop signalling and flow control in asynchronous serial communication. Sharing of the transmission medium among multiple network participants can be handled by simple circuit switching or multiplexing. More complex medium access control protocols for sharing the transmission medium may use carrier sense and collision detection such as in Ethernet's Carrier-sense multiple access with collision detection (CSMA/CD).\nTo optimize reliability and efficiency, signal processing techniques such as equalization, training sequences and pulse shaping may be used. Error correction codes and techniques including forward error correction may be applied to further improve reliability."]], "0_corpus": "wikipedia", "1_conv_id": "4a10a067951246b3b9caec626adc8f8b", "1_model_name": "BM25", "1_prompt": "what type of coding scheme is used to prepare data for transmission or storage on a physical medium", "1_output": [["what type of coding scheme is used to prepare data for transmission or storage on a physical medium", "Line code\n\nTransmission and storage After line coding, the signal is put through a physical communication channel, either a transmission medium or data storage medium. The most common physical channels are:\nthe line-coded signal can directly be put on a transmission line, in the form of variations of the voltage or current (often using differential signaling).\nthe line-coded signal (the baseband signal) undergoes further pulse shaping (to reduce its frequency bandwidth) and then is modulated (to shift its frequency) to create an RF signal that can be sent through free space.\nthe line-coded signal can be used to turn on and off a light source in free-space optical communication, most commonly used in an infrared remote control.\nthe line-coded signal can be printed on paper to create a bar code.\nthe line-coded signal can be converted to magnetized spots on a hard drive or tape drive.\nthe line-coded signal can be converted to pits on an optical disc.\nSome of the more common binary line codes include:\nEach line code has advantages and disadvantages. Line codes are chosen to meet one or more of the following criteria:\nMinimize transmission hardware\nFacilitate synchronization\nEase error detection and correction\nAchieve a target spectral density\nEliminate a DC component"]], "1_corpus": "wikipedia"}
 {"tstamp": 1723094855.8037, "task_type": "retrieval", "type": "tievote", "models": ["", ""], "ip": "", "0_conv_id": "3f42d1721e62405b94ad31cf12186e73", "0_model_name": "BAAI/bge-large-en-v1.5", "0_prompt": "which US president created the EPA", "0_output": [["which US president created the EPA", "United States Environmental Protection Agency\n\nThe Environmental Protection Agency (EPA) is an independent agency of the United States government tasked with environmental protection matters. President Richard Nixon proposed the establishment of EPA on July 9, 1970; it began operation on December 2, 1970, after Nixon signed an executive order. The order establishing the EPA was ratified by committee hearings in the House and Senate.\nThe agency is led by its administrator, who is appointed by the president and approved by the Senate. The current administrator is Michael S. Regan. The EPA is not a Cabinet department, but the administrator is normally given cabinet rank. The EPA has its headquarters in Washington, D.C.. There are regional offices for each of the agency's ten regions, as well as 27 laboratories around the country.\nThe agency conducts environmental assessment, research, and education. It has the responsibility of maintaining and enforcing national standards under a variety of environmental laws, in consultation with state, tribal, and local governments. EPA enforcement powers include fines, sanctions, and other measures.\nIt delegates some permitting, monitoring, and enforcement responsibility to U.S. states and the federally recognized tribes. The agency also works with industries and all levels of government in a wide variety of voluntary pollution prevention programs and energy conservation efforts."]], "0_corpus": "wikipedia", "1_conv_id": "2e26300716634d749db0e5d26982b3ef", "1_model_name": "mixedbread-ai/mxbai-embed-large-v1", "1_prompt": "which US president created the EPA", "1_output": [["which US president created the EPA", "United States Environmental Protection Agency\n\nFinally the Nixon administration made the environment a priority in 1969-1971 and set up a series of major agencies headed by the new EPA. The National Environmental Policy Act of 1969 (NEPA) President Nixon signed NEPA into law on January 1, 1970. The law created the Council on Environmental Quality (CEQ) in the Executive Office of the President. NEPA required that a detailed statement of environmental impacts be prepared for all major federal actions significantly affecting the environment. The \"detailed statement\" would ultimately be referred to as an environmental impact statement (EIS).\nEstablishment\nOn July 9, 1970, Nixon proposed an executive reorganization that consolidated many environmental responsibilities of the federal government under one agency, a new Environmental Protection Agency. This proposal included merging pollution control programs from a number of departments, such as the combination of pesticide programs from the United States Department of Agriculture and the United States Department of the Interior. After conducting hearings during that summer, the House and Senate approved the proposal. The EPA was created 90 days before it had to operate, and officially opened its doors on December 2, 1970. The agency's first administrator, William Ruckelshaus, took the oath of office on December 4, 1970."]], "1_corpus": "wikipedia"}
+{"tstamp": 1723101936.302, "task_type": "retrieval", "type": "tievote", "models": ["", ""], "ip": "", "0_conv_id": "705c17f96bcb4cc8ba65b0bc409145db", "0_model_name": "text-embedding-004", "0_prompt": "Quel est le record au 100m ?", "0_output": [["Quel est le record au 100m ?", "100 metres\n\nThe men's world record has been improved upon twelve times since electronic timing became mandatory in 1977. The current men's world record of 9.58 s is held by Usain Bolt of Jamaica, set at the 2009 World Athletics Championships final in Berlin, Germany on 16 August 2009, breaking his own previous world record by 0.11 s. The current women's world record of 10.49 s was set by Florence Griffith-Joyner of the US, at the 1988 United States Olympic Trials in Indianapolis, Indiana, on 16 July 1988 breaking Evelyn Ashford's four-year-old world record by 0.27 seconds. The extraordinary nature of this result and those of several other sprinters in this race raised the possibility of a technical malfunction with the wind gauge which read at 0.0 m/s- a reading which was at complete odds to the windy conditions on the day with high wind speeds being recorded in all other sprints before and after this race as well as the parallel long jump runway at the time of the Griffith-Joyner performance. All scientific studies commissioned by the IAAF and independent organisations since have confirmed there was certainly an illegal tailwind of between 5 m/s – 7 m/s at the time. This should have annulled the legality of this result, although the IAAF has chosen not to take this course of action. The legitimate next best wind legal performance would therefore be Elaine Thompson-Herah's 10.54 second clocking in 2021 at the Prefontaine Classic. Griffith-Joyner's next best legal performance of 10.61 from 1988, would have her third on the all-time list behind Thompson-Herah and Shelly-Ann Fraser-Pryce (10.60).\nSome records have been marred by prohibited drug use – in particular, the scandal at the 1988 Summer Olympics when the winner, Canadian Ben Johnson was stripped of his medal and world record.\nJim Hines, Ronnie Ray Smith and Charles Greene were the first to break the 10-second barrier in the 100 m, all on 20 June 1968, the Night of Speed. Hines also recorded the first legal electronically timed sub-10 second 100 m in winning the 100 metres at the 1968 Olympics. Bob Hayes ran a wind-assisted 9.91 seconds at the 1964 Olympics.\nContinental records"]], "0_corpus": "wikipedia", "1_conv_id": "adf4730893014b129cdcd36798946e69", "1_model_name": "text-embedding-3-large", "1_prompt": "Quel est le record au 100m ?", "1_output": [["Quel est le record au 100m ?", "100 metres\n\nThe men's world record has been improved upon twelve times since electronic timing became mandatory in 1977. The current men's world record of 9.58 s is held by Usain Bolt of Jamaica, set at the 2009 World Athletics Championships final in Berlin, Germany on 16 August 2009, breaking his own previous world record by 0.11 s. The current women's world record of 10.49 s was set by Florence Griffith-Joyner of the US, at the 1988 United States Olympic Trials in Indianapolis, Indiana, on 16 July 1988 breaking Evelyn Ashford's four-year-old world record by 0.27 seconds. The extraordinary nature of this result and those of several other sprinters in this race raised the possibility of a technical malfunction with the wind gauge which read at 0.0 m/s- a reading which was at complete odds to the windy conditions on the day with high wind speeds being recorded in all other sprints before and after this race as well as the parallel long jump runway at the time of the Griffith-Joyner performance. All scientific studies commissioned by the IAAF and independent organisations since have confirmed there was certainly an illegal tailwind of between 5 m/s – 7 m/s at the time. This should have annulled the legality of this result, although the IAAF has chosen not to take this course of action. The legitimate next best wind legal performance would therefore be Elaine Thompson-Herah's 10.54 second clocking in 2021 at the Prefontaine Classic. Griffith-Joyner's next best legal performance of 10.61 from 1988, would have her third on the all-time list behind Thompson-Herah and Shelly-Ann Fraser-Pryce (10.60).\nSome records have been marred by prohibited drug use – in particular, the scandal at the 1988 Summer Olympics when the winner, Canadian Ben Johnson was stripped of his medal and world record.\nJim Hines, Ronnie Ray Smith and Charles Greene were the first to break the 10-second barrier in the 100 m, all on 20 June 1968, the Night of Speed. Hines also recorded the first legal electronically timed sub-10 second 100 m in winning the 100 metres at the 1968 Olympics. Bob Hayes ran a wind-assisted 9.91 seconds at the 1964 Olympics.\nContinental records"]], "1_corpus": "wikipedia"}
+{"tstamp": 1723101985.8806, "task_type": "retrieval", "type": "bothbadvote", "models": ["", ""], "ip": "", "0_conv_id": "ed56e63ad6dc4ba9b73beed4da0e17e9", "0_model_name": "text-embedding-004", "0_prompt": "what are the best practice in single cell analysis ?", "0_output": [["what are the best practice in single cell analysis ?", "Title: Initial recommendations for performing, benchmarking, and reporting single-cell proteomics experiments\n\nAbstract: Analyzing proteins from single cells by tandem mass spectrometry (MS) has become technically feasible. While such analysis has the potential to accurately quantify thousands of proteins across thousands of single cells, the accuracy and reproducibility of the results may be undermined by numerous factors affecting experimental design, sample preparation, data acquisition, and data analysis. Broadly accepted community guidelines and standardized metrics will enhance rigor, data quality, and alignment between laboratories. Here we propose best practices, quality controls, and data reporting recommendations to assist in the broad adoption of reliable quantitative workflows for single-cell proteomics."]], "0_corpus": "arxiv", "1_conv_id": "6320e2d9b35b4fc08fc0257dd4b347fb", "1_model_name": "jinaai/jina-embeddings-v2-base-en", "1_prompt": "what are the best practice in single cell analysis ?", "1_output": [["what are the best practice in single cell analysis ?", "Title: On Two-cavity Coupling\n\nAbstract: This work presents research results on a novel analytical model of electromagnetic systems coupling through small size holes. The key problem regarding coupling of two cavities through an aperture in separating screen of finite thickness without making assumption on smallness of any parameters is considered. We are the first to calculate on the base of rigorous electromagnetic approach the coupling coefficients of the cylindrical cavities within the limit of small aperture and infinitely thin separating screen. The numeric results of electromagnetic characteristic dependencies that have been impossible to perform on the base of previous models are given."]], "1_corpus": "arxiv"}

data/retrieval_individual-5b78c89f-244c-4bc4-9693-2c3c7acb6977.jsonl

@@ -112,3 +112,7 @@
 {"tstamp": 1723099474.8717, "task_type": "retrieval", "type": "chat", "model": "Alibaba-NLP/gte-Qwen2-7B-instruct", "gen_params": {}, "start": 1723099474.5558, "finish": 1723099474.8717, "ip": "", "conv_id": "6ea2c93aba104681adabc1e463204208", "model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "prompt": "Which test was devised to determine whether robots can think?", "output": [["Which test was devised to determine whether robots can think?", "Turing test\n\nThe Turing test, originally called the imitation game by Alan Turing in 1950, is a test of a machine's ability to exhibit intelligent behaviour equivalent to, or indistinguishable from, that of a human. Turing proposed that a human evaluator would judge natural language conversations between a human and a machine designed to generate human-like responses. The evaluator would be aware that one of the two partners in conversation was a machine, and all participants would be separated from one another. The conversation would be limited to a text-only channel, such as a computer keyboard and screen, so the result would not depend on the machine's ability to render words as speech. If the evaluator could not reliably tell the machine from the human, the machine would be said to have passed the test. The test results would not depend on the machine's ability to give correct answers to questions, only on how closely its answers resembled those a human would give. Since the Turing test is a test of indistinguishability in performance capacity, the verbal version generalizes naturally to all of human performance capacity, verbal as well as nonverbal (robotic)."]], "corpus": "wikipedia"}
 {"tstamp": 1723099534.0832, "task_type": "retrieval", "type": "chat", "model": "nomic-ai/nomic-embed-text-v1.5", "gen_params": {}, "start": 1723099533.858, "finish": 1723099534.0832, "ip": "", "conv_id": "5880811a538d4375845ac62d23833282", "model_name": "nomic-ai/nomic-embed-text-v1.5", "prompt": "Please find me the paper on training code large language models to follow instructions via git commits.", "output": [["Please find me the paper on training code large language models to follow instructions via git commits.", "Title: OctoPack: Instruction Tuning Code Large Language Models\n\nAbstract: Finetuning large language models (LLMs) on instructions leads to vast performance improvements on natural language tasks. We apply instruction tuning using code, leveraging the natural structure of Git commits, which pair code changes with human instructions. We compile CommitPack: 4 terabytes of Git commits across 350 programming languages. We benchmark CommitPack against other natural and synthetic code instructions (xP3x, Self-Instruct, OASST) on the 16B parameter StarCoder model, and achieve state-of-the-art performance among models not trained on OpenAI outputs, on the HumanEval Python benchmark (46.2% pass@1). We further introduce HumanEvalPack, expanding the HumanEval benchmark to a total of 3 coding tasks (Code Repair, Code Explanation, Code Synthesis) across 6 languages (Python, JavaScript, Java, Go, C++, Rust). Our models, OctoCoder and OctoGeeX, achieve the best performance across HumanEvalPack among all permissive models, demonstrating CommitPack's benefits in generalizing to a wider set of languages and natural coding tasks. Code, models and data are freely available at https://github.com/bigcode-project/octopack."]], "corpus": "arxiv"}
 {"tstamp": 1723099534.0832, "task_type": "retrieval", "type": "chat", "model": "Alibaba-NLP/gte-Qwen2-7B-instruct", "gen_params": {}, "start": 1723099533.858, "finish": 1723099534.0832, "ip": "", "conv_id": "1871a3a8dd9546aa87435271717b0d63", "model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "prompt": "Please find me the paper on training code large language models to follow instructions via git commits.", "output": [["Please find me the paper on training code large language models to follow instructions via git commits.", "Title: OctoPack: Instruction Tuning Code Large Language Models\n\nAbstract: Finetuning large language models (LLMs) on instructions leads to vast performance improvements on natural language tasks. We apply instruction tuning using code, leveraging the natural structure of Git commits, which pair code changes with human instructions. We compile CommitPack: 4 terabytes of Git commits across 350 programming languages. We benchmark CommitPack against other natural and synthetic code instructions (xP3x, Self-Instruct, OASST) on the 16B parameter StarCoder model, and achieve state-of-the-art performance among models not trained on OpenAI outputs, on the HumanEval Python benchmark (46.2% pass@1). We further introduce HumanEvalPack, expanding the HumanEval benchmark to a total of 3 coding tasks (Code Repair, Code Explanation, Code Synthesis) across 6 languages (Python, JavaScript, Java, Go, C++, Rust). Our models, OctoCoder and OctoGeeX, achieve the best performance across HumanEvalPack among all permissive models, demonstrating CommitPack's benefits in generalizing to a wider set of languages and natural coding tasks. Code, models and data are freely available at https://github.com/bigcode-project/octopack."]], "corpus": "arxiv"}
+{"tstamp": 1723101924.2672, "task_type": "retrieval", "type": "chat", "model": "text-embedding-004", "gen_params": {}, "start": 1723101921.6325, "finish": 1723101924.2672, "ip": "", "conv_id": "705c17f96bcb4cc8ba65b0bc409145db", "model_name": "text-embedding-004", "prompt": "Quel est le record au 100m ?", "output": [["Quel est le record au 100m ?", "100 metres\n\nThe men's world record has been improved upon twelve times since electronic timing became mandatory in 1977. The current men's world record of 9.58 s is held by Usain Bolt of Jamaica, set at the 2009 World Athletics Championships final in Berlin, Germany on 16 August 2009, breaking his own previous world record by 0.11 s. The current women's world record of 10.49 s was set by Florence Griffith-Joyner of the US, at the 1988 United States Olympic Trials in Indianapolis, Indiana, on 16 July 1988 breaking Evelyn Ashford's four-year-old world record by 0.27 seconds. The extraordinary nature of this result and those of several other sprinters in this race raised the possibility of a technical malfunction with the wind gauge which read at 0.0 m/s- a reading which was at complete odds to the windy conditions on the day with high wind speeds being recorded in all other sprints before and after this race as well as the parallel long jump runway at the time of the Griffith-Joyner performance. All scientific studies commissioned by the IAAF and independent organisations since have confirmed there was certainly an illegal tailwind of between 5 m/s – 7 m/s at the time. This should have annulled the legality of this result, although the IAAF has chosen not to take this course of action. The legitimate next best wind legal performance would therefore be Elaine Thompson-Herah's 10.54 second clocking in 2021 at the Prefontaine Classic. Griffith-Joyner's next best legal performance of 10.61 from 1988, would have her third on the all-time list behind Thompson-Herah and Shelly-Ann Fraser-Pryce (10.60).\nSome records have been marred by prohibited drug use – in particular, the scandal at the 1988 Summer Olympics when the winner, Canadian Ben Johnson was stripped of his medal and world record.\nJim Hines, Ronnie Ray Smith and Charles Greene were the first to break the 10-second barrier in the 100 m, all on 20 June 1968, the Night of Speed. Hines also recorded the first legal electronically timed sub-10 second 100 m in winning the 100 metres at the 1968 Olympics. Bob Hayes ran a wind-assisted 9.91 seconds at the 1964 Olympics.\nContinental records"]], "corpus": "wikipedia"}
+{"tstamp": 1723101924.2672, "task_type": "retrieval", "type": "chat", "model": "text-embedding-3-large", "gen_params": {}, "start": 1723101921.6325, "finish": 1723101924.2672, "ip": "", "conv_id": "adf4730893014b129cdcd36798946e69", "model_name": "text-embedding-3-large", "prompt": "Quel est le record au 100m ?", "output": [["Quel est le record au 100m ?", "100 metres\n\nThe men's world record has been improved upon twelve times since electronic timing became mandatory in 1977. The current men's world record of 9.58 s is held by Usain Bolt of Jamaica, set at the 2009 World Athletics Championships final in Berlin, Germany on 16 August 2009, breaking his own previous world record by 0.11 s. The current women's world record of 10.49 s was set by Florence Griffith-Joyner of the US, at the 1988 United States Olympic Trials in Indianapolis, Indiana, on 16 July 1988 breaking Evelyn Ashford's four-year-old world record by 0.27 seconds. The extraordinary nature of this result and those of several other sprinters in this race raised the possibility of a technical malfunction with the wind gauge which read at 0.0 m/s- a reading which was at complete odds to the windy conditions on the day with high wind speeds being recorded in all other sprints before and after this race as well as the parallel long jump runway at the time of the Griffith-Joyner performance. All scientific studies commissioned by the IAAF and independent organisations since have confirmed there was certainly an illegal tailwind of between 5 m/s – 7 m/s at the time. This should have annulled the legality of this result, although the IAAF has chosen not to take this course of action. The legitimate next best wind legal performance would therefore be Elaine Thompson-Herah's 10.54 second clocking in 2021 at the Prefontaine Classic. Griffith-Joyner's next best legal performance of 10.61 from 1988, would have her third on the all-time list behind Thompson-Herah and Shelly-Ann Fraser-Pryce (10.60).\nSome records have been marred by prohibited drug use – in particular, the scandal at the 1988 Summer Olympics when the winner, Canadian Ben Johnson was stripped of his medal and world record.\nJim Hines, Ronnie Ray Smith and Charles Greene were the first to break the 10-second barrier in the 100 m, all on 20 June 1968, the Night of Speed. Hines also recorded the first legal electronically timed sub-10 second 100 m in winning the 100 metres at the 1968 Olympics. Bob Hayes ran a wind-assisted 9.91 seconds at the 1964 Olympics.\nContinental records"]], "corpus": "wikipedia"}
+{"tstamp": 1723101965.8552, "task_type": "retrieval", "type": "chat", "model": "text-embedding-004", "gen_params": {}, "start": 1723101965.5983, "finish": 1723101965.8552, "ip": "", "conv_id": "ed56e63ad6dc4ba9b73beed4da0e17e9", "model_name": "text-embedding-004", "prompt": "what are the best practice in single cell analysis ?", "output": [["what are the best practice in single cell analysis ?", "Title: Initial recommendations for performing, benchmarking, and reporting single-cell proteomics experiments\n\nAbstract: Analyzing proteins from single cells by tandem mass spectrometry (MS) has become technically feasible. While such analysis has the potential to accurately quantify thousands of proteins across thousands of single cells, the accuracy and reproducibility of the results may be undermined by numerous factors affecting experimental design, sample preparation, data acquisition, and data analysis. Broadly accepted community guidelines and standardized metrics will enhance rigor, data quality, and alignment between laboratories. Here we propose best practices, quality controls, and data reporting recommendations to assist in the broad adoption of reliable quantitative workflows for single-cell proteomics."]], "corpus": "arxiv"}
+{"tstamp": 1723101965.8552, "task_type": "retrieval", "type": "chat", "model": "jinaai/jina-embeddings-v2-base-en", "gen_params": {}, "start": 1723101965.5983, "finish": 1723101965.8552, "ip": "", "conv_id": "6320e2d9b35b4fc08fc0257dd4b347fb", "model_name": "jinaai/jina-embeddings-v2-base-en", "prompt": "what are the best practice in single cell analysis ?", "output": [["what are the best practice in single cell analysis ?", "Title: On Two-cavity Coupling\n\nAbstract: This work presents research results on a novel analytical model of electromagnetic systems coupling through small size holes. The key problem regarding coupling of two cavities through an aperture in separating screen of finite thickness without making assumption on smallness of any parameters is considered. We are the first to calculate on the base of rigorous electromagnetic approach the coupling coefficients of the cylindrical cavities within the limit of small aperture and infinitely thin separating screen. The numeric results of electromagnetic characteristic dependencies that have been impossible to perform on the base of previous models are given."]], "corpus": "arxiv"}