Add files using upload-large-folder tool

data/sampled_jsons/'RL_on_Incorrect_Synthetic_Data'_spurious_steps_underlying_cause_'Section_5'_year_2024.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "RL on Incorrect Synthetic Data Scales", "date": "", "ddg_snippet": "Negative Data Identifies Spurious Steps with Advantage Estimates. RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold.", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=9m87e9Keq1", "content": "Negative Data Identifies Spurious Steps with Advantage Estimates. RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold."}
+{"idx": 1, "title": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math...", "date": "", "ddg_snippet": "It addresses the limitations of solely relying on positive examples, which can lead to overfitting on spurious correlations and hinder generalization. By carefully constructing negative data that emphasizes critical intermediate steps in the problem-solving process...", "subpage_snippet": "", "source": "deep-diver.github.io", "link": "https://deep-diver.github.io/neurips2024/posters/9m87e9keq1/", "content": "It addresses the limitations of solely relying on positive examples, which can lead to overfitting on spurious correlations and hinder generalization. By carefully constructing negative data that emphasizes critical intermediate steps in the problem-solving process..."}
+{"idx": 2, "title": "(PDF) RL on Incorrect Synthetic Data Scales the Efficiency of LLM...", "date": "", "ddg_snippet": "positive response co ntains incorrect /irrelevant intermediate steps , training on such data often incentivizes. the model to ov erfit on spurious correlations, l eading to a flat or even inverse scaling with m ore data .", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/381604579_RL_on_Incorrect_Synthetic_Data_Scales_the_Efficiency_of_LLM_Math_Reasoning_by_Eight-Fold", "content": "positive response co ntains incorrect /irrelevant intermediate steps , training on such data often incentivizes. the model to ov erfit on spurious correlations, l eading to a flat or even inverse scaling with m ore data ."}
+{"idx": 3, "title": "Bayesian beagle - RL on Incorrect Synthetic Data Scales the...", "date": "", "ddg_snippet": "Training on positive self-generated synthetic data alone often amplifies the model’s dependence on spurious steps , that erroneously appear to lead to a good solution but do not generalize to novel problems and hurt test performance.", "subpage_snippet": "", "source": "bayesian-beagle.netlify.app", "link": "https://bayesian-beagle.netlify.app/posts/rl_on_incorrect_synthetic_data_scales_the_efficiency_of_llm_math_reasoning_by_eight_fold/2024-06-20-rl_on_incorrect_synthetic_data_scales_the_efficiency_of_llm_math_reasoning_by_eight_fold", "content": "Training on positive self-generated synthetic data alone often amplifies the model’s dependence on spurious steps , that erroneously appear to lead to a good solution but do not generalize to novel problems and hurt test performance."}
+{"idx": 4, "title": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math...", "date": "", "ddg_snippet": "The study investigates the impact of synthetic data , both correct and incorrect , on the fine-tuning of LLMs for enhanced math reasoning using supervised fine-tuning (SFT) and reinforcement learning ( RL ).", "subpage_snippet": "", "source": "www.emergentmind.com", "link": "https://www.emergentmind.com/papers/2406.14532", "content": "The study investigates the impact of synthetic data , both correct and incorrect , on the fine-tuning of LLMs for enhanced math reasoning using supervised fine-tuning (SFT) and reinforcement learning ( RL )."}
+{"idx": 5, "title": "RL on Incorrect Synthetic Data Scales the Efficiency of... | TheaterFire", "date": "", "ddg_snippet": "Training on positive synthetic data alone can lead to suboptimal performance due to the amplification of spurious correlations.", "subpage_snippet": "", "source": "theaterfi.re", "link": "https://theaterfi.re/post/1142871", "content": "Training on positive synthetic data alone can lead to suboptimal performance due to the amplification of spurious correlations."}
+{"idx": 6, "title": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math...", "date": "", "ddg_snippet": "This paper investigates the use of reinforcement learning ( RL ) on incorrect synthetic data to improve the math reasoning abilities of large language models (LLMs).", "subpage_snippet": "", "source": "www.aimodels.fyi", "link": "https://www.aimodels.fyi/papers/arxiv/rl-incorrect-synthetic-data-scales-efficiency-llm", "content": "This paper investigates the use of reinforcement learning ( RL ) on incorrect synthetic data to improve the math reasoning abilities of large language models (LLMs)."}
+{"idx": 7, "title": "AI-Powered Paper Summarization about the arXiv paper 2406.14532 v 1", "date": "", "ddg_snippet": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold.AI-generated Key Points. Authors explore training language models on model-generated synthetic data for math reasoning tasks.", "subpage_snippet": "", "source": "summarizepaper.com", "link": "https://summarizepaper.com/en/arxiv-id/2406.14532v1/", "content": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold.AI-generated Key Points. Authors explore training language models on model-generated synthetic data for math reasoning tasks."}
+{"idx": 8, "title": "Vidhyanand (Vick) Mahase PharmD, PhD. on LinkedIn: RL on ...", "date": "", "ddg_snippet": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold.", "subpage_snippet": "", "source": "www.linkedin.com", "link": "https://www.linkedin.com/posts/vick-mahase-pharmd-phd_rl-on-incorrect-synthetic-data-scales-the-activity-7213284801527582720-_eRW", "content": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold."}
+{"idx": 9, "title": "Reinforcement Learning", "date": "", "ddg_snippet": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold. NeurIPS 2024. 11. Mental Picture: Spurious Steps Derail the Model. prompts. “ Spurious ” steps • Can somehow learn to recover on train. • But not on test, where it will derail. Correct.", "subpage_snippet": "", "source": "cs224r.stanford.edu", "link": "https://cs224r.stanford.edu/slides/10_cs224r-rl_for_reasoning_lecture.pdf", "content": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold. NeurIPS 2024. 11. Mental Picture: Spurious Steps Derail the Model. prompts. “ Spurious ” steps • Can somehow learn to recover on train. • But not on test, where it will derail. Correct."}

data/sampled_jsons/'Towards_Realistic_Semi-Supervised_Learning'_Rizve_Kardan_Shah_ECCV_2022_abstract_Sinkhorn-Knopp_year_2022.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "[2207.02269] Towards Realistic Semi-Supervised Learning", "date": "", "ddg_snippet": "Deep learning is pushing the state-of-the-art in many computer vision applications. However, it relies on large annotated data repositories, and capturing the unconstrained nature of the real-world data is yet to be solved. Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data to reduce annotation cost. The standard SSL approach assumes ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2207.02269", "content": "Deep learning is pushing the state-of-the-art in many computer vision applications. However, it relies on large annotated data repositories, and capturing the unconstrained nature of the real-world data is yet to be solved. Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data to reduce annotation cost. The standard SSL approach assumes ..."}
+{"idx": 1, "title": "PDF Towards Realistic Semi-Supervised Learning - ECVA", "date": "", "ddg_snippet": "Towards Realistic Semi-Supervised Learning Mamshad Nayeem Rizve , Navid Kardan , and Mubarak Shah Center for Research in Computer Vision, UCF, USA {nayeemrizve, kardan}@knights.ucf.edu, shah@crcv.ucf.edu Abstract . Deep learning is pushing the state-of-the-art in many com-puter vision applications.", "subpage_snippet": "", "source": "www.ecva.net", "link": "https://www.ecva.net/papers/eccv_2022/papers_ECCV/papers/136910423.pdf", "content": "Towards Realistic Semi-Supervised Learning Mamshad Nayeem Rizve , Navid Kardan , and Mubarak Shah Center for Research in Computer Vision, UCF, USA {nayeemrizve, kardan}@knights.ucf.edu, shah@crcv.ucf.edu Abstract . Deep learning is pushing the state-of-the-art in many com-puter vision applications."}
+{"idx": 2, "title": "Towards Realistic Semi-supervised Learning | Computer Vision - ECCV 2022", "date": "", "ddg_snippet": "Towards Realistic Semi-supervised Learning Authors: Mamshad Nayeem Rizve , Navid Kardan , Mubarak Shah", "subpage_snippet": "", "source": "dl.acm.org", "link": "https://dl.acm.org/doi/abs/10.1007/978-3-031-19821-2_25", "content": "Towards Realistic Semi-supervised Learning Authors: Mamshad Nayeem Rizve , Navid Kardan , Mubarak Shah"}
+{"idx": 3, "title": "Towards Realistic Semi-supervised Learning | SpringerLink", "date": "", "ddg_snippet": "Towards Realistic Semi-supervised Learning Conference paper First Online: 23 October 2022 pp 437-455 Cite this conference paper Download book PDF Download book EPUB Computer Vision - ECCV 2022 ( ECCV 2022 ) Mamshad Nayeem Rizve , Navid Kardan & Mubarak Shah", "subpage_snippet": "", "source": "link.springer.com", "link": "https://link.springer.com/chapter/10.1007/978-3-031-19821-2_25", "content": "Towards Realistic Semi-supervised Learning Conference paper First Online: 23 October 2022 pp 437-455 Cite this conference paper Download book PDF Download book EPUB Computer Vision - ECCV 2022 ( ECCV 2022 ) Mamshad Nayeem Rizve , Navid Kardan & Mubarak Shah"}
+{"idx": 4, "title": "(PDF) Towards Realistic Semi-Supervised Learning - ResearchGate", "date": "", "ddg_snippet": "Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data to reduce annotation cost.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/361807923_Towards_Realistic_Semi-Supervised_Learning", "content": "Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data to reduce annotation cost."}
+{"idx": 5, "title": "Towards Realistic Semi-Supervised Learning - Semantic Scholar", "date": "", "ddg_snippet": "OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learning Mamshad Nayeem Rizve , Navid Kardan , Salman Khan, F. Khan, M. Shah Computer Science ECCV 2022 TLDR", "subpage_snippet": "", "source": "www.semanticscholar.org", "link": "https://www.semanticscholar.org/paper/Towards-Realistic-Semi-Supervised-Learning-Rizve-Kardan/cce61de63d6691fce13bf1a1c8d231553df12f31/figure/0", "content": "OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learning Mamshad Nayeem Rizve , Navid Kardan , Salman Khan, F. Khan, M. Shah Computer Science ECCV 2022 TLDR"}
+{"idx": 6, "title": "Towards Realistic Semi-Supervised Learning - GitHub", "date": "", "ddg_snippet": "Implementation of Towards Realistic Semi-Supervised Learning . Deep learning is pushing the state-of-the-art in many computer vision applications. However, it relies on large annotated data repositories, and capturing the unconstrained nature of the real-world data is yet to be solved. Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data ...", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/nayeemrizve/TRSSL", "content": "Implementation of Towards Realistic Semi-Supervised Learning . Deep learning is pushing the state-of-the-art in many computer vision applications. However, it relies on large annotated data repositories, and capturing the unconstrained nature of the real-world data is yet to be solved. Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data ..."}
+{"idx": 7, "title": "Towards Realistic Semi-Supervised Learning - NASA/ADS", "date": "", "ddg_snippet": "Deep learning is pushing the state-of-the-art in many computer vision applications. However, it relies on large annotated data repositories, and capturing the unconstrained nature of the real-world data is yet to be solved. Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data to reduce annotation cost. The standard SSL approach assumes ...", "subpage_snippet": "", "source": "ui.adsabs.harvard.edu", "link": "https://ui.adsabs.harvard.edu/abs/2022arXiv220702269N/abstract", "content": "Deep learning is pushing the state-of-the-art in many computer vision applications. However, it relies on large annotated data repositories, and capturing the unconstrained nature of the real-world data is yet to be solved. Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data to reduce annotation cost. The standard SSL approach assumes ..."}
+{"idx": 8, "title": "Towards Realistic Semi-Supervised Learning - arXiv.org", "date": "", "ddg_snippet": "Towards Realistic Semi-Supervised Learning Mamshad Nayeem Rizve , Navid Kardan , and Mubarak Shah Center for Research in Computer Vision, UCF, USA Abstract . Deep learning is pushing the state-of-the-art in many com-puter vision applications.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2207.02269v1", "content": "Towards Realistic Semi-Supervised Learning Mamshad Nayeem Rizve , Navid Kardan , and Mubarak Shah Center for Research in Computer Vision, UCF, USA Abstract . Deep learning is pushing the state-of-the-art in many com-puter vision applications."}
+{"idx": 9, "title": "Towards Realistic Semi-supervised Learning | Computer Vision - ECCV 2022", "date": "", "ddg_snippet": "Abstract Deep learning is pushing the state-of-the-art in many computer vision applications. However, it relies on large annotated data repositories, and capturing the unconstrained nature of the real-world data is yet to be solved. Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data to reduce annotation cost. The standard SSL approach ...", "subpage_snippet": "", "source": "dlnext.acm.org", "link": "https://dlnext.acm.org/doi/abs/10.1007/978-3-031-19821-2_25", "content": "Abstract Deep learning is pushing the state-of-the-art in many computer vision applications. However, it relies on large annotated data repositories, and capturing the unconstrained nature of the real-world data is yet to be solved. Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data to reduce annotation cost. The standard SSL approach ..."}

data/sampled_jsons/1BaC3AdG1i_ATA_Adaptive_Task_Allocation_Distributed_Machine_Learning_GTA_strategy.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Machine learning - Wikipedia", "date": "", "ddg_snippet": "tasks in which machine learning is concerned offers a fundamentally operational definition rather than defining the field in cognitive terms. This follows Alan Turing's proposal in his paper \"Computing Machinery and Intelligence\", in which the question \"Can machines think...", "subpage_snippet": "", "source": "en.wikipedia.org", "link": "https://en.wikipedia.org/wiki/Machine_learning", "content": "tasks in which machine learning is concerned offers a fundamentally operational definition rather than defining the field in cognitive terms. This follows Alan Turing's proposal in his paper \"Computing Machinery and Intelligence\", in which the question \"Can machines think..."}
+{"idx": 1, "title": "Postdoc, KAUST - Cited by 12 - Online learning - Bandits theory", "date": "", "ddg_snippet": "Ata : Adaptive task allocation for efficient resource management in distributed machine learning .Covariance- adaptive best arm identification. EM Saad, G Blanchard, N Verzelen.", "subpage_snippet": "", "source": "scholar.google.com", "link": "https://scholar.google.com/citations?user=Jih_bwsAAAAJ&hl=en", "content": "Ata : Adaptive task allocation for efficient resource management in distributed machine learning .Covariance- adaptive best arm identification. EM Saad, G Blanchard, N Verzelen."}
+{"idx": 2, "title": "ATA : Adaptive Task Allocation", "date": "", "ddg_snippet": "ATA : Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning .", "subpage_snippet": "", "source": "artomaranjyan.github.io", "link": "https://artomaranjyan.github.io/assets/pdf/ATA_Rising_Stars.pdf", "content": "ATA : Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning ."}
+{"idx": 3, "title": "History of Machine Learning : How We Got Here", "date": "", "ddg_snippet": "A machine learning algorithm visualized by MidJourney AI. The roots of machine learning can be traced back to the 1940s, when researchers began to explore very basic pattern recognition and the development of the first neural networks.", "subpage_snippet": "", "source": "www.akkio.com", "link": "https://www.akkio.com/post/history-of-machine-learning", "content": "A machine learning algorithm visualized by MidJourney AI. The roots of machine learning can be traced back to the 1940s, when researchers began to explore very basic pattern recognition and the development of the first neural networks."}
+{"idx": 4, "title": "Online Multi-Robot Coordination and Cooperation with Task ...", "date": "", "ddg_snippet": "Offline Task Allocation Experiments. Experiment 1: Number of Tasks .The online algorithm enables replanning in response to actual task outcomes, making it adaptive to task failures and inaccuracies in the mission model.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2509.15052", "content": "Offline Task Allocation Experiments. Experiment 1: Number of Tasks .The online algorithm enables replanning in response to actual task outcomes, making it adaptive to task failures and inaccuracies in the mission model."}
+{"idx": 5, "title": "machinelearningmastery.com/precision-recall-and-f-measure-for...", "date": "", "ddg_snippet": "Precision And Recall Machine Learning .", "subpage_snippet": "", "source": "machinelearningmastery.com", "link": "https://machinelearningmastery.com/precision-recall-and-f-measure-for-imbalanced-classification/", "content": "Precision And Recall Machine Learning ."}
+{"idx": 6, "title": "10 предложений в Present Simple - примеры... - EnglishMore.ru", "date": "", "ddg_snippet": "English language learning Blog. Главная. Грамматика.", "subpage_snippet": "", "source": "englishmore.ru", "link": "https://englishmore.ru/grammar/10-predlozheniy-v-present-simple-s-perevodom", "content": "English language learning Blog. Главная. Грамматика."}
+{"idx": 7, "title": "Top 10 Big Data Frameworks In 2024 - Jelvix", "date": "", "ddg_snippet": "It has machine - learning capabilities and integration with other popular Big Data frameworks. You can read our article to find out more about machine learning services.Hive’s main competitor Apache Impala is distributed by Cloudera. 5. Storm. Twitter first big data framework.", "subpage_snippet": "", "source": "jelvix.com", "link": "https://jelvix.com/blog/top-5-big-data-frameworks", "content": "It has machine - learning capabilities and integration with other popular Big Data frameworks. You can read our article to find out more about machine learning services.Hive’s main competitor Apache Impala is distributed by Cloudera. 5. Storm. Twitter first big data framework."}
+{"idx": 8, "title": "Axiom", "date": "", "ddg_snippet": "Auto - Strategies . Set your entire trading strategy in motion with a single click.To learn more about our wallet infrastructure, visit our partner Turnkey. Can I buy crypto on Axiom?", "subpage_snippet": "", "source": "axiom.trade", "link": "https://axiom.trade/", "content": "Auto - Strategies . Set your entire trading strategy in motion with a single click.To learn more about our wallet infrastructure, visit our partner Turnkey. Can I buy crypto on Axiom?"}
+{"idx": 9, "title": "SnowRunner: Аляска - карта машин и улучшений | WTFTime.ru", "date": "", "ddg_snippet": "Промокоды Genshin Impact. Русский язык в GTA 6.Когда выйдет GTA 6 на ПК.", "subpage_snippet": "", "source": "wtftime.ru", "link": "https://wtftime.ru/guides/508/snowrunner-aljaska---karta-mashin-i-uluchshenij/", "content": "Промокоды Genshin Impact. Русский язык в GTA 6.Когда выйдет GTA 6 на ПК."}

data/sampled_jsons/1BaC3AdG1i_ATA_Theorem_6.1_regret_bound_vanishing.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "ProvableRegretBounds forDeepOnlineLearningandControl", "date": "", "ddg_snippet": "Motivated by applications in control, we give a general black-box reduction from deep learn-ing to online convex optimization. This allows us to decouple optimization, regret , expressive-ness, and derive agnostic online learning guarantees for fully-connected deep neural networks with ReLU activations. We quantify convergence and regret guarantees for any range of pa-rameters and allow any ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2110.07807.pdf", "content": "Motivated by applications in control, we give a general black-box reduction from deep learn-ing to online convex optimization. This allows us to decouple optimization, regret , expressive-ness, and derive agnostic online learning guarantees for fully-connected deep neural networks with ReLU activations. We quantify convergence and regret guarantees for any range of pa-rameters and allow any ..."}
+{"idx": 1, "title": "High-Probability Regret Bounds for Bandit Online Linear ...", "date": "", "ddg_snippet": "We presented p an algorithm that achieves the desired regret bound of O ( T) with high probability. However, the quest for an efficient algorithm with the same high-probability guar-antee, even for the special case of bandit online shortest paths, is still open.", "subpage_snippet": "", "source": "www.mit.edu", "link": "https://www.mit.edu/~rakhlin/papers/bandit_merged.pdf", "content": "We presented p an algorithm that achieves the desired regret bound of O ( T) with high probability. However, the quest for an efficient algorithm with the same high-probability guar-antee, even for the special case of bandit online shortest paths, is still open."}
+{"idx": 2, "title": "Bandits: Regret Lower Bound and Instance-Dependent Regret", "date": "", "ddg_snippet": "Want to construct a lower bound regret ? on the achievable regret So far we our theoretical analysis has always considered a fixed algorithm and analyzed it (by deriving a regret upper bound with high probability) To get a lower bound , we need to consider what regret could be achieved by algorithm, and show it can’t be better than some rate any", "subpage_snippet": "", "source": "shamulent.github.io", "link": "https://shamulent.github.io/RL_2022/Lectures/Lecture4_prelecture.pdf", "content": "Want to construct a lower bound regret ? on the achievable regret So far we our theoretical analysis has always considered a fixed algorithm and analyzed it (by deriving a regret upper bound with high probability) To get a lower bound , we need to consider what regret could be achieved by algorithm, and show it can’t be better than some rate any"}
+{"idx": 3, "title": "Make the Minority Great Again: First-Order Regret Bound for ...", "date": "", "ddg_snippet": "The more refined concept of first-order p regret bound replaces this with a scal- p ing L , which may be much smaller than T. It is well known that minor variants of standard algorithms satisfy first-order regret bounds in the full information and multi-armed bandit settings.", "subpage_snippet": "", "source": "proceedings.mlr.press", "link": "https://proceedings.mlr.press/v80/allen-zhu18b/allen-zhu18b.pdf", "content": "The more refined concept of first-order p regret bound replaces this with a scal- p ing L , which may be much smaller than T. It is well known that minor variants of standard algorithms satisfy first-order regret bounds in the full information and multi-armed bandit settings."}
+{"idx": 4, "title": "Regret Bounds for Learning State Representations in ...", "date": "", "ddg_snippet": "Note that the bound of Theorem 4 holds for any Markov model in . Thus, in case there is a Markov model with smaller state space the regret bound shows that UCB-MS automatically adapts to this preferable model.", "subpage_snippet": "", "source": "papers.nips.cc", "link": "https://papers.nips.cc/paper/2019/file/9b8b50fb590c590ffbf1295ce92258dc-Paper.pdf", "content": "Note that the bound of Theorem 4 holds for any Markov model in . Thus, in case there is a Markov model with smaller state space the regret bound shows that UCB-MS automatically adapts to this preferable model."}
+{"idx": 5, "title": "KL-UCB-Switch: Optimal Regret Bounds for Stochastic Bandits ...", "date": "", "ddg_snippet": "The appendix also features the proof of the sophisticated distribution-dependent regret bound of Theorem 2, with an optimal second order term of order ln ln T in the case of a known T (Appendix C).", "subpage_snippet": "", "source": "jmlr.org", "link": "https://jmlr.org/papers/volume23/20-717/20-717.pdf", "content": "The appendix also features the proof of the sophisticated distribution-dependent regret bound of Theorem 2, with an optimal second order term of order ln ln T in the case of a known T (Appendix C)."}
+{"idx": 6, "title": "CS292FStatRLLecture 8 Exploration in Bandits", "date": "", "ddg_snippet": "T 13 / 21 Plan of the proof 1. First prove the Proposition that bounds the sum of square regret • By bounding instantaneous regret e um of squares with “Information 2. Prove the uniform confidence bound", "subpage_snippet": "", "source": "cseweb.ucsd.edu", "link": "https://cseweb.ucsd.edu/~yuxiangw/classes/RLCourse-2021Spring/Lectures/LinearBandits_annotated.pdf", "content": "T 13 / 21 Plan of the proof 1. First prove the Proposition that bounds the sum of square regret • By bounding instantaneous regret e um of squares with “Information 2. Prove the uniform confidence bound"}
+{"idx": 7, "title": "The Four Pillars", "date": "", "ddg_snippet": "The theorem about concurrence of medians generalizes beautifully to three di-mensions, where the gure corresponding to a triangle is a tetrahedron: a solid with four vertices joined by six lines that bound the tetrahedron’s four triangular faces (Figure 4.7).", "subpage_snippet": "", "source": "johnval.nl", "link": "https://johnval.nl/school/wiskunde/wiskundeD/gratis_studieboeken/FourPillarsOfGeometry.pdf", "content": "The theorem about concurrence of medians generalizes beautifully to three di-mensions, where the gure corresponding to a triangle is a tetrahedron: a solid with four vertices joined by six lines that bound the tetrahedron’s four triangular faces (Figure 4.7)."}
+{"idx": 8, "title": "Горящие туры из Москвы 2025– Египет, Турция, ОАЭ до -70...", "date": "", "ddg_snippet": "Тысячи предложений от надежных туроператоров. Без переплаты и поездок в офис.", "subpage_snippet": "", "source": "travelata.ru", "link": "https://travelata.ru/tury", "content": "Тысячи предложений от надежных туроператоров. Без переплаты и поездок в офис."}
+{"idx": 9, "title": "Номер 1, страница 7 - гдз по английскому языку 6 класс (spotlight)...", "date": "", "ddg_snippet": "Английский язык (english), 6 класс Рабочая тетрадь (workbook), авторы: Ваулина Юлия Евгеньевна (Vaulina Julia), Дули Дженни (Dooley Jenny), Подоляко Ольга Евгеньевна (Podolyako Olga), Эванс Вирджиния (Evans Virginia), издательство Просвещение, Москва...", "subpage_snippet": "", "source": "gdz.top", "link": "https://gdz.top/6-klass/english/vaulina-spotlight-rabochaja-tetrad/01-3-1", "content": "Английский язык (english), 6 класс Рабочая тетрадь (workbook), авторы: Ваулина Юлия Евгеньевна (Vaulina Julia), Дули Дженни (Dooley Jenny), Подоляко Ольга Евгеньевна (Podolyako Olga), Эванс Вирджиния (Evans Virginia), издательство Просвещение, Москва..."}

data/sampled_jsons/2.5_Score_Based_Denoising_Training_with_noise_augmentation_introduces_posterior_mean.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Adaptive and Iterative Point Cloud Denoising with Score‐ ...", "date": "", "ddg_snippet": "21 Apr 2025 — In this paper, we propose an adaptive and iterative point cloud denoising method based on the score - based diffusion model.", "subpage_snippet": "", "source": "onlinelibrary.wiley.com", "link": "https://onlinelibrary.wiley.com/doi/10.1111/cgf.70149?af=R", "content": "21 Apr 2025 — In this paper, we propose an adaptive and iterative point cloud denoising method based on the score - based diffusion model."}
+{"idx": 1, "title": "DiffAug: A Diffuse-and-Denoise Augmentation for Training ...", "date": "", "ddg_snippet": "9 Dec 2024 — We introduce DiffAug , a simple and efficient diffusion-based augmentation technique to train image classifiers for the crucial yet challenging goal of improved ...", "subpage_snippet": "", "source": "neurips.cc", "link": "https://neurips.cc/virtual/2024/poster/95014", "content": "9 Dec 2024 — We introduce DiffAug , a simple and efficient diffusion-based augmentation technique to train image classifiers for the crucial yet challenging goal of improved ..."}
+{"idx": 2, "title": "TRAINING DIFFUSION CLASSIFIERS WITH DENOISING ...", "date": "", "ddg_snippet": "In this work, we focus on supervised and semi-supervised training of classifier-guided diffusion models that utilize the gradients of classifier with respect to ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf/1d590a6e3e5fe6af67636df247d95d431f17631c.pdf", "content": "In this work, we focus on supervised and semi-supervised training of classifier-guided diffusion models that utilize the gradients of classifier with respect to ..."}
+{"idx": 3, "title": "CVPR Poster Transfer CLIP for Generalizable Image Denoising", "date": "", "ddg_snippet": "These datasets comprise natural noisy sRGB images from smartphones and commercial cameras. During training , we simulate sRGB noise based on DIV2K dataset [2] ...", "subpage_snippet": "", "source": "cvpr.thecvf.com", "link": "https://cvpr.thecvf.com/virtual/2024/poster/31724", "content": "These datasets comprise natural noisy sRGB images from smartphones and commercial cameras. During training , we simulate sRGB noise based on DIV2K dataset [2] ..."}
+{"idx": 4, "title": "Enhancing signal-to-noise ratio in real-time LED-based ...", "date": "", "ddg_snippet": "by A Paul · 2025 · Cited by 6 — In this study, we systematically assess the efficacy of various Encoder-Decoder- based CNN architectures for enhancing SNR in real-time LED- based PA imaging.", "subpage_snippet": "", "source": "www.sciencedirect.com", "link": "https://www.sciencedirect.com/science/article/pii/S2213597924000910", "content": "by A Paul · 2025 · Cited by 6 — In this study, we systematically assess the efficacy of various Encoder-Decoder- based CNN architectures for enhancing SNR in real-time LED- based PA imaging."}
+{"idx": 5, "title": "Self-supervised learning for CT image denoising and ...", "date": "", "ddg_snippet": "12 Sept 2024 — Note that the posterior mean in (34) requires the noise model p ( y J | x J ) and the posterior distribution p ( x J | Ω J ) given the context.", "subpage_snippet": "", "source": "pmc.ncbi.nlm.nih.gov", "link": "https://pmc.ncbi.nlm.nih.gov/articles/PMC11502646/", "content": "12 Sept 2024 — Note that the posterior mean in (34) requires the noise model p ( y J | x J ) and the posterior distribution p ( x J | Ω J ) given the context."}
+{"idx": 6, "title": "Posterior Transition Modeling for Unsupervised Diffusion- ...", "date": "", "ddg_snippet": "by M Sadeghi · 2025 — Abstract—We explore unsupervised speech enhancement using diffusion models as expressive generative priors for clean speech.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2507.02391", "content": "by M Sadeghi · 2025 — Abstract—We explore unsupervised speech enhancement using diffusion models as expressive generative priors for clean speech."}
+{"idx": 7, "title": "Denoising Diffusion-Augmented Hybrid Video Anomaly ...", "date": "", "ddg_snippet": "Video Anomaly Detection (VAD) aims to detect unexpected events in various applications, such as security and surveil- lance, industrial safety, transportation, ...", "subpage_snippet": "", "source": "www.ijcai.org", "link": "https://www.ijcai.org/proceedings/2024/0077.pdf", "content": "Video Anomaly Detection (VAD) aims to detect unexpected events in various applications, such as security and surveil- lance, industrial safety, transportation, ..."}
+{"idx": 8, "title": "Normalizing Flows are Capable Generative Models", "date": "", "ddg_snippet": "6 Jun 2025 — 2.5 Score Based Denoising Training with noise augmentation introduces an additional challenge: models trained on the noisy distribution q ( y ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2412.06329v3", "content": "6 Jun 2025 — 2.5 Score Based Denoising Training with noise augmentation introduces an additional challenge: models trained on the noisy distribution q ( y ..."}
+{"idx": 9, "title": "Normalizing Flows are Capable Generative Models", "date": "", "ddg_snippet": "2.5 . Score Based Denoising . Training with noise augmentation introduces an additional challenge: models trained on the noisy distribution q(y) naturally ...", "subpage_snippet": "", "source": "icml.cc", "link": "https://icml.cc/virtual/2025/poster/46564", "content": "2.5 . Score Based Denoising . Training with noise augmentation introduces an additional challenge: models trained on the noisy distribution q(y) naturally ..."}

data/sampled_jsons/2024_critical_windows_language_models_reasoning_arxiv.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Reasoning language model - Wikipedia", "date": "", "ddg_snippet": "Reasoning language models are large language models that are trained further to solve tasks that take several steps of reasoning . They tend to do better on logic, math, and programming tasks than standard LLMs, can revisit and revise earlier steps...", "subpage_snippet": "", "source": "en.wikipedia.org", "link": "https://en.wikipedia.org/wiki/Reasoning_language_model", "content": "Reasoning language models are large language models that are trained further to solve tasks that take several steps of reasoning . They tend to do better on logic, math, and programming tasks than standard LLMs, can revisit and revise earlier steps..."}
+{"idx": 1, "title": "Claude ( language model ) - Wikipedia", "date": "", "ddg_snippet": "Claude is a family of large language models developed by Anthropic. The first model , Claude, was released in March 2023. The Claude 3 family, released in March 2024 , consists of three models : Haiku, optimized for speed; Sonnet, which balances capabil...", "subpage_snippet": "", "source": "en.m.wikipedia.org", "link": "https://en.m.wikipedia.org/wiki/Claude_(language_model)", "content": "Claude is a family of large language models developed by Anthropic. The first model , Claude, was released in March 2023. The Claude 3 family, released in March 2024 , consists of three models : Haiku, optimized for speed; Sonnet, which balances capabil..."}
+{"idx": 2, "title": "The CLRS-Text Algorithmic Reasoning Language Benchmark", "date": "", "ddg_snippet": "Eliciting reasoning capabilities from language models (LMs) is a critical direction on the path towards building intelligent systems.Gemma: Open models based on gemini research and technology. arXiv preprint arXiv :2403.08295, 2024 .", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2406.04229v1", "content": "Eliciting reasoning capabilities from language models (LMs) is a critical direction on the path towards building intelligent systems.Gemma: Open models based on gemini research and technology. arXiv preprint arXiv :2403.08295, 2024 ."}
+{"idx": 3, "title": "Training Large Language Models to Reason in a", "date": "", "ddg_snippet": "Large language models (LLMs) are restricted to reason in the “ language space”, where they typically express the reasoning process with a chain-of-thought (CoT) to solve a complex reasoning problem. However, we argue that language space may not always be optimal for reasoning .", "subpage_snippet": "", "source": "gptrack.ai", "link": "https://gptrack.ai/paper/Coconut.pdf", "content": "Large language models (LLMs) are restricted to reason in the “ language space”, where they typically express the reasoning process with a chain-of-thought (CoT) to solve a complex reasoning problem. However, we argue that language space may not always be optimal for reasoning ."}
+{"idx": 4, "title": "(PDF) The CLRS-Text Algorithmic Reasoning Language Benchmark", "date": "", "ddg_snippet": "Eliciting reasoning capabilities from language models (LMs) is a critical direction on the path towards building intelligent systems. arXiv :2406.04229v1 [cs.LG] 6 Jun 2024 . The CLRS-Text Algorithmic Reasoning Language Benchmark.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/381227056_The_CLRS-Text_Algorithmic_Reasoning_Language_Benchmark", "content": "Eliciting reasoning capabilities from language models (LMs) is a critical direction on the path towards building intelligent systems. arXiv :2406.04229v1 [cs.LG] 6 Jun 2024 . The CLRS-Text Algorithmic Reasoning Language Benchmark."}
+{"idx": 5, "title": "Understanding Chain-of-Thought Reasoning in Large Language ...", "date": "", "ddg_snippet": "Large Language Models (LLMs) like GPT-4 and Claude 3 have transformed natural language processing by enabling sophisticated reasoning , planning, and problem-solving capabilities.", "subpage_snippet": "", "source": "www.linkedin.com", "link": "https://www.linkedin.com/pulse/understanding-chain-of-thought-reasoning-large-models-vispi-karkaria-wlljc", "content": "Large Language Models (LLMs) like GPT-4 and Claude 3 have transformed natural language processing by enabling sophisticated reasoning , planning, and problem-solving capabilities."}
+{"idx": 6, "title": "What Are Large Language Models (LLMs)? | IBM", "date": "", "ddg_snippet": "Large language models (LLMs) are a category of deep learning models trained on immense amounts of data, making them capable of understanding and generating natural language and other types of content to perform a wide range of tasks.", "subpage_snippet": "", "source": "www.ibm.com", "link": "https://www.ibm.com/think/topics/large-language-models", "content": "Large language models (LLMs) are a category of deep learning models trained on immense amounts of data, making them capable of understanding and generating natural language and other types of content to perform a wide range of tasks."}
+{"idx": 7, "title": "AI Reasoning Is Exploding in 2025 — What’s Hype, What’s Real, and...", "date": "", "ddg_snippet": "Formal, verifiable reasoning . DeepMind paired language models with symbolic proof systems. In July 2024 , AlphaProof + AlphaGeometry 2 solved four of six International Mathematical Olympiad problems—28/42 points, equivalent to an IMO silver medal.", "subpage_snippet": "", "source": "informacja-lokalna.pl", "link": "https://informacja-lokalna.pl/2025/09/23/ai-reasoning-is-exploding-in-2025-whats-hype-whats-real-and-how-it-changes-everything/", "content": "Formal, verifiable reasoning . DeepMind paired language models with symbolic proof systems. In July 2024 , AlphaProof + AlphaGeometry 2 solved four of six International Mathematical Olympiad problems—28/42 points, equivalent to an IMO silver medal."}
+{"idx": 8, "title": "QwenLM/Qwen3-VL: Qwen3-VL is the multimodal large language ...", "date": "", "ddg_snippet": "2024 .08.30: We have released the Qwen2-VL series. The 2B and 7B models are now available, and the 72B model for open source is coming soon.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/QwenLM/Qwen3-VL", "content": "2024 .08.30: We have released the Qwen2-VL series. The 2B and 7B models are now available, and the 72B model for open source is coming soon."}
+{"idx": 9, "title": "Proceedings of the 2024 Conference on Empirical Methods in Natural...", "date": "", "ddg_snippet": "2024 a. Conceptual and unbiased reasoning in language models . arXiv preprint arXiv :2404.00205. Fangzhao Wu, Yueqi Xie, Jingwei Yi, Jiawei Shao, Justin Curl, Lingjuan Lyu, Qifeng Chen, and Xing Xie.", "subpage_snippet": "", "source": "aclanthology.org", "link": "https://aclanthology.org/2024.emnlp-tutorials.pdf", "content": "2024 a. Conceptual and unbiased reasoning in language models . arXiv preprint arXiv :2404.00205. Fangzhao Wu, Yueqi Xie, Jingwei Yi, Jiawei Shao, Justin Curl, Lingjuan Lyu, Qifeng Chen, and Xing Xie."}

data/sampled_jsons/23zxLtvder_SPD_Sync-Point_Drop_Section_4.1.2_bias_linear_layers_parallelization.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "SPD: Sync-Point Drop for Efficient Tensor Parallelism of ... - OpenReview", "date": "", "ddg_snippet": "In this paper, we present Sync-Point Drop ( SPD ), a novel optimization technique that improves the latency of LLMs on distributed inference systems. By adopting a new block design and separated approaches based on block-wisely identified sensitivity for lack of sync-point , SPD enables eficient deployment across multiple computing units with ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=23zxLtvder", "content": "In this paper, we present Sync-Point Drop ( SPD ), a novel optimization technique that improves the latency of LLMs on distributed inference systems. By adopting a new block design and separated approaches based on block-wisely identified sensitivity for lack of sync-point , SPD enables eficient deployment across multiple computing units with ..."}
+{"idx": 1, "title": "(PDF) SPD: Sync-Point Drop for efficient tensor ... - ResearchGate", "date": "", "ddg_snippet": "Decoder block structure with sync-point drop (in 2-GPUs distributed inference case). 'Wi' and 'b' represent weight and bias of linear layer on each device (i).", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/389510233_SPD_Sync-Point_Drop_for_efficient_tensor_parallelism_of_Large_Language_Models", "content": "Decoder block structure with sync-point drop (in 2-GPUs distributed inference case). 'Wi' and 'b' represent weight and bias of linear layer on each device (i)."}
+{"idx": 2, "title": "Decoder block structure with sync-point drop (in 2-GPUs distributed ...", "date": "", "ddg_snippet": "Decoder block structure with sync-point drop (in 2-GPUs distributed inference case). 'Wi' and 'b' represent weight and bias of linear layer on each device (i).", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/figure/Decoder-block-structure-with-sync-point-drop-in-2-GPUs-distributed-inference-case-Wi_fig2_389510233", "content": "Decoder block structure with sync-point drop (in 2-GPUs distributed inference case). 'Wi' and 'b' represent weight and bias of linear layer on each device (i)."}
+{"idx": 3, "title": "SPD: Sync-Point Drop for efficient tensor parallelism of Large Language ...", "date": "", "ddg_snippet": "In this paper, we present Sync-Point Drop ( SPD ), a novel optimization technique that improves the latency of LLMs on distributed inference systems. By adopting a new block design and separated approaches based on block-wisely identified sensitivity for lack of sync-point , SPD enables efficient deployment across multiple computing units with ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2502.20727v2", "content": "In this paper, we present Sync-Point Drop ( SPD ), a novel optimization technique that improves the latency of LLMs on distributed inference systems. By adopting a new block design and separated approaches based on block-wisely identified sensitivity for lack of sync-point , SPD enables efficient deployment across multiple computing units with ..."}
+{"idx": 4, "title": "Spd: Sync-point Drop for Efficient Tensor Par Allelism of Large ...", "date": "", "ddg_snippet": "ability and low latency. Therefore, we introduce a novel optimization technique, Sync-Point Drop ( SPD ) to reduce communication overheads in tensor parallelism by dropping synchronization on attention outputs. In detail, we first propose a block design that allows execution to proceed without communication through SPD . Second, we identify regions of communication redundancy, where dropping ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=uoU4ypjAmN", "content": "ability and low latency. Therefore, we introduce a novel optimization technique, Sync-Point Drop ( SPD ) to reduce communication overheads in tensor parallelism by dropping synchronization on attention outputs. In detail, we first propose a block design that allows execution to proceed without communication through SPD . Second, we identify regions of communication redundancy, where dropping ..."}
+{"idx": 5, "title": "PDF spColumn-v10.20-User-Manual - StructurePoint", "date": "", "ddg_snippet": "Looking in plan at the section with z-axis pointing outwards, the positive x-axis points to the right and the positive y-axis points up. For this section , vectors of positive bending moments have the same orientation as their corresponding axes x and y.", "subpage_snippet": "", "source": "structurepoint.org", "link": "https://structurepoint.org/pdfs/manuals/spColumn-Manual.pdf", "content": "Looking in plan at the section with z-axis pointing outwards, the positive x-axis points to the right and the positive y-axis points up. For this section , vectors of positive bending moments have the same orientation as their corresponding axes x and y."}
+{"idx": 6, "title": "[2502.20727] SPD: Sync-Point Drop for Efficient Tensor Parallelism of ...", "date": "", "ddg_snippet": "Therefore, we introduce a novel optimization technique, Sync-Point Drop ( SPD ), to reduce communication overheads in tensor parallelism by selectively dropping synchronization on attention outputs. In detail, we first propose a block design that allows execution to proceed without communication through SPD .", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2502.20727", "content": "Therefore, we introduce a novel optimization technique, Sync-Point Drop ( SPD ), to reduce communication overheads in tensor parallelism by selectively dropping synchronization on attention outputs. In detail, we first propose a block design that allows execution to proceed without communication through SPD ."}
+{"idx": 7, "title": "Optimizing LLM Inference Across Multiple GPUs - zerna.io", "date": "", "ddg_snippet": "Sync-Point Drop ( SPD ) is a novel optimization technique that selectively eliminates synchronization points in tensor-parallel LLM inference, significantly reducing communication overhead.", "subpage_snippet": "", "source": "zerna.io", "link": "http://zerna.io/en/page/engineering/presentation_set/engineering-llm-research/presentation/engineering-model-optimization/slide/engineering-paper-2502_20727", "content": "Sync-Point Drop ( SPD ) is a novel optimization technique that selectively eliminates synchronization points in tensor-parallel LLM inference, significantly reducing communication overhead."}
+{"idx": 8, "title": "Parallel Computing - The Art of HPC", "date": "", "ddg_snippet": "It is easy to characterize the gain in memory, as the total memory is the sum of the individual memories. The speed of a parallel computer is harder to characterize. This section will have an extended discussion on theoretical measures for expressing and judging the gain in execution speed from going to a parallel architecture.", "subpage_snippet": "", "source": "theartofhpc.com", "link": "https://theartofhpc.com/istc/parallel.html", "content": "It is easy to characterize the gain in memory, as the total memory is the sum of the individual memories. The speed of a parallel computer is harder to characterize. This section will have an extended discussion on theoretical measures for expressing and judging the gain in execution speed from going to a parallel architecture."}
+{"idx": 9, "title": "PDF A Primer on Coordinate Descent Algorithms - UCLA Mathematics", "date": "", "ddg_snippet": "Abstract This monograph presents a class of algorithms called coordinate descent algorithms for mathemati-cians, statisticians, and engineers outside the eld of optimization. This particular class of algo-rithms has recently gained popularity due to their e ectiveness in solving large-scale optimization problems in machine learning, compressed sensing, image processing, and computational ...", "subpage_snippet": "", "source": "ww3.math.ucla.edu", "link": "https://ww3.math.ucla.edu/camreport/cam16-67.pdf", "content": "Abstract This monograph presents a class of algorithms called coordinate descent algorithms for mathemati-cians, statisticians, and engineers outside the eld of optimization. This particular class of algo-rithms has recently gained popularity due to their e ectiveness in solving large-scale optimization problems in machine learning, compressed sensing, image processing, and computational ..."}

data/sampled_jsons/23zxLtvder_SPD_Sync-Point_Drop_Tensor_Parallelism_Data_Parallelism_Section_C.2.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "SPD : Sync - Point Drop for Efficient Tensor Parallelism of Large...", "date": "", "ddg_snippet": "Tensor parallelism provides an effective way to increase server large language model (LLM) inference efficiency despite adding an additional communication cost.", "subpage_snippet": "", "source": "machinelearning.apple.com", "link": "https://machinelearning.apple.com/research/sync-point-drop", "content": "Tensor parallelism provides an effective way to increase server large language model (LLM) inference efficiency despite adding an additional communication cost."}
+{"idx": 1, "title": "[2502.20727v4] SPD : Sync - Point Drop for Efficient Tensor ...", "date": "", "ddg_snippet": "View a PDF of the paper titled SPD : Sync - Point Drop for Efficient Tensor Parallelism of Large Language Models, by Han-Byul Kim and 4 other authors.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2502.20727v4", "content": "View a PDF of the paper titled SPD : Sync - Point Drop for Efficient Tensor Parallelism of Large Language Models, by Han-Byul Kim and 4 other authors."}
+{"idx": 2, "title": "ICML Poster SPD : Sync - Point Drop for efficient tensor parallelism of...", "date": "", "ddg_snippet": "Therefore, we introduce a novel optimization technique, Sync - Point Drop ( SPD ), to reduce communication overheads in tensor parallelism by selectively dropping synchronization on attention outputs. In detail, we first propose a block design that allows execution to proceed without...", "subpage_snippet": "", "source": "icml.cc", "link": "https://icml.cc/virtual/2025/poster/46606", "content": "Therefore, we introduce a novel optimization technique, Sync - Point Drop ( SPD ), to reduce communication overheads in tensor parallelism by selectively dropping synchronization on attention outputs. In detail, we first propose a block design that allows execution to proceed without..."}
+{"idx": 3, "title": "SPD : Sync - Point Drop for efficient tensor parallelism of Large...", "date": "", "ddg_snippet": "Overview SPD ( Sync - Point Drop ) reduces communication overhead in tensor parallelism for LLMsIdentifies and selectively drops unnecessary synchronization operations", "subpage_snippet": "", "source": "www.aimodels.fyi", "link": "https://www.aimodels.fyi/papers/arxiv/spd-sync-point-drop-efficient-tensor-parallelism", "content": "Overview SPD ( Sync - Point Drop ) reduces communication overhead in tensor parallelism for LLMsIdentifies and selectively drops unnecessary synchronization operations"}
+{"idx": 4, "title": "Figure 1. Tensor parallelism applied on transformer decoder block (in...", "date": "", "ddg_snippet": "SPD : Sync - Point Drop for efficient tensor parallelism of Large Language Models.However, communication overheads from popular distributed inference techniques such as Tensor Parallelism pose a significant challenge to achieve scalability and low latenc...", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/figure/Tensor-parallelism-applied-on-transformer-decoder-block-in-2-GPUs-distributed-inference_fig1_389510233", "content": "SPD : Sync - Point Drop for efficient tensor parallelism of Large Language Models.However, communication overheads from popular distributed inference techniques such as Tensor Parallelism pose a significant challenge to achieve scalability and low latenc..."}
+{"idx": 5, "title": "Data - parallel types (SIMD) (since C ++26) - cppreference.com", "date": "", "ddg_snippet": "The data - parallel type refers to all enabled specializations of the class templates basic_simd and basic_simd_mask.This simple rule expresses data - parallelism and will be used by the compiler to generate SIMD instructions and/or independent execution streams.", "subpage_snippet": "", "source": "w.cppreference.com", "link": "https://w.cppreference.com/cpp/numeric/simd.html", "content": "The data - parallel type refers to all enabled specializations of the class templates basic_simd and basic_simd_mask.This simple rule expresses data - parallelism and will be used by the compiler to generate SIMD instructions and/or independent execution streams."}
+{"idx": 6, "title": "Added second 5090 and turne on tensor parallel ... - vLLM Forums", "date": "", "ddg_snippet": "After running docker with tensor - parallel -size 2 I got these errors with dual 5090: (VllmWorker TP0 pid=152) INFO 09-18 08:07:19 [custom_all_reduce.py:35] Skipping P 2 P check and trusting the driver’s P 2 P report.", "subpage_snippet": "", "source": "discuss.vllm.ai", "link": "https://discuss.vllm.ai/t/added-second-5090-and-turne-on-tensor-parallel-2/1629", "content": "After running docker with tensor - parallel -size 2 I got these errors with dual 5090: (VllmWorker TP0 pid=152) INFO 09-18 08:07:19 [custom_all_reduce.py:35] Skipping P 2 P check and trusting the driver’s P 2 P report."}
+{"idx": 7, "title": "Reducing LLM Inference Costs While Preserving Performance", "date": "", "ddg_snippet": "Recent research on efficient tensor parallelism (e.g. Sync - Point Drop ) shows it’s possible to alleviate communication bottlenecks and scale LLM inference nearly linearly with minimal accuracy impact ( SPD : Sync - Point Drop for efficient tensor parallelism of Large...).", "subpage_snippet": "", "source": "www.rohan-paul.com", "link": "https://www.rohan-paul.com/p/reducing-llm-inference-costs-while", "content": "Recent research on efficient tensor parallelism (e.g. Sync - Point Drop ) shows it’s possible to alleviate communication bottlenecks and scale LLM inference nearly linearly with minimal accuracy impact ( SPD : Sync - Point Drop for efficient tensor parallelism of Large...)."}
+{"idx": 8, "title": "Распределённый инференс и шардирование LLM. Часть... / Хабр", "date": "", "ddg_snippet": "Tensor parallel size — сколько «кусочков» весов распределять по GPU внутри узла. Tensor Parallelism дробит модель на куски, загружает их на все локальные GPU и с помощью NCCL синхронизирует параметры. Между узлами.", "subpage_snippet": "", "source": "habr.com", "link": "https://habr.com/ru/companies/flant/articles/906702/", "content": "Tensor parallel size — сколько «кусочков» весов распределять по GPU внутри узла. Tensor Parallelism дробит модель на куски, загружает их на все локальные GPU и с помощью NCCL синхронизирует параметры. Между узлами."}
+{"idx": 9, "title": "Ускорение инференса LLM через тензорный... — AI на vc.ru", "date": "", "ddg_snippet": "Data Parallel или реплики моделей. Data Parallel (DP) - еще один метод параллизма, но на уровне данных.Настройки в vLLM. Выберите tensor _ parallel _size, кратный hidden_size и num_heads модели. Llama- 2 -7B: 4096 hidden / 32 heads ⇒ TP = 2 , 4, 8 допустимы.", "subpage_snippet": "", "source": "vc.ru", "link": "https://vc.ru/ai/2052555-uskorenie-inferensa-llm-s-pomoshchyu-tenzornogo-parallelizma", "content": "Data Parallel или реплики моделей. Data Parallel (DP) - еще один метод параллизма, но на уровне данных.Настройки в vLLM. Выберите tensor _ parallel _size, кратный hidden_size и num_heads модели. Llama- 2 -7B: 4096 hidden / 32 heads ⇒ TP = 2 , 4, 8 допустимы."}

data/sampled_jsons/2406.13909_appendix_table_1_Two-Room_(2x11)_training_steps_Rnd._Experts.jsonl

@@ -0,0 +1,7 @@
+{"idx": 0, "title": "Multiplication chart 1-100 and 1-12 on Timestables.com", "date": "", "ddg_snippet": "Multiplication chart On this page there is a multiplication chart 1 -100 and 1 -12. This web page is aimed at primary school children. You can practice on your own or with your parents. Learn the tables and say them out loud so you can remember them better. What may also be interesting is the tables table . This is a handy way of quickly viewing of all the tables . Learning the times tables is a ...", "subpage_snippet": "", "source": "www.timestables.com", "link": "https://www.timestables.com/times-table-chart/", "content": "Multiplication chart On this page there is a multiplication chart 1 -100 and 1 -12. This web page is aimed at primary school children. You can practice on your own or with your parents. Learn the tables and say them out loud so you can remember them better. What may also be interesting is the tables table . This is a handy way of quickly viewing of all the tables . Learning the times tables is a ..."}
+{"idx": 1, "title": "Appendices Setup - APA Style", "date": "", "ddg_snippet": "Appendices can consist of text along with tables and/or figures. In this case, write the appendix label and title on separate lines in bold and centered, as described in the section about basic appendix format, and present tables and figures as you would in the text (see the Table Setup and Figure Setup pages for more). Ensure you call out the appendix tables or figures within the text of the ...", "subpage_snippet": "", "source": "apastyle.apa.org", "link": "https://apastyle.apa.org/style-grammar-guidelines/paper-format/appendices", "content": "Appendices can consist of text along with tables and/or figures. In this case, write the appendix label and title on separate lines in bold and centered, as described in the section about basic appendix format, and present tables and figures as you would in the text (see the Table Setup and Figure Setup pages for more). Ensure you call out the appendix tables or figures within the text of the ..."}
+{"idx": 2, "title": "COMPANION TO THE AISC STEEL CONSTRUCTION MANUAL", "date": "", "ddg_snippet": "The objective of this Companion is to provide additional design tables beyond what are incorporated into the 15th Edition AISC Steel Construction Manual. Tables in this Companion that present available strengths are developed using the geometric conditions indicated and applicable limits states from the 2016 AISC Specification for Structural Steel Buildings(ANSI/AISC 360-16). Given the nature ...", "subpage_snippet": "", "source": "www.aisc.org", "link": "https://www.aisc.org/globalassets/aisc/manual/v15.1-companion/v15.1_vol-2_design-tables.pdf", "content": "The objective of this Companion is to provide additional design tables beyond what are incorporated into the 15th Edition AISC Steel Construction Manual. Tables in this Companion that present available strengths are developed using the geometric conditions indicated and applicable limits states from the 2016 AISC Specification for Structural Steel Buildings(ANSI/AISC 360-16). Given the nature ..."}
+{"idx": 3, "title": "Mathway | Algebra Problem Solver", "date": "", "ddg_snippet": "Free math problem solver answers your algebra homework questions with step-by-step explanations.", "subpage_snippet": "", "source": "www.mathway.com", "link": "https://www.mathway.com/algebra", "content": "Free math problem solver answers your algebra homework questions with step-by-step explanations."}
+{"idx": 4, "title": "Joint Travel Regulations | Defense Travel Management Office", "date": "", "ddg_snippet": "The Joint Travel Regulations (JTR) implements policy and law to establish travel and transportation allowances for Uniformed Service members, DoD civilian employees, and others traveling at the DoD’s expense.", "subpage_snippet": "", "source": "www.travel.dod.mil", "link": "https://www.travel.dod.mil/Policy-Regulations/Joint-Travel-Regulations/", "content": "The Joint Travel Regulations (JTR) implements policy and law to establish travel and transportation allowances for Uniformed Service members, DoD civilian employees, and others traveling at the DoD’s expense."}
+{"idx": 5, "title": "Desmos | Graphing Calculator", "date": "", "ddg_snippet": "Explore math with our beautiful, free online graphing calculator. Graph functions, plot points, visualize algebraic equations, add sliders, animate graphs, and more.", "subpage_snippet": "", "source": "www.desmos.com", "link": "https://www.desmos.com/calculator", "content": "Explore math with our beautiful, free online graphing calculator. Graph functions, plot points, visualize algebraic equations, add sliders, animate graphs, and more."}
+{"idx": 6, "title": "Equation Solver - Mathway", "date": "", "ddg_snippet": "The equation solver allows you to enter your problem and solve the equation to see the result. Solve in one variable or many.", "subpage_snippet": "", "source": "www.mathway.com", "link": "https://www.mathway.com/Calculator/equation-solver", "content": "The equation solver allows you to enter your problem and solve the equation to see the result. Solve in one variable or many."}

data/sampled_jsons/2406.13909_neural_network_S-functions_continuous_MDPs_input_output.jsonl

@@ -0,0 +1,3 @@
+{"idx": 0, "title": "Beyond Optimism: Exploration With Partially Observable Rewards", "date": "", "ddg_snippet": "First, we considered tabular MDPs, thus we plan to follow up on continuous MDPs .. For discrete actions, the network would take the (current state, goal state) pair and output the value for all (action, goal action) pairs, similarly to how deep Q-networks [48] work.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2406.13909v2", "content": "First, we considered tabular MDPs, thus we plan to follow up on continuous MDPs .. For discrete actions, the network would take the (current state, goal state) pair and output the value for all (action, goal action) pairs, similarly to how deep Q-networks [48] work."}
+{"idx": 1, "title": "Exploration With Partially Observable Rewards", "date": "", "ddg_snippet": "by S Parisi · 2024 · Cited by 4 — For example, the finite set of S-functions ̂Ssiaj (st,at) could be replaced by a neural network ̂S(st,at,si,aj). For discrete actions, the network ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2406.13909?", "content": "by S Parisi · 2024 · Cited by 4 — For example, the finite set of S-functions ̂Ssiaj (st,at) could be replaced by a neural network ̂S(st,at,si,aj). For discrete actions, the network ..."}
+{"idx": 2, "title": "", "date": "", "ddg_snippet": "", "subpage_snippet": "", "source": "", "link": "", "content": ""}

data/sampled_jsons/2406.14532_base_models_Llama-2_Mistral_7B_experiments_Section_5.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Enhancing Decision-Making of Large Language Models ...", "date": "", "ddg_snippet": "by H Dong · Cited by 1 — Our method outperforms other state-of-the-art baselines using the same 7B /8B open-source LLMs and even exceeds a strong baseline ReAct using GPT-4 in most ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/forum?id=0tXmtd0vZG", "content": "by H Dong · Cited by 1 — Our method outperforms other state-of-the-art baselines using the same 7B /8B open-source LLMs and even exceeds a strong baseline ReAct using GPT-4 in most ..."}
+{"idx": 1, "title": "mistral:7b - Ollama", "date": "", "ddg_snippet": "Mistral is a 7B parameter model , distributed with the Apache license. It is available in both instruct (instruction following) and text completion. The Mistral AI team has noted that Mistral 7B : Outperforms Llama 2 13B on all benchmarks Outperforms Llama 1 34B on many benchmarks Approaches CodeLlama 7B performance on code, while remaining good at English tasks Versions", "subpage_snippet": "", "source": "ollama.com", "link": "https://ollama.com/library/mistral:7b", "content": "Mistral is a 7B parameter model , distributed with the Apache license. It is available in both instruct (instruction following) and text completion. The Mistral AI team has noted that Mistral 7B : Outperforms Llama 2 13B on all benchmarks Outperforms Llama 1 34B on many benchmarks Approaches CodeLlama 7B performance on code, while remaining good at English tasks Versions"}
+{"idx": 2, "title": "[2310.06825] Mistral 7B - arXiv.org", "date": "", "ddg_snippet": "We introduce Mistral 7B v0.1, a 7-billion-parameter language model engineered for superior performance and efficiency. Mistral 7B outperforms Llama 2 13B across all evaluated benchmarks, and Llama 1 34B in reasoning, mathematics, and code generation. Our model leverages grouped-query attention (GQA) for faster inference, coupled with sliding window attention (SWA) to effectively handle ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2310.06825", "content": "We introduce Mistral 7B v0.1, a 7-billion-parameter language model engineered for superior performance and efficiency. Mistral 7B outperforms Llama 2 13B across all evaluated benchmarks, and Llama 1 34B in reasoning, mathematics, and code generation. Our model leverages grouped-query attention (GQA) for faster inference, coupled with sliding window attention (SWA) to effectively handle ..."}
+{"idx": 3, "title": "The Battle of 7B in LLMs: Mistral 7B vs Llama-2 7B - Data Science Dojo", "date": "", "ddg_snippet": "Each option has a particular model size, providing a wide range of choices to users. However, in this blog we will explore two LLMs of 7B - Mistral 7B and Llama-2 7B , navigating the differences and similarities between the two options. Before we dig deeper into the showdown of the two 7B LLMs, let's do a quick recap of the language models .", "subpage_snippet": "", "source": "datasciencedojo.com", "link": "https://datasciencedojo.com/blog/mistral-7b-vs-llama-2-7b/", "content": "Each option has a particular model size, providing a wide range of choices to users. However, in this blog we will explore two LLMs of 7B - Mistral 7B and Llama-2 7B , navigating the differences and similarities between the two options. Before we dig deeper into the showdown of the two 7B LLMs, let's do a quick recap of the language models ."}
+{"idx": 4, "title": "Mistral 7B | Mistral AI", "date": "", "ddg_snippet": "Results on MMLU, Commonsense Reasoning, World Knowledge and Reading comprehension for Mistral 7B and Llama 2 ( 7B /13/70B). Mistral 7B largely outperforms Llama 2 13B on all evaluations, except on knowledge benchmarks, where it is on par (this is likely due to its limited parameter count, which restricts the amount of knowledge it can compress).", "subpage_snippet": "", "source": "mistral.ai", "link": "https://mistral.ai/en/news/announcing-mistral-7b", "content": "Results on MMLU, Commonsense Reasoning, World Knowledge and Reading comprehension for Mistral 7B and Llama 2 ( 7B /13/70B). Mistral 7B largely outperforms Llama 2 13B on all evaluations, except on knowledge benchmarks, where it is on par (this is likely due to its limited parameter count, which restricts the amount of knowledge it can compress)."}
+{"idx": 5, "title": "Comprehensive Examination of Instruction-Based Language Models: A ...", "date": "", "ddg_snippet": "In natural language processing and computer vision, both of which are developing rapidly, analytical methods are very important to improve modeling This study uses ROUGE, BLEU, and CIDEr as key performance indicators to assess how Mistral-7b and Llama-2-7b are better model perform. The preliminary results show that the LAMA-2-7b is the best in all dimensions. Despite these results, the new ...", "subpage_snippet": "", "source": "ieeexplore.ieee.org", "link": "https://ieeexplore.ieee.org/document/10434081", "content": "In natural language processing and computer vision, both of which are developing rapidly, analytical methods are very important to improve modeling This study uses ROUGE, BLEU, and CIDEr as key performance indicators to assess how Mistral-7b and Llama-2-7b are better model perform. The preliminary results show that the LAMA-2-7b is the best in all dimensions. Despite these results, the new ..."}
+{"idx": 6, "title": "Llama 2 vs Mistral 7B: Comparison of Two Leading LLM - HyScaler", "date": "", "ddg_snippet": "Large Language Models (LLMs) are revolutionizing the way we interact with computers. Here is a comparison between Llama 2 vs Mistral 7B .", "subpage_snippet": "", "source": "hyscaler.com", "link": "https://hyscaler.com/insights/llama-2-vs-mistral-7b-large-language-models/", "content": "Large Language Models (LLMs) are revolutionizing the way we interact with computers. Here is a comparison between Llama 2 vs Mistral 7B ."}
+{"idx": 7, "title": "Mistral 7B vs. LLama2: The 5 Key Differences Between the ... - Skim AI", "date": "", "ddg_snippet": "Dive into the competitive landscape of AI language models with a detailed comparison between Mistral 7B and LLama 2 , exploring their performance, adaptability, efficiency, and more to understand their unique impacts in the AI realm.", "subpage_snippet": "", "source": "skimai.com", "link": "https://skimai.com/mistral-7b-vs-llama2-the-5-key-differences-between-the-leading-open-source-llms/", "content": "Dive into the competitive landscape of AI language models with a detailed comparison between Mistral 7B and LLama 2 , exploring their performance, adaptability, efficiency, and more to understand their unique impacts in the AI realm."}
+{"idx": 8, "title": "Mistral 7B vs. Llama 3 70B vs. Gemma 2 9B: A Comprehensive ... - Medium", "date": "", "ddg_snippet": "Mistral 7B and Gemma 2 9B are commendable models , offering valuable capabilities and performance in their respective niches. However, for those seeking the most robust and versatile model , Llama 3 ...", "subpage_snippet": "", "source": "medium.com", "link": "https://medium.com/@samir20/mistral-7b-vs-llama-3-70b-vs-gemma-2-9b-a-comprehensive-benchmark-showdown-9c3128f24b23", "content": "Mistral 7B and Gemma 2 9B are commendable models , offering valuable capabilities and performance in their respective niches. However, for those seeking the most robust and versatile model , Llama 3 ..."}
+{"idx": 9, "title": "Comprehensive Benchmarking of Top LLMs: Qwen2, Llama, Mistral, Gemma ...", "date": "", "ddg_snippet": "Explore our in-depth analysis and benchmarking of the latest large language models , including Qwen2- 7B , Llama-3.1-8B, Mistral-7B , Gemma-2-9B, and Phi-3-medium-128k. Discover which models and libraries deliver the best performance in terms of tokens/sec and TTFT, helping you optimize your AI applications for maximum efficiency", "subpage_snippet": "", "source": "www.inferless.com", "link": "https://www.inferless.com/learn/exploring-llms-speed-benchmarks-independent-analysis---part-3", "content": "Explore our in-depth analysis and benchmarking of the latest large language models , including Qwen2- 7B , Llama-3.1-8B, Mistral-7B , Gemma-2-9B, and Phi-3-medium-128k. Discover which models and libraries deliver the best performance in terms of tokens/sec and TTFT, helping you optimize your AI applications for maximum efficiency"}

data/sampled_jsons/2407.01511_CRAB_abstract.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "[2407.01511] CRAB: Cross-environment Agent Benchmark for ... CRAB: Cross-environment Agent Benchmark for Multimodal ... GitHub - civicW/crabAgent: ️ CRAB: Cross-environment Agent ... CRAB: Cross-Environment Agent Benchmark crab-framework · PyPI CRAB: Cross-environment Agent Benchmark for Multimodal ... 同时操控手机和电脑，100项任务，跨系统智能体评测基准有了", "date": "", "ddg_snippet": "Jul 1, 2024 · Abstract page for arXiv paper 2407.01511 : CRAB : Cross-environment Agent Benchmark for Multimodal Language Model Agents To overcome these limitations, we introduce Crab , the first agent benchmark framework designed to support cross-environment tasks, incorporating a graph-based fine-grained evaluation method and an efficient mechanism for task and evaluator construction. Experiment on CRAB -Benchmark-v0 All datasets and experiment code are in crab -benchmark-v0 directory. Please carefully read the benchmark tutorial before using our benchmark. Cross-environment Tasks: CRAB is the first framework to facilitate agent benchmarking across multiple environments, allowing agents to interact with desktop computers and mobile devices through a standardized interface. This approach mirrors complex real-world applications more closely than traditional single-environment tasks. Jul 31, 2024 · 🦀 CRAB : Cross-platform Agent Benchmark for Multimodal Embodied Language Model Agents Overview CRAB is a framework for building LLM agent benchmark environments in a Python-centric way. Key Features 🌐 Cross-platform and Multi-environment Create build agent environments that support various deployment options including in-memory, Docker-hosted, virtual machines, or distributed physical ... CRAB aims to become a general-purpose agent benchmark framework for Multimodal Language Model (MLM) agents. CRAB provides an end-to-end while easy-to-use framework to build agents, operate environments, and create benchmarks to evaluate them, featuring three key components: cross-environment support, a graph evaluator, and task generation. Aug 15, 2024 · 基于 CRAB ，本文提出了 Crab Benchmark-v0，同时支持智能体在 Ubuntu 和 Android 系统上执行多种复杂的跨环境任务，这一基准的提出，不仅可以推动自主智能体评价体系的发展，也为未来设计更加高效的智能体系统提供全新灵感。", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2407.01511", "content": "Jul 1, 2024 · Abstract page for arXiv paper 2407.01511 : CRAB : Cross-environment Agent Benchmark for Multimodal Language Model Agents To overcome these limitations, we introduce Crab , the first agent benchmark framework designed to support cross-environment tasks, incorporating a graph-based fine-grained evaluation method and an efficient mechanism for task and evaluator construction. Experiment on CRAB -Benchmark-v0 All datasets and experiment code are in crab -benchmark-v0 directory. Please carefully read the benchmark tutorial before using our benchmark. Cross-environment Tasks: CRAB is the first framework to facilitate agent benchmarking across multiple environments, allowing agents to interact with desktop computers and mobile devices through a standardized interface. This approach mirrors complex real-world applications more closely than traditional single-environment tasks. Jul 31, 2024 · 🦀 CRAB : Cross-platform Agent Benchmark for Multimodal Embodied Language Model Agents Overview CRAB is a framework for building LLM agent benchmark environments in a Python-centric way. Key Features 🌐 Cross-platform and Multi-environment Create build agent environments that support various deployment options including in-memory, Docker-hosted, virtual machines, or distributed physical ... CRAB aims to become a general-purpose agent benchmark framework for Multimodal Language Model (MLM) agents. CRAB provides an end-to-end while easy-to-use framework to build agents, operate environments, and create benchmarks to evaluate them, featuring three key components: cross-environment support, a graph evaluator, and task generation. Aug 15, 2024 · 基于 CRAB ，本文提出了 Crab Benchmark-v0，同时支持智能体在 Ubuntu 和 Android 系统上执行多种复杂的跨环境任务，这一基准的提出，不仅可以推动自主智能体评价体系的发展，也为未来设计更加高效的智能体系统提供全新灵感。"}
+{"idx": 1, "title": "CRAB: Cross-environment Agent Benchmark for ...", "date": "", "ddg_snippet": "9 Aug 2024 — A cross-platform/environment multimodal agent benchmark framework: CRAB , innovatively enabling agents to operate multiple devices simultaneously.", "subpage_snippet": "", "source": "www.camel-ai.org", "link": "https://www.camel-ai.org/blogs/crab-cross-platform-agent-benchmark", "content": "9 Aug 2024 — A cross-platform/environment multimodal agent benchmark framework: CRAB , innovatively enabling agents to operate multiple devices simultaneously."}
+{"idx": 2, "title": "CRAB: Cross-environment Agent Benchmark for Multimodal ...", "date": "", "ddg_snippet": "To overcome these limitations, we introduce Crab , the first agent benchmark framework designed to support cross-environment tasks, incorporating a graph-based fine-grained evaluation method and an efficient mechanism for task and evaluator construction.", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/papers/2407.01511", "content": "To overcome these limitations, we introduce Crab , the first agent benchmark framework designed to support cross-environment tasks, incorporating a graph-based fine-grained evaluation method and an efficient mechanism for task and evaluator construction."}
+{"idx": 3, "title": "GitHub - civicW/crabAgent: ️ CRAB: Cross-environment Agent ...", "date": "", "ddg_snippet": "Experiment on CRAB -Benchmark-v0 All datasets and experiment code are in crab -benchmark-v0 directory. Please carefully read the benchmark tutorial before using our benchmark.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/civicW/crabAgent", "content": "Experiment on CRAB -Benchmark-v0 All datasets and experiment code are in crab -benchmark-v0 directory. Please carefully read the benchmark tutorial before using our benchmark."}
+{"idx": 4, "title": "CRAB: Cross-Environment Agent Benchmark", "date": "", "ddg_snippet": "Cross-environment Tasks: CRAB is the first framework to facilitate agent benchmarking across multiple environments, allowing agents to interact with desktop computers and mobile devices through a standardized interface. This approach mirrors complex real-world applications more closely than traditional single-environment tasks.", "subpage_snippet": "", "source": "www.emergentmind.com", "link": "https://www.emergentmind.com/papers/2407.01511", "content": "Cross-environment Tasks: CRAB is the first framework to facilitate agent benchmarking across multiple environments, allowing agents to interact with desktop computers and mobile devices through a standardized interface. This approach mirrors complex real-world applications more closely than traditional single-environment tasks."}
+{"idx": 5, "title": "crab-framework · PyPI", "date": "", "ddg_snippet": "Jul 31, 2024 · 🦀 CRAB : Cross-platform Agent Benchmark for Multimodal Embodied Language Model Agents Overview CRAB is a framework for building LLM agent benchmark environments in a Python-centric way. Key Features 🌐 Cross-platform and Multi-environment Create build agent environments that support various deployment options including in-memory, Docker-hosted, virtual machines, or distributed physical ...", "subpage_snippet": "", "source": "pypi.org", "link": "https://pypi.org/project/crab-framework/", "content": "Jul 31, 2024 · 🦀 CRAB : Cross-platform Agent Benchmark for Multimodal Embodied Language Model Agents Overview CRAB is a framework for building LLM agent benchmark environments in a Python-centric way. Key Features 🌐 Cross-platform and Multi-environment Create build agent environments that support various deployment options including in-memory, Docker-hosted, virtual machines, or distributed physical ..."}
+{"idx": 6, "title": "CRAB: Cross-environment Agent Benchmark for Multimodal ...", "date": "", "ddg_snippet": "CRAB aims to become a general-purpose agent benchmark framework for Multimodal Language Model (MLM) agents. CRAB provides an end-to-end while easy-to-use framework to build agents, operate environments, and create benchmarks to evaluate them, featuring three key components: cross-environment support, a graph evaluator, and task generation.", "subpage_snippet": "", "source": "crab.camel-ai.org", "link": "https://crab.camel-ai.org/", "content": "CRAB aims to become a general-purpose agent benchmark framework for Multimodal Language Model (MLM) agents. CRAB provides an end-to-end while easy-to-use framework to build agents, operate environments, and create benchmarks to evaluate them, featuring three key components: cross-environment support, a graph evaluator, and task generation."}
+{"idx": 7, "title": "Towards Evaluating Generalist Agents: An Automated ...", "date": "", "ddg_snippet": "The CRAB framework Xu et al. (2024) introduces a cross-environment benchmark that leverages multimodal language models to perform tasks across various GUI ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2310.08367v2", "content": "The CRAB framework Xu et al. (2024) introduces a cross-environment benchmark that leverages multimodal language models to perform tasks across various GUI ..."}
+{"idx": 8, "title": "Cybernaut: Towards Reliable Web Automation", "date": "", "ddg_snippet": "Crab: cross-environment agent benchmark for multimodal language model agents . arXiv Preprint, (2024). https://arxiv.org/abs/2407.01511. arXiv: 2407.01511 [cs.AI] ...", "subpage_snippet": "", "source": "assets.amazon.science", "link": "https://assets.amazon.science/13/50/30cd0cfc4fe381691ab383e0fb85/scipub-approval152129-38644753-cybernaut-towards-reliable-web-automation.pdf", "content": "Crab: cross-environment agent benchmark for multimodal language model agents . arXiv Preprint, (2024). https://arxiv.org/abs/2407.01511. arXiv: 2407.01511 [cs.AI] ..."}
+{"idx": 9, "title": "ENHANCING MULTI-AGENT LEARNING IN REAL", "date": "", "ddg_snippet": "CRAB: Cross-environment agent benchmark for multimodal language model agents . arXiv preprint arXiv:2407.01511, 2024. An Yan, Zhengyuan Yang, Wanrong Zhu ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf/b1851fd67b48f3112fd964abed7d39f7cb86d4eb.pdf", "content": "CRAB: Cross-environment agent benchmark for multimodal language model agents . arXiv preprint arXiv:2407.01511, 2024. An Yan, Zhengyuan Yang, Wanrong Zhu ..."}

data/sampled_jsons/2410.02025_MNIST_Wasserstein_distance_sparse_network_empirical_results_table.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Wasserstein Distances, Neuronal Entanglement, and Sparsity", "date": "", "ddg_snippet": "The Wasserstein distance between a neuron's normalized sparse output distribution and the Gaussian distribution is shown as sparsity increases for the top 3% of entangled Wasserstein neurons, the same number of bottom 3% WD neurons, and a random sample of 3% of the neurons.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2405.15756v4", "content": "The Wasserstein distance between a neuron's normalized sparse output distribution and the Gaussian distribution is shown as sparsity increases for the top 3% of entangled Wasserstein neurons, the same number of bottom 3% WD neurons, and a random sample of 3% of the neurons."}
+{"idx": 1, "title": "PDF Quantifying the Empirical Wasserstein Distance to a Set of ... - NeurIPS", "date": "", "ddg_snippet": "We consider the problem of estimating the Wasserstein distance between the empirical measure and a set of probability measures whose expectations over a class of functions (hypothesis class) are constrained.", "subpage_snippet": "", "source": "proceedings.neurips.cc", "link": "https://proceedings.neurips.cc/paper/2020/file/f3507289cfdc8c9ae93f4098111a13f9-Paper.pdf", "content": "We consider the problem of estimating the Wasserstein distance between the empirical measure and a set of probability measures whose expectations over a class of functions (hypothesis class) are constrained."}
+{"idx": 2, "title": "Empirical Wasserstein distance between MNIST and MNIST with varied ...", "date": "", "ddg_snippet": "Figure 4 compares the empirical Wasserstein distance using orthonomal projection compared to Gaussian random projection and sparse random projection when MNIST is perturbed with additive Gaussian ...", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/figure/Empirical-Wasserstein-distance-between-MNIST-and-MNIST-with-varied-levels-of-additive_fig3_368687851", "content": "Figure 4 compares the empirical Wasserstein distance using orthonomal projection compared to Gaussian random projection and sparse random projection when MNIST is perturbed with additive Gaussian ..."}
+{"idx": 3, "title": "A new robust partial p-Wasserstein-based metric for comparing ...", "date": "", "ddg_snippet": "The 2-Wasserstein distance is sensitive to minor geometric differences between distributions, making it a very powerful dissimilarity metric. However, due to this sensitivity, a small outlier mass can also cause a significant increase in the 2-Wasserstein distance between two similar distributions. Similarly, sampling discrepancy can cause the empirical 2-Wasserstein distance on n samples in ...", "subpage_snippet": "", "source": "dl.acm.org", "link": "https://dl.acm.org/doi/10.5555/3692070.3693773", "content": "The 2-Wasserstein distance is sensitive to minor geometric differences between distributions, making it a very powerful dissimilarity metric. However, due to this sensitivity, a small outlier mass can also cause a significant increase in the 2-Wasserstein distance between two similar distributions. Similarly, sampling discrepancy can cause the empirical 2-Wasserstein distance on n samples in ..."}
+{"idx": 4, "title": "PDF Smooth p-Wasserstein Distance: Structure, Empirical Approximation, and ...", "date": "", "ddg_snippet": "We adopt the smooth Wasserstein distance as the figure of merit and use the empirical distri-bution ^n as an estimate for . Generative modeling is thus formulated as the following minimum smooth Wasserstein estimation (M-SWE) problem:", "subpage_snippet": "", "source": "proceedings.mlr.press", "link": "https://proceedings.mlr.press/v139/nietert21a/nietert21a.pdf", "content": "We adopt the smooth Wasserstein distance as the figure of merit and use the empirical distri-bution ^n as an estimate for . Generative modeling is thus formulated as the following minimum smooth Wasserstein estimation (M-SWE) problem:"}
+{"idx": 5, "title": "Wasserstein Distances Made Explainable: Insights into Dataset Shifts ...", "date": "", "ddg_snippet": "This confirms the technical strength of our method, namely its sensitivity to the actual Wasserstein distance model and its ability to attribute the Wasserstein distance comprehensively through its fulfillment of the conservation property (cf. Table II for a technical comparison between methods).", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2505.06123", "content": "This confirms the technical strength of our method, namely its sensitivity to the actual Wasserstein distance model and its ability to attribute the Wasserstein distance comprehensively through its fulfillment of the conservation property (cf. Table II for a technical comparison between methods)."}
+{"idx": 6, "title": "Quantifying the Empirical Wasserstein Distance to a Set of ... - NeurIPS", "date": "", "ddg_snippet": "Authors Nian Si, Jose Blanchet, Soumyadip Ghosh, Mark Squillante Abstract We consider the problem of estimating the Wasserstein distance between the empirical measure and a set of probability measures whose expectations over a class of functions (hypothesis class) are constrained. If this class is sufficiently rich to characterize a particular distribution (e.g., all Lipschitz functions), then ...", "subpage_snippet": "", "source": "proceedings.neurips.cc", "link": "https://proceedings.neurips.cc/paper/2020/hash/f3507289cfdc8c9ae93f4098111a13f9-Abstract.html", "content": "Authors Nian Si, Jose Blanchet, Soumyadip Ghosh, Mark Squillante Abstract We consider the problem of estimating the Wasserstein distance between the empirical measure and a set of probability measures whose expectations over a class of functions (hypothesis class) are constrained. If this class is sufficiently rich to characterize a particular distribution (e.g., all Lipschitz functions), then ..."}
+{"idx": 7, "title": "Published as a conference paper at ICLR 2025 - OpenReview", "date": "", "ddg_snippet": "ers. Sparsity is set to 90% for each expert. Here, WD refers to the Wasserstein distance between the Sparse Expansion sparse and dense out ut distributions, rather than to a Gaussian. RI repres", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=cnKhHxN3xj", "content": "ers. Sparsity is set to 90% for each expert. Here, WD refers to the Wasserstein distance between the Sparse Expansion sparse and dense out ut distributions, rather than to a Gaussian. RI repres"}
+{"idx": 8, "title": "Feature selection via risk-bound utility maximization", "date": "", "ddg_snippet": "The 1-Wasserstein distance , originating from OT, has gained popularity in machine learning tasks due to its geometric interpretability and empirical stability. Unlike divergence-based metrics, it incorporates the geometry of the data and offers robustness to support mismatches.", "subpage_snippet": "", "source": "www.sciencedirect.com", "link": "https://www.sciencedirect.com/science/article/pii/S0925231225022441", "content": "The 1-Wasserstein distance , originating from OT, has gained popularity in machine learning tasks due to its geometric interpretability and empirical stability. Unlike divergence-based metrics, it incorporates the geometry of the data and offers robustness to support mismatches."}
+{"idx": 9, "title": "2310.01973v1 (1) | PDF | Computer Science | Learning - Scribd", "date": "", "ddg_snippet": "In Section 4, we conduct an empirical analysis of FedWaD on different use-cases ( Wasserstein coresets and Optimal Transport Dataset distance ) which rely on the computation of the Wasserstein distance . We unveil how these problems can be solved in our FL setting and demonstrates the remarkable versatility of our approach. In particular, we expose", "subpage_snippet": "", "source": "www.scribd.com", "link": "https://www.scribd.com/document/918012031/2310-01973v1-1", "content": "In Section 4, we conduct an empirical analysis of FedWaD on different use-cases ( Wasserstein coresets and Optimal Transport Dataset distance ) which rely on the computation of the Wasserstein distance . We unveil how these problems can be solved in our FL setting and demonstrates the remarkable versatility of our approach. In particular, we expose"}

data/sampled_jsons/2410.09536_random_segment_length_fixed_segment_Section_5.3.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Random segments and broken sticks - The DO Loop - SAS Blogs", "date": "", "ddg_snippet": "A classical problem in elementary probability asks for the expected lengths of line segments that result from randomly selecting k points along a segment of unit length .", "subpage_snippet": "", "source": "blogs.sas.com", "link": "https://blogs.sas.com/content/iml/2017/07/26/random-segments-broken-sticks.html", "content": "A classical problem in elementary probability asks for the expected lengths of line segments that result from randomly selecting k points along a segment of unit length ."}
+{"idx": 1, "title": "Length of a Line Segment Calculator", "date": "", "ddg_snippet": "\"A line segment is a section of a line that has two endpoints, A and B, and a fixed length . Being different from a line, which does not have a beginning or an end.", "subpage_snippet": "", "source": "www.omnicalculator.com", "link": "https://www.omnicalculator.com/math/length-of-a-line-segment", "content": "\"A line segment is a section of a line that has two endpoints, A and B, and a fixed length . Being different from a line, which does not have a beginning or an end."}
+{"idx": 2, "title": "Self-supervision through Random Segments with Autoregressive Coding", "date": "", "ddg_snippet": "To achieve this, we introduce a novel strategy called \" Random Segments with Autoregressive Coding (RandSAC)\" where we group patch representations (image tokens) into hierarchically arranged segments . Within each segment , tokens are predicted in parallel (similar to BERT), while across segment predictions are sequential (similar to GPT).", "subpage_snippet": "", "source": "open.library.ubc.ca", "link": "https://open.library.ubc.ca/media/download/pdf/24/1.0431370/3", "content": "To achieve this, we introduce a novel strategy called \" Random Segments with Autoregressive Coding (RandSAC)\" where we group patch representations (image tokens) into hierarchically arranged segments . Within each segment , tokens are predicted in parallel (similar to BERT), while across segment predictions are sequential (similar to GPT)."}
+{"idx": 3, "title": "Random Points on a Segment - Alexander Bogomolny", "date": "", "ddg_snippet": "Random Points on a Segment Problem 1 Two random points, chosen uniformly and independently on a unit segment , split the segment into 3 pieces. What is the average of the length of the left segment ?", "subpage_snippet": "", "source": "www.cut-the-knot.org", "link": "https://www.cut-the-knot.org/m/Probability/RandomPointsOnSegment.shtml", "content": "Random Points on a Segment Problem 1 Two random points, chosen uniformly and independently on a unit segment , split the segment into 3 pieces. What is the average of the length of the left segment ?"}
+{"idx": 4, "title": "PDF Conditional Random Fields: Probabilistic Models for Segmenting and ...", "date": "", "ddg_snippet": "Conditional random fields offer a unique combination of properties: discriminatively trained models for sequence segmentation and labeling; combination of arbitrary, over-lapping and agglomerative observation features from both the past and future; efficient training and decoding based on dynamic programming; and parameter estimation guar ...", "subpage_snippet": "", "source": "www.aladdin.cs.cmu.edu", "link": "http://www.aladdin.cs.cmu.edu/papers/pdfs/y2001/crf.pdf", "content": "Conditional random fields offer a unique combination of properties: discriminatively trained models for sequence segmentation and labeling; combination of arbitrary, over-lapping and agglomerative observation features from both the past and future; efficient training and decoding based on dynamic programming; and parameter estimation guar ..."}
+{"idx": 5, "title": "Segment Area Calculator", "date": "", "ddg_snippet": "With our segment area calculator you can find area of a segment given the central angle or segment height.", "subpage_snippet": "", "source": "www.omnicalculator.com", "link": "https://www.omnicalculator.com/math/segment-area", "content": "With our segment area calculator you can find area of a segment given the central angle or segment height."}
+{"idx": 6, "title": "Random points on interval, expected lengths of pieces", "date": "", "ddg_snippet": "The reader in hurry can skip some paragraphs and go to the compact solution int he sequel. The following long and possibly boring exposition gives the sincere solution, showing how it was found. One observation maybe at this point. The above numbers are building the triangle (1) divided by 1 (1, 3) divided by 4 (2, 5, 11) divided by 18 (3, 7, 13, 25) divided by 48 (12, 27, 47, 77, 137) divided ...", "subpage_snippet": "", "source": "math.stackexchange.com", "link": "https://math.stackexchange.com/questions/3245901/random-points-on-interval-expected-lengths-of-pieces", "content": "The reader in hurry can skip some paragraphs and go to the compact solution int he sequel. The following long and possibly boring exposition gives the sincere solution, showing how it was found. One observation maybe at this point. The above numbers are building the triangle (1) divided by 1 (1, 3) divided by 4 (2, 5, 11) divided by 18 (3, 7, 13, 25) divided by 48 (12, 27, 47, 77, 137) divided ..."}
+{"idx": 7, "title": "PDF 5.3 - Connecting Algebra & Geometry using Coordinates", "date": "", "ddg_snippet": "12. Find the point I that breaks the directed segment 1:4, given (−3,3) and point (4, −2). ⃗ in a ratio of", "subpage_snippet": "", "source": "mattsmathlabs.com", "link": "https://mattsmathlabs.com/grade10GSE/Unit05/05-03-MidpointsDirectedLineSegments-Complete.pdf", "content": "12. Find the point I that breaks the directed segment 1:4, given (−3,3) and point (4, −2). ⃗ in a ratio of"}
+{"idx": 8, "title": "cnata/3/wireshark_lab.md at master · xxyzz/cnata · GitHub", "date": "", "ddg_snippet": "What is the value of the Acknowledgement field in the SYNACK segment ? How did gaia.cs.umass.edu determine that value? What is it in the segment that identifies the segment as a SYNACK segment ? 0 1 set SYN and ACK flag What is the sequence number of the TCP segment containing the HTTP POST command?", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/xxyzz/cnata/blob/master/3/wireshark_lab.md", "content": "What is the value of the Acknowledgement field in the SYNACK segment ? How did gaia.cs.umass.edu determine that value? What is it in the segment that identifies the segment as a SYNACK segment ? 0 1 set SYN and ACK flag What is the sequence number of the TCP segment containing the HTTP POST command?"}
+{"idx": 9, "title": "PDF Wireshark_TCP.pdf - Temple University", "date": "", "ddg_snippet": "The whole transmission time is the difference of the time instant of the first TCP segment (i.e., 0.026477 second for No.4 segment ) and the time instant of the last ACK (i.e., 5.455830 second for No. 202 segment ).", "subpage_snippet": "", "source": "cis.temple.edu", "link": "https://cis.temple.edu/~tug29203/18spring-3329/reading//Lab_7_Solutions.pdf", "content": "The whole transmission time is the difference of the time instant of the first TCP segment (i.e., 0.026477 second for No.4 segment ) and the time instant of the last ACK (i.e., 5.455830 second for No. 202 segment )."}

data/sampled_jsons/2502.10875_arxiv_Table_1_dataset_statistics_ML-1M_Amazon_Beauty_Amazon_Toys_Amazon_Games_train_Du_us.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "luminati-io/Amazon-dataset-samples - GitHub", "date": "", "ddg_snippet": "A sample dataset of over 1,000 Amazon product listings, extracted using the Bright Data API, perfect for competitive analysis, market trends, and eCommerce insights.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/luminati-io/Amazon-dataset-samples", "content": "A sample dataset of over 1,000 Amazon product listings, extracted using the Bright Data API, perfect for competitive analysis, market trends, and eCommerce insights."}
+{"idx": 1, "title": "PALR: Personalization Aware LLMs for Recommendation - arXiv.org", "date": "", "ddg_snippet": "Evaluation under PALR framework on two public datasets demonstrate its competitive performance against state-of-the-art methods. We experimented with two datasets , MovieLens-1M[5], and Amazon Beauty[14] and demonstrated the strong potential of an LLM for recommendation in comparison to SOTA.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2305.07622", "content": "Evaluation under PALR framework on two public datasets demonstrate its competitive performance against state-of-the-art methods. We experimented with two datasets , MovieLens-1M[5], and Amazon Beauty[14] and demonstrated the strong potential of an LLM for recommendation in comparison to SOTA."}
+{"idx": 2, "title": "MovieLens 1M Dataset - GroupLens", "date": "", "ddg_snippet": "MovieLens 1M movie ratings. Stable benchmark dataset . 1 million ratings from 6000 users on 4000 movies. Released 2/2003. README.txt ml - 1m .zip (size: 6 MB, checksum) Permalink:", "subpage_snippet": "", "source": "grouplens.org", "link": "https://grouplens.org/datasets/movielens/1m/", "content": "MovieLens 1M movie ratings. Stable benchmark dataset . 1 million ratings from 6000 users on 4000 movies. Released 2/2003. README.txt ml - 1m .zip (size: 6 MB, checksum) Permalink:"}
+{"idx": 3, "title": "Do LLMs Memorize Recommendation Datasets? - arXiv.org", "date": "", "ddg_snippet": "The results in Table 2 are computed on MovieLens- 1M without any filtering, splitting the dataset into 80% for training and 20% for testing, using the leave-n-out paradigm, following (Zangerle and Bauer, 2023; Celma and Herrera, 2008; Cremonesi et al., 2008).", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2505.10212v1", "content": "The results in Table 2 are computed on MovieLens- 1M without any filtering, splitting the dataset into 80% for training and 20% for testing, using the leave-n-out paradigm, following (Zangerle and Bauer, 2023; Celma and Herrera, 2008; Cremonesi et al., 2008)."}
+{"idx": 4, "title": "GitHub - amazon-science/esci-data: Shopping Queries Dataset: A Large ...", "date": "", "ddg_snippet": "We introduce the \"Shopping Queries Data Set\", a large dataset of difficult search queries, released with the aim of fostering research in the area of semantic matching of queries and products. For each query, the dataset provides a list of up to 40 potentially relevant results, together with ESCI relevance judgements (Exact, Substitute, Complement, Irrelevant) indicating the relevance of ...", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/amazon-science/esci-data", "content": "We introduce the \"Shopping Queries Data Set\", a large dataset of difficult search queries, released with the aim of fostering research in the area of semantic matching of queries and products. For each query, the dataset provides a list of up to 40 potentially relevant results, together with ESCI relevance judgements (Exact, Substitute, Complement, Irrelevant) indicating the relevance of ..."}
+{"idx": 5, "title": "Amazon Sales Dataset | Kaggle", "date": "", "ddg_snippet": "This dataset is having the data of 1K+ Amazon Product's Ratings and Reviews", "subpage_snippet": "", "source": "www.kaggle.com", "link": "https://www.kaggle.com/datasets/karkavelrajaj/amazon-sales-dataset", "content": "This dataset is having the data of 1K+ Amazon Product's Ratings and Reviews"}
+{"idx": 6, "title": "MASSIVE - Amazon Science", "date": "", "ddg_snippet": "MASSIVE is a parallel dataset of > 1M utterances across 52 languages with annotations for the Natural Language Understanding tasks of intent prediction and slot annotation.", "subpage_snippet": "", "source": "www.amazon.science", "link": "https://www.amazon.science/code-and-datasets/massive", "content": "MASSIVE is a parallel dataset of > 1M utterances across 52 languages with annotations for the Natural Language Understanding tasks of intent prediction and slot annotation."}
+{"idx": 7, "title": "Performance Comparison of different methods on ML-1m, ML-10m, and...", "date": "", "ddg_snippet": "Performance Comparison of different methods on ML-1m , ML -10m, and Amazon datasets . Bold Scores are the best, while underlines scores are the second best.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/figure/Performance-Comparison-of-different-methods-on-ML-1m-ML-10m-and-Amazon-datasets-Bold_tbl2_354574308", "content": "Performance Comparison of different methods on ML-1m , ML -10m, and Amazon datasets . Bold Scores are the best, while underlines scores are the second best."}
+{"idx": 8, "title": "MASSIVE: A 1M-Example Multilingual Natural Language Understanding ...", "date": "", "ddg_snippet": "We present the MASSIVE dataset --Multilingual Amazon Slu resource package (SLURP) for Slot-filling, Intent classification, and Virtual assistant Evaluation. MASSIVE contains 1M realistic, parallel, labeled virtual assistant utterances spanning 51 languages, 18 domains, 60 intents, and 55 slots. MASSIVE was created by tasking professional translators to localize the English-only SLURP dataset ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2204.08582", "content": "We present the MASSIVE dataset --Multilingual Amazon Slu resource package (SLURP) for Slot-filling, Intent classification, and Virtual assistant Evaluation. MASSIVE contains 1M realistic, parallel, labeled virtual assistant utterances spanning 51 languages, 18 domains, 60 intents, and 55 slots. MASSIVE was created by tasking professional translators to localize the English-only SLURP dataset ..."}
+{"idx": 9, "title": "Review-based Recommender Systems: A Survey of Approaches, Challenges ...", "date": "", "ddg_snippet": "Amazon Review 1 : Amazon is one of the largest and most commonly used product review datasets , originating from Amazon website. With 21 distinct categories of items, this dataset is the most extensive publicly accessible product review dataset to date, encompassing user ID, item ID, rating, and textual feedback.", "subpage_snippet": "", "source": "dlnext.acm.org", "link": "https://dlnext.acm.org/doi/full/10.1145/3742421", "content": "Amazon Review 1 : Amazon is one of the largest and most commonly used product review datasets , originating from Amazon website. With 21 distinct categories of items, this dataset is the most extensive publicly accessible product review dataset to date, encompassing user ID, item ID, rating, and textual feedback."}

data/sampled_jsons/2502.10875_equation_1_FBox_score.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "[2502.10875] A Geometric Approach to Personalized ... DoniaGasmii/MNLP_M3_dpo_dataset · Datasets at Hugging Face w | PDF - Scribd Delivering contraceptive vaginal rings: Review of postpartum ... Australia’s response to the Horn of Africa humanitarian ... The Regulation of the Sea: Consolidation and Perspectives South Africa's 2024 election: What you need to know", "date": "", "ddg_snippet": "Feb 15, 2025 · Abstract page for arXiv paper 2502.10875 : A Geometric Approach to Personalized Recommendation with Set-Theoretic Constraints Using Box Embeddings We’re on a journey to advance and democratize artificial intelligence through open source and open science. formula and predicted the future of fanimation. This essay argues that Atlantis: The Lost Empire is a culturally fsignificant artifact: a bridge between traditional hand-drawn animation fand modern genre storytelling, a commentary on colonialism fand cultural preservation , and a philosophica l meditation on the cost of progress. fThrough its ... African journal of reproductive health, 2018 The progesterone vaginal ring (PVR) is a ring-shaped device designed for use by women in the postpartum period to regulate fertility by complementing and extending the contraceptive effectiveness of lactational amenorrhea to suppress ovulation. 1 It is available in eight Latin American countries for use by breastfeeding women who want more effective ... This evaluation rated Australia’s HoA response based on the 23 principles of good humanitarian donorship (Table 7). A simple numerical score was used, with 2 representing compliance, 1 representing partial compliance and 0 no compliance. The International Journal of Marine and Coastal Law, 2012 There is widespread and persisting non-compliance in relation to many of the provisions of the UN Convention on the Law of the Sea. This is a matter of serious concern because it undermines the integrity and legitimacy of the Convention, causes disputes and harms the marine environment. To remedy the situation, more use should be made ... May 6, 2024 · South Africans vote in national and provincial elections on May 29 that could test the African National Congress' 30-year rule. Item 1 of 2 Election posters of different political parties are seen, as South Afr ica prepares for national and provincial elections to be held on May 29, in Pretoria, South Afr ica, April 5, 2024.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2502.10875", "content": "Feb 15, 2025 · Abstract page for arXiv paper 2502.10875 : A Geometric Approach to Personalized Recommendation with Set-Theoretic Constraints Using Box Embeddings We’re on a journey to advance and democratize artificial intelligence through open source and open science. formula and predicted the future of fanimation. This essay argues that Atlantis: The Lost Empire is a culturally fsignificant artifact: a bridge between traditional hand-drawn animation fand modern genre storytelling, a commentary on colonialism fand cultural preservation , and a philosophica l meditation on the cost of progress. fThrough its ... African journal of reproductive health, 2018 The progesterone vaginal ring (PVR) is a ring-shaped device designed for use by women in the postpartum period to regulate fertility by complementing and extending the contraceptive effectiveness of lactational amenorrhea to suppress ovulation. 1 It is available in eight Latin American countries for use by breastfeeding women who want more effective ... This evaluation rated Australia’s HoA response based on the 23 principles of good humanitarian donorship (Table 7). A simple numerical score was used, with 2 representing compliance, 1 representing partial compliance and 0 no compliance. The International Journal of Marine and Coastal Law, 2012 There is widespread and persisting non-compliance in relation to many of the provisions of the UN Convention on the Law of the Sea. This is a matter of serious concern because it undermines the integrity and legitimacy of the Convention, causes disputes and harms the marine environment. To remedy the situation, more use should be made ... May 6, 2024 · South Africans vote in national and provincial elections on May 29 that could test the African National Congress' 30-year rule. Item 1 of 2 Election posters of different political parties are seen, as South Afr ica prepares for national and provincial elections to be held on May 29, in Pretoria, South Afr ica, April 5, 2024."}
+{"idx": 1, "title": "Nationals 5-3 Mets (Sep 20, 2025) Box Score - ESPN", "date": "", "ddg_snippet": "Box score for the Washington Nationals vs. New York Mets MLB game from September 20, 2025 on ESPN. Includes all pitching and batting stats.", "subpage_snippet": "", "source": "www.espn.com", "link": "https://www.espn.com/mlb/boxscore/_/gameId/401697223", "content": "Box score for the Washington Nationals vs. New York Mets MLB game from September 20, 2025 on ESPN. Includes all pitching and batting stats."}
+{"idx": 2, "title": "Systems of Equations Solver: Step-by-Step Solutions - Wolfram|Alpha", "date": "", "ddg_snippet": "Free Systems of Equations Calculator helps you solve sets of two or more equations . Linear, nonlinear, inequalities or general constraints. Answers, graphs, alternate forms.", "subpage_snippet": "", "source": "www.wolframalpha.com", "link": "https://www.wolframalpha.com/calculators/system-equation-calculator", "content": "Free Systems of Equations Calculator helps you solve sets of two or more equations . Linear, nonlinear, inequalities or general constraints. Answers, graphs, alternate forms."}
+{"idx": 3, "title": "Boxed and aligned Equations - LaTeX.org | Forum", "date": "", "ddg_snippet": "1 ) Can I set box in such a way that it includes equation numbers also? 2) Can I define a length and width for the box ?For boxing a group of aligned equations along with their labels, I use TikZ", "subpage_snippet": "", "source": "latex.org", "link": "https://latex.org/forum/viewtopic.php?t=19600", "content": "1 ) Can I set box in such a way that it includes equation numbers also? 2) Can I define a length and width for the box ?For boxing a group of aligned equations along with their labels, I use TikZ"}
+{"idx": 4, "title": "Australia’s response to the Horn of Africa humanitarian ...", "date": "", "ddg_snippet": "This evaluation rated Australia’s HoA response based on the 23 principles of good humanitarian donorship (Table 7). A simple numerical score was used, with 2 representing compliance, 1 representing partial compliance and 0 no compliance.", "subpage_snippet": "", "source": "www.academia.edu", "link": "https://www.academia.edu/144080488/Australia_s_response_to_the_Horn_of_Africa_humanitarian_crisis_2011", "content": "This evaluation rated Australia’s HoA response based on the 23 principles of good humanitarian donorship (Table 7). A simple numerical score was used, with 2 representing compliance, 1 representing partial compliance and 0 no compliance."}
+{"idx": 5, "title": "Cancer Cup Champions Meeting... | Umamusume: Pretty Derby｜Game8", "date": "", "ddg_snippet": "Winning Equation ...)Support Points x2,000. CM 1 Win Finish Box 1 Win. Carats x15. Goddess Statue x2.", "subpage_snippet": "", "source": "game8.co", "link": "https://game8.co/games/Umamusume-Pretty-Derby/archives/550078", "content": "Winning Equation ...)Support Points x2,000. CM 1 Win Finish Box 1 Win. Carats x15. Goddess Statue x2."}
+{"idx": 6, "title": "Дифференциальные уравнения. Пошаговый калькулятор", "date": "", "ddg_snippet": "integral icon Интегралы. equation icon Уравнения.", "subpage_snippet": "", "source": "mathdf.com", "link": "https://mathdf.com/dif/ru/", "content": "integral icon Интегралы. equation icon Уравнения."}
+{"idx": 7, "title": "Brazil vs USA - Semi-Finals - FIBA AmeriCup 2025 | FIBA Basketball", "date": "", "ddg_snippet": "Brazil vs USA - Browse boxscore , game information & live stats, team rosters, photos and video highlights.", "subpage_snippet": "", "source": "www.fiba.basketball", "link": "https://www.fiba.basketball/en/events/fiba-americup-2025/games/121420-BRA-USA", "content": "Brazil vs USA - Browse boxscore , game information & live stats, team rosters, photos and video highlights."}
+{"idx": 8, "title": "DoniaGasmii/MNLP_M3_dpo_dataset · Datasets at Hugging Face", "date": "", "ddg_snippet": "We’re on a journey to advance and democratize artificial intelligence through open source and open science.", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/datasets/DoniaGasmii/MNLP_M3_dpo_dataset/viewer", "content": "We’re on a journey to advance and democratize artificial intelligence through open source and open science."}
+{"idx": 9, "title": "w | PDF - Scribd", "date": "", "ddg_snippet": "formula and predicted the future of fanimation. This essay argues that Atlantis: The Lost Empire is a culturally fsignificant artifact: a bridge between traditional hand-drawn animation fand modern genre storytelling, a commentary on colonialism fand cultural preservation , and a philosophica l meditation on the cost of progress. fThrough its ...", "subpage_snippet": "", "source": "www.scribd.com", "link": "https://www.scribd.com/document/918780359/w", "content": "formula and predicted the future of fanimation. This essay argues that Atlantis: The Lost Empire is a culturally fsignificant artifact: a bridge between traditional hand-drawn animation fand modern genre storytelling, a commentary on colonialism fand cultural preservation , and a philosophica l meditation on the cost of progress. fThrough its ..."}

data/sampled_jsons/2502.20727_SPD_Figure_7c_LLaMA2-70B_1-node_LBW_70%_speedup_19.7%.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "SPD: Sync-Point Drop for Efficient Tensor Parallelism of ...", "date": "", "ddg_snippet": "by HB Kim · 2025 — For LLaMA2-70B with. LBW in Figure 7c, 70% SPD offers about 19.7% speedup while only sacrificing 0.94% accuracy. Note that this result is from ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2502.20727", "content": "by HB Kim · 2025 — For LLaMA2-70B with. LBW in Figure 7c, 70% SPD offers about 19.7% speedup while only sacrificing 0.94% accuracy. Note that this result is from ..."}
+{"idx": 1, "title": "[2502.20727] SPD: Sync-Point Drop for efficient tensor parallelism of ...", "date": "", "ddg_snippet": "For LLaMA2-70B with LBW in Figure 7 (c), 70% SPD offers about 19.7% speedup while only sacrificing 0.94% accuracy. Note that this result is from simple zero-shot dropping from ISBs.", "subpage_snippet": "", "source": "ar5iv.labs.arxiv.org", "link": "https://ar5iv.labs.arxiv.org/html/2502.20727", "content": "For LLaMA2-70B with LBW in Figure 7 (c), 70% SPD offers about 19.7% speedup while only sacrificing 0.94% accuracy. Note that this result is from simple zero-shot dropping from ISBs."}
+{"idx": 2, "title": "(PDF) SPD: Sync-Point Drop for efficient tensor ... - ResearchGate", "date": "", "ddg_snippet": "Therefore, we introduce a novel optimization technique, Sync-Point Drop ( SPD ), to reduce communication overheads in tensor parallelism by selectively dropping synchronization on attention outputs.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/389510233_SPD_Sync-Point_Drop_for_efficient_tensor_parallelism_of_Large_Language_Models", "content": "Therefore, we introduce a novel optimization technique, Sync-Point Drop ( SPD ), to reduce communication overheads in tensor parallelism by selectively dropping synchronization on attention outputs."}
+{"idx": 3, "title": "meta-llama/Llama-2-70b · Hugging Face", "date": "", "ddg_snippet": "Bigger models - 70B -- use Grouped-Query Attention (GQA) for improved inference scalability. Model Dates Llama 2 was trained between January 2023 and July 2023. Status This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/meta-llama/Llama-2-70b", "content": "Bigger models - 70B -- use Grouped-Query Attention (GQA) for improved inference scalability. Model Dates Llama 2 was trained between January 2023 and July 2023. Status This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback."}
+{"idx": 4, "title": "llama2:70b", "date": "", "ddg_snippet": "Llama 2 is a collection of foundation language models ranging from 7B to 70B parameters.", "subpage_snippet": "", "source": "ollama.com", "link": "https://ollama.com/library/llama2:70b", "content": "Llama 2 is a collection of foundation language models ranging from 7B to 70B parameters."}
+{"idx": 5, "title": "Fine-Tune Llama 2 70B on Intel® Gaudi® 2 AI Accelerators", "date": "", "ddg_snippet": "The Intel® Gaudi® software 1.13.0 release enabled Llama 2 70B fine-tuning on eight Intel® Gaudi® 2 cards with DeepSpeed ZeRO-3 optimization and LoRA. To improve the model's training performance, support was added for running the softmax in the attention layer in bfloat16 precision without compromising the accuracy of the outputs.", "subpage_snippet": "", "source": "www.intel.com", "link": "https://www.intel.com/content/www/us/en/developer/articles/llm/fine-tuning-llama2-70b-and-lora-on-gaudi2.html", "content": "The Intel® Gaudi® software 1.13.0 release enabled Llama 2 70B fine-tuning on eight Intel® Gaudi® 2 cards with DeepSpeed ZeRO-3 optimization and LoRA. To improve the model's training performance, support was added for running the softmax in the attention layer in bfloat16 precision without compromising the accuracy of the outputs."}
+{"idx": 6, "title": "Meta Code Llama 70B | Model Cards and Prompt formats", "date": "", "ddg_snippet": "Meta Code Llama 70B has a different prompt template compared to 34B, 13B and 7B. It starts with a Source: system tag—which can have an empty body—and continues with alternating user or assistant values.", "subpage_snippet": "", "source": "www.llama.com", "link": "https://www.llama.com/docs/model-cards-and-prompt-formats/meta-code-llama-70b/", "content": "Meta Code Llama 70B has a different prompt template compared to 34B, 13B and 7B. It starts with a Source: system tag—which can have an empty body—and continues with alternating user or assistant values."}
+{"idx": 7, "title": "LLama2 70B-LoRA - SMC", "date": "", "ddg_snippet": "Llama2 70B -LoRA is a large language model (LLM) that uses a technique called Low-Rank Adaptation (LoRA) to fine-tune its parameters on specific tasks or domains. This allows for efficient customization of the model without the need to retrain the entire 70 billion parameter model from scratch. The Llama2 70B -LoRA model can be used for various tasks, such as natural language generation ...", "subpage_snippet": "", "source": "smc.co", "link": "https://smc.co/mlperfv40/results/llama2-70b-lora", "content": "Llama2 70B -LoRA is a large language model (LLM) that uses a technique called Low-Rank Adaptation (LoRA) to fine-tune its parameters on specific tasks or domains. This allows for efficient customization of the model without the need to retrain the entire 70 billion parameter model from scratch. The Llama2 70B -LoRA model can be used for various tasks, such as natural language generation ..."}
+{"idx": 8, "title": "Llama 3.1 70B 24.11.1 (DGXC Benchmarking) | NVIDIA NGC", "date": "", "ddg_snippet": "This recipe contains information and scripts to produce performance results for the Llama 3.1 70B training workload.", "subpage_snippet": "", "source": "catalog.ngc.nvidia.com", "link": "https://catalog.ngc.nvidia.com/orgs/nvidia/teams/dgxc-benchmarking/resources/llama31-70b-dgxc-benchmarking-a", "content": "This recipe contains information and scripts to produce performance results for the Llama 3.1 70B training workload."}
+{"idx": 9, "title": "What's the difference between Llama 2 7B, 13B, and 70B?", "date": "", "ddg_snippet": "Let's break down the differences between the Llama 2 models and help you choose the right one for your use case.", "subpage_snippet": "", "source": "replicate.com", "link": "https://replicate.com/blog/all-the-llamas", "content": "Let's break down the differences between the Llama 2 models and help you choose the right one for your use case."}

data/sampled_jsons/2503.06366_mHeight_definition_3412_patterns_sitearxiv.org.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "[2503.05652] BEHAVIOR Robot Suite: Streamlining Real-World...", "date": "", "ddg_snippet": "Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG). Cite as(or arXiv:2503.05652v1 [cs.RO] for this version).", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2503.05652", "content": "Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG). Cite as(or arXiv:2503.05652v1 [cs.RO] for this version)."}
+{"idx": 1, "title": "[ 2503 . 06366 ] Machine Learning meets Algebraic Combinatorics...", "date": "", "ddg_snippet": "Further differentiating our dataset collection is the fact that it aims at the conjecturing process. Each dataset includes an open-ended research-level question and a large collection of examples (up to 10M in some cases) from which conjectures should be generated.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2503.06366", "content": "Further differentiating our dataset collection is the fact that it aims at the conjecturing process. Each dataset includes an open-ended research-level question and a large collection of examples (up to 10M in some cases) from which conjectures should be generated."}
+{"idx": 2, "title": "[2503.06311v1] Hybrid CNN-Dilated Self-attention Model Using Inertial...", "date": "", "ddg_snippet": "While human body capacitance ($HBC$) has been explored as a novel wearable motion sensing modality, its competence has never been quantitatively demonstrated compared to that of the dominant inertial measurement unit ($IMU$) in practical scenarios.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2503.06311v1", "content": "While human body capacitance ($HBC$) has been explored as a novel wearable motion sensing modality, its competence has never been quantitatively demonstrated compared to that of the dominant inertial measurement unit ($IMU$) in practical scenarios."}
+{"idx": 3, "title": "Machine Learning meets Algebraic Combinatorics: A Suite of Datasets...", "date": "", "ddg_snippet": "arXiv: 2503 . 06366 v1 [cs.LG] 9 Mar 2025.ML task: Predict the mHeight of a permutation. Since mHeight can take a limited number of values for small n, this is framed as a classification problem.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2503.06366", "content": "arXiv: 2503 . 06366 v1 [cs.LG] 9 Mar 2025.ML task: Predict the mHeight of a permutation. Since mHeight can take a limited number of values for small n, this is framed as a classification problem."}
+{"idx": 4, "title": "World Attack Frequency Rates – International Shark Attack File", "date": "", "ddg_snippet": "Jan 6, 2022 · This graph of unprovoked shark attacks in the world from 1960 through the present also shows the portion occurring in the United States for comparison. The drop in recorded attacks in 1969 is a result of a loss of funding to support the ISAF, and hence less effort reporting and organizing of attack", "subpage_snippet": "", "source": "www.floridamuseum.ufl.edu", "link": "https://www.floridamuseum.ufl.edu/shark-attacks/trends/frequency-rates/world/", "content": "Jan 6, 2022 · This graph of unprovoked shark attacks in the world from 1960 through the present also shows the portion occurring in the United States for comparison. The drop in recorded attacks in 1969 is a result of a loss of funding to support the ISAF, and hence less effort reporting and organizing of attack"}
+{"idx": 5, "title": "Machine Learning meets Algebraic Combinatorics: A Suite of Datasets...", "date": "", "ddg_snippet": "arXiv: 2503 . 06366 v1 [cs.LG] 09 Mar 2025.The definition of a semistandard Young tableau is analogous except that the entries are only assumed to weakly increase as one moves from left to right along a row (see Figure 1 (right)).", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2503.06366v1", "content": "arXiv: 2503 . 06366 v1 [cs.LG] 09 Mar 2025.The definition of a semistandard Young tableau is analogous except that the entries are only assumed to weakly increase as one moves from left to right along a row (see Figure 1 (right))."}
+{"idx": 6, "title": "Let G be a simply connected semisimple complex Lie group, with Borel", "date": "", "ddg_snippet": "The following statistic on occurrences of the pattern 3412 will be of special importance for us (see Theorem 1.6).Let mCont(w) be the minimum content of a 3412 pattern in w. Lemma 3.9. For w ∈ Sn that contains 3412 , mHeight (w) = mCont(w).", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2303.13695", "content": "The following statistic on occurrences of the pattern 3412 will be of special importance for us (see Theorem 1.6).Let mCont(w) be the minimum content of a 3412 pattern in w. Lemma 3.9. For w ∈ Sn that contains 3412 , mHeight (w) = mCont(w)."}
+{"idx": 7, "title": "Shark Attacks Plummeted Worldwide Last Year", "date": "", "ddg_snippet": "Feb 11, 2025 · TUESDAY, Feb. 11, 2025 (HealthDay News) -- It was a bit safer to go into the water last year: Unprovoked shark attacks worldwide fell by about a third in 2024 compared to the year before, new data show. The United States continues to be the most likely country in which to get bitten by a shark , according to the annual International Shark Attack File from the Florida Museum of Natural History ...", "subpage_snippet": "", "source": "www.healthday.com", "link": "https://www.healthday.com/health-news/first-aid-and-emergencies/shark-attacks-plummeted-worldwide-last-year", "content": "Feb 11, 2025 · TUESDAY, Feb. 11, 2025 (HealthDay News) -- It was a bit safer to go into the water last year: Unprovoked shark attacks worldwide fell by about a third in 2024 compared to the year before, new data show. The United States continues to be the most likely country in which to get bitten by a shark , according to the annual International Shark Attack File from the Florida Museum of Natural History ..."}
+{"idx": 8, "title": "Maps & Data – International Shark Attack File - Florida Museum", "date": "", "ddg_snippet": "Aug 30, 2018 · Explore maps and data on shark attacks worldwide from the International Shark Attack File at Florida Museum .", "subpage_snippet": "", "source": "www.floridamuseum.ufl.edu", "link": "https://www.floridamuseum.ufl.edu/shark-attacks/maps/", "content": "Aug 30, 2018 · Explore maps and data on shark attacks worldwide from the International Shark Attack File at Florida Museum ."}
+{"idx": 9, "title": "Deaths from shark attacks rise dramatically, new data show", "date": "", "ddg_snippet": "Feb 5, 2024 · There were 69 unprovoked shark attacks in 2023, higher than the five-year average of 63 attacks per year, according to the University of Florida 's International Shark Attack File.", "subpage_snippet": "", "source": "www.upi.com", "link": "https://www.upi.com/Science_News/2024/02/05/shark-attacks-deaths/8221707145849/", "content": "Feb 5, 2024 · There were 69 unprovoked shark attacks in 2023, higher than the five-year average of 63 attacks per year, according to the University of Florida 's International Shark Attack File."}

data/sampled_jsons/2503.16979_Equation_6_motion_interpolation_Gaussian.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Polynomial interpolation - Wikipedia", "date": "", "ddg_snippet": "In numerical analysis, polynomial interpolation is the interpolation of a given data set by the polynomial of lowest possible degree that passes through the points in the dataset. Given a set of n + 1 data points. , with no two. the same, a polynomia...", "subpage_snippet": "", "source": "en.wikipedia.org", "link": "https://en.wikipedia.org/wiki/Polynomial_interpolation", "content": "In numerical analysis, polynomial interpolation is the interpolation of a given data set by the polynomial of lowest possible degree that passes through the points in the dataset. Given a set of n + 1 data points. , with no two. the same, a polynomia..."}
+{"idx": 1, "title": "[2503.16979] Instant Gaussian Stream: Fast and Generalizable ... GitHub - riotofbug/dynamic-3dgs-papers: Dynamic Scene ... Motion Matters: Compact Gaussian Streaming for Free-Viewpoint ... Ronggang Wang - dblp arXiv:2503.16979v1 [cs.CV] 21 Mar 2025 Electron. Commun. Probab. 19 (2014), no. 35, DOI: 10.1214/ECP ... arXiv: 2503 .16979v1 [cs.CV] 21 Mar 2025 Instant Gaussian Stream : Fast and Generalizable Streaming of Dyn… arXiv: 2503 .16979v1 [cs.CV] 21 Mar 2025 A Semi-Analytical Approach to Solving the Black-Scholes ...", "date": "", "ddg_snippet": "Mar 21, 2025 · In this paper, we propose Instant Gaussian Stream (IGS), a fast and generalizable streaming framework, to address these issues. First, we introduce a generalized Anchor-driven Gaussian Motion Network, which projects multi-view 2D motion features into 3D space, using anchor points to drive the motion of all Gaussians. Dynamic Scene Representation Gaussian Splatting. Contribute to riotofbug/dynamic-3dgs-papers development by creating an account on GitHub . May 22, 2025 · To address this limitation, we propose a novel Compact Gaussian Streaming (ComGS) framework, leveraging the locality and consistency of motion in dynamic scene, that models object-consistent ... [i38] Jiacong Chen, Qingyu Mao, Youneng Bao, Xiandong Meng, Fanyang Meng, Ronggang Wang, Yongsheng Liang: Motion Matters: Compact Gaussian Streaming for Free-Viewpoint Video Reconstruction. CoRR abs/2505.16533 (2025) We propose a generalized Anchor-driven Gaussian Mo-tion Network that captures Gaussian motion between ad-jacent frames with a single inference, eliminating the need for frame-by-frame optimization. R weighted by the standard Gaussian measure (see [12]). The case p = 2, k = 0, b = 0, and q = 1 was considered in [4] for the one-dimensional case (in particular, the process X is a martingale). How can a generalized network drive the motion of all Gaussians? to drive the mo-tion of all Gaussians. This generalized Network gener-ates the motion of Gaussians for each target frame in th time required for a single inference. Second, we propose a Key-frame-guided Streaming Strategy that refines each key frame, enabling accurate reconstruction of temporally complex scen What is a generalized anchor-driven Gaussian motion network? First, we introduce a generalized Anchor-driven Gaussian Motion Network, which projects multi-view 2D motion features into 3D space , using anchor points to drive the motion of all Gaussians. This generalized Network generates the motion of Gaussians for each target frame in the time required for a single inference. What is anchor-driven Gaussian motion Network (AGM-net)? n overview of the pipeline in Sec.3.1. Then, in Sec. 3.3, we introduce the Anchor-driven Gaussian Motion Network (AGM-Net), a general-ized model that drives Gaussian motion from the previous frame using anchor points, whi h serve as key points in the 3D scene. Following this, we present our Key-fram Abstract This paper proposes a semi-analytical method for solving the Black-Scholes equation using the framework of Reproducing Kernel Hilbert Spaces (RKHS). By embedding the solution space into an RKHS defined by a positive definite kernel, the problem is reformulated as a regularized interpolation task based on observed data.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2503.16979", "content": "Mar 21, 2025 · In this paper, we propose Instant Gaussian Stream (IGS), a fast and generalizable streaming framework, to address these issues. First, we introduce a generalized Anchor-driven Gaussian Motion Network, which projects multi-view 2D motion features into 3D space, using anchor points to drive the motion of all Gaussians. Dynamic Scene Representation Gaussian Splatting. Contribute to riotofbug/dynamic-3dgs-papers development by creating an account on GitHub . May 22, 2025 · To address this limitation, we propose a novel Compact Gaussian Streaming (ComGS) framework, leveraging the locality and consistency of motion in dynamic scene, that models object-consistent ... [i38] Jiacong Chen, Qingyu Mao, Youneng Bao, Xiandong Meng, Fanyang Meng, Ronggang Wang, Yongsheng Liang: Motion Matters: Compact Gaussian Streaming for Free-Viewpoint Video Reconstruction. CoRR abs/2505.16533 (2025) We propose a generalized Anchor-driven Gaussian Mo-tion Network that captures Gaussian motion between ad-jacent frames with a single inference, eliminating the need for frame-by-frame optimization. R weighted by the standard Gaussian measure (see [12]). The case p = 2, k = 0, b = 0, and q = 1 was considered in [4] for the one-dimensional case (in particular, the process X is a martingale). How can a generalized network drive the motion of all Gaussians? to drive the mo-tion of all Gaussians. This generalized Network gener-ates the motion of Gaussians for each target frame in th time required for a single inference. Second, we propose a Key-frame-guided Streaming Strategy that refines each key frame, enabling accurate reconstruction of temporally complex scen What is a generalized anchor-driven Gaussian motion network? First, we introduce a generalized Anchor-driven Gaussian Motion Network, which projects multi-view 2D motion features into 3D space , using anchor points to drive the motion of all Gaussians. This generalized Network generates the motion of Gaussians for each target frame in the time required for a single inference. What is anchor-driven Gaussian motion Network (AGM-net)? n overview of the pipeline in Sec.3.1. Then, in Sec. 3.3, we introduce the Anchor-driven Gaussian Motion Network (AGM-Net), a general-ized model that drives Gaussian motion from the previous frame using anchor points, whi h serve as key points in the 3D scene. Following this, we present our Key-fram Abstract This paper proposes a semi-analytical method for solving the Black-Scholes equation using the framework of Reproducing Kernel Hilbert Spaces (RKHS). By embedding the solution space into an RKHS defined by a positive definite kernel, the problem is reformulated as a regularized interpolation task based on observed data."}
+{"idx": 2, "title": "COLMAP-Free 3D Gaussian Splatting | PDF", "date": "", "ddg_snippet": "Abstract approaches in view synthesis and camera pose estimation. under large motion changes. Our project page is https: While neural rendering has led to ...", "subpage_snippet": "", "source": "www.scribd.com", "link": "https://www.scribd.com/document/789740501/2312-07504v2", "content": "Abstract approaches in view synthesis and camera pose estimation. under large motion changes. Our project page is https: While neural rendering has led to ..."}
+{"idx": 3, "title": "Motion Matters: Compact Gaussian Streaming for Free-Viewpoint ...", "date": "", "ddg_snippet": "May 22, 2025 · To address this limitation, we propose a novel Compact Gaussian Streaming (ComGS) framework, leveraging the locality and consistency of motion in dynamic scene, that models object-consistent ...", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/391992317_Motion_Matters_Compact_Gaussian_Streaming_for_Free-Viewpoint_Video_Reconstruction", "content": "May 22, 2025 · To address this limitation, we propose a novel Compact Gaussian Streaming (ComGS) framework, leveraging the locality and consistency of motion in dynamic scene, that models object-consistent ..."}
+{"idx": 4, "title": "arXiv:2503.16979v1 [cs.CV] 21 Mar 2025", "date": "", "ddg_snippet": "We propose a generalized Anchor-driven Gaussian Mo-tion Network that captures Gaussian motion between ad-jacent frames with a single inference, eliminating the need for frame-by-frame optimization.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2503.16979", "content": "We propose a generalized Anchor-driven Gaussian Mo-tion Network that captures Gaussian motion between ad-jacent frames with a single inference, eliminating the need for frame-by-frame optimization."}
+{"idx": 5, "title": "Gauss 's Forward Interpolation - GeeksforGeeks", "date": "", "ddg_snippet": "Interpolation refers to the process of creating new data points given within the given set of data. The below code computes the desired data point within the given range of discrete data sets using the formula given by Gauss and this method is known as Gauss 's Forward Method.", "subpage_snippet": "", "source": "www.geeksforgeeks.org", "link": "https://www.geeksforgeeks.org/python/gausss-forward-interpolation/", "content": "Interpolation refers to the process of creating new data points given within the given set of data. The below code computes the desired data point within the given range of discrete data sets using the formula given by Gauss and this method is known as Gauss 's Forward Method."}
+{"idx": 6, "title": "math - How to use Gaussian interpolation ? - Stack Overflow", "date": "", "ddg_snippet": "I heard about Gaussian interpolation , but don't know how to use it.Then he used \" Gaussian interpolation \" and was able to make the path either smooth and rounded or straight and anywhere in between.", "subpage_snippet": "", "source": "stackoverflow.com", "link": "https://stackoverflow.com/questions/62345860/how-to-use-gaussian-interpolation", "content": "I heard about Gaussian interpolation , but don't know how to use it.Then he used \" Gaussian interpolation \" and was able to make the path either smooth and rounded or straight and anywhere in between."}
+{"idx": 7, "title": "Interpolation - Gauss Backward Central Difference Formula Video...", "date": "", "ddg_snippet": "The notes and questions for Interpolation - Gauss Backward Central Difference Formula have been prepared according to the Civil Engineering (CE) exam syllabus.", "subpage_snippet": "", "source": "edurev.in", "link": "https://edurev.in/v/252835/Interpolation-Gauss-Backward-Central-Difference-Formula", "content": "The notes and questions for Interpolation - Gauss Backward Central Difference Formula have been prepared according to the Civil Engineering (CE) exam syllabus."}
+{"idx": 8, "title": "Gaussian interpolation revisited | Request PDF", "date": "", "ddg_snippet": "... Another popular RBF is Gaussian which was first proposed by a geophysicist Krige [23] for the data interpolation and further developed by Matheron [24] . Wendland [25] refined the error bound for Gaussian which was known for spectral convergence as MQs.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/268885345_Gaussian_interpolation_revisited", "content": "... Another popular RBF is Gaussian which was first proposed by a geophysicist Krige [23] for the data interpolation and further developed by Matheron [24] . Wendland [25] refined the error bound for Gaussian which was known for spectral convergence as MQs."}
+{"idx": 9, "title": "Gauss interpolation formula - Encyclopedia of Mathematics", "date": "", "ddg_snippet": "A formula in which the nodes nearest to the interpolation point $ x $ are used as interpolation nodes. If $ x = x _ {0} + th $, the formula. $$ \\tag{1 } G _ {2n + 1 } = \\ f _ {0} + f _ {1/2} ^ { 1 } t + f _ {0} ^ { 2 } \\frac{t }{2!} + \\dots + ... +...", "subpage_snippet": "", "source": "encyclopediaofmath.org", "link": "https://encyclopediaofmath.org/wiki/Gauss_interpolation_formula", "content": "A formula in which the nodes nearest to the interpolation point $ x $ are used as interpolation nodes. If $ x = x _ {0} + th $, the formula. $$ \\tag{1 } G _ {2n + 1 } = \\ f _ {0} + f _ {1/2} ^ { 1 } t + f _ {0} ^ { 2 } \\frac{t }{2!} + \\dots + ... +..."}

data/sampled_jsons/27tMzmzDjO_A_Geometric_Approach_to_Personalized_Recommendation_with_Set-Theoretic_Constraints_Using_.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "A Geometric Approach to Personalized Recommendation with ...", "date": "", "ddg_snippet": "In this work, we formulate the problem of personalized item recommendation as matrix completion where rows are set - theoretically dependent. To capture this set - theoretic dependence we represent each user and attribute by a hyperrectangle or box (i.e. a Cartesian product of intervals).", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/forum?id=27tMzmzDjO&referrer=[the+profile+of+Andrew+McCallum](/profile?id=~Andrew_McCallum1)", "content": "In this work, we formulate the problem of personalized item recommendation as matrix completion where rows are set - theoretically dependent. To capture this set - theoretic dependence we represent each user and attribute by a hyperrectangle or box (i.e. a Cartesian product of intervals)."}
+{"idx": 1, "title": "A Geometric Approach to Personalized Recommendation with ...", "date": "", "ddg_snippet": "Box embeddings , with their geometric set operations, sig-nificantly outperform all vector-based methods.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2502.10875", "content": "Box embeddings , with their geometric set operations, sig-nificantly outperform all vector-based methods."}
+{"idx": 2, "title": "(PDF) A Geometric Approach to Personalized Recommendation ...", "date": "", "ddg_snippet": "with Set - Theoretic Constraints Using Box Embeddings .of set - theoretic query recommendation . Box embeddings , with their geometric set operations, sig-. nificantly outperform all vector-based methods.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/389091382_A_Geometric_Approach_to_Personalized_Recommendation_with_Set-Theoretic_Constraints_Using_Box_Embeddings", "content": "with Set - Theoretic Constraints Using Box Embeddings .of set - theoretic query recommendation . Box embeddings , with their geometric set operations, sig-. nificantly outperform all vector-based methods."}
+{"idx": 3, "title": "A Geometric Approach to Personalized Recommendation with ...", "date": "", "ddg_snippet": "Box embeddings , with their geometric set operations, significantly outperform all vector-based methods.This paper aims to advance the field of Machine Learning by introducing a geometric approach to personalized recommendation under set - theoretic constraints .", "subpage_snippet": "", "source": "bytez.com", "link": "https://bytez.com/docs/icml/46603/paper", "content": "Box embeddings , with their geometric set operations, significantly outperform all vector-based methods.This paper aims to advance the field of Machine Learning by introducing a geometric approach to personalized recommendation under set - theoretic constraints ."}
+{"idx": 4, "title": "GitHub - pko89403/ Recommendation -paper-daily: Automatically...", "date": "", "ddg_snippet": "A Geometric Approach to Personalized Recommendation with Set - Theoretic Constraints Using Box Embeddings .A multi-theoretical kernel-based approach to social network-based recommendation . Xin Li et.al. 2412.12202.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/pko89403/Recommendation-paper-daily", "content": "A Geometric Approach to Personalized Recommendation with Set - Theoretic Constraints Using Box Embeddings .A multi-theoretical kernel-based approach to social network-based recommendation . Xin Li et.al. 2412.12202."}
+{"idx": 5, "title": "PhD Thesis Defense: Shib Dasgupta, Box Embeddings as...", "date": "", "ddg_snippet": "To address this, I frame personalized item search as matrix completion with set - theoretic constraints , where both users and attributes are represented as hyper-rectangles ( boxes ).", "subpage_snippet": "", "source": "www.cics.umass.edu", "link": "https://www.cics.umass.edu/events/phd-thesis-defense-shib-dasgupta", "content": "To address this, I frame personalized item search as matrix completion with set - theoretic constraints , where both users and attributes are represented as hyper-rectangles ( boxes )."}
+{"idx": 6, "title": "Knowledge Representation: The Promising Power of Box Embeddings", "date": "", "ddg_snippet": "Introducing Box Embeddings : A Geometric Approach to Knowledge Representation.For instance, the concept of “sci-fi movies” could be represented as a large box that contains the individual movie embeddings within its boundaries, naturally capturing the set -based relationship.", "subpage_snippet": "", "source": "ai.plainenglish.io", "link": "https://ai.plainenglish.io/knowledge-representation-the-promising-power-of-box-embeddings-d0cbcf1dc93c", "content": "Introducing Box Embeddings : A Geometric Approach to Knowledge Representation.For instance, the concept of “sci-fi movies” could be represented as a large box that contains the individual movie embeddings within its boundaries, naturally capturing the set -based relationship."}
+{"idx": 7, "title": "When Box Meets Graph Neural Network in Tag-aware... | CoLab", "date": "", "ddg_snippet": "Word2 Box : Capturing Set - Theoretic Semantics of Words using Box Embeddings .", "subpage_snippet": "", "source": "colab.ws", "link": "https://colab.ws/articles/10.1145/3637528.3671973", "content": "Word2 Box : Capturing Set - Theoretic Semantics of Words using Box Embeddings ."}
+{"idx": 8, "title": "Box embeddings for extending ontologies: a data-driven and...", "date": "", "ddg_snippet": "Thus, we used box -shaped embeddings of ontology classes to learn and transparently represent logical relationships that are only implicit in multi-label datasets.", "subpage_snippet": "", "source": "pubmed.ncbi.nlm.nih.gov", "link": "https://pubmed.ncbi.nlm.nih.gov/40890838/", "content": "Thus, we used box -shaped embeddings of ontology classes to learn and transparently represent logical relationships that are only implicit in multi-label datasets."}
+{"idx": 9, "title": "Shib Sankar Dasgupta - Google Scholar", "date": "", "ddg_snippet": "A Geometric Approach to Personalized Recommendation with Set - Theoretic Constraints Using Box Embeddings .", "subpage_snippet": "", "source": "scholar.google.com", "link": "https://scholar.google.com/citations?user=0KpQR94AAAAJ&hl=en", "content": "A Geometric Approach to Personalized Recommendation with Set - Theoretic Constraints Using Box Embeddings ."}

data/sampled_jsons/27tMzmzDjO_Table_1_statistics_ML-1M_Amazon_Beauty_Amazon_Toys_Amazon_Games_train_Du_users.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "luminati-io/Amazon-dataset-samples - GitHub", "date": "", "ddg_snippet": "What are the Amazon datasets use cases? 1 . Optimize Your Inventory & Pricing Strategy Discover top-selling products that customers are searching for and buying. Enhance your entire inventory, pricing, supply chain, and marketing approach for maximum efficiency.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/luminati-io/Amazon-dataset-samples", "content": "What are the Amazon datasets use cases? 1 . Optimize Your Inventory & Pricing Strategy Discover top-selling products that customers are searching for and buying. Enhance your entire inventory, pricing, supply chain, and marketing approach for maximum efficiency."}
+{"idx": 1, "title": "Amazon Review Dataset数据集介绍 - CSDN博客 Creating and Using Datasources - Amazon Machine Learning Amazon Review Data (2018) - GitHub Pages Amazon-book, Yelp, and ML-1M datasets - Dataset - LDM kaiboon0216/Amazon-beauty-products-analysis - GitHub", "date": "", "ddg_snippet": "Jan 29, 2021 · Amazon Review Dataset 数据集 记录了用户对亚马逊网站商品的评价，是推荐系统的经典数据集，并且 Amazon 一直在更新这个数据集，根据时间顺序， Amazon 数据集可以分成三类： You can use Amazon ML datasources to train an ML model, evaluate an ML model, and generate batch predictions using an ML model. Datasource objects contain metadata about your input data. When you create a datasource, Amazon ML reads your input data, computes descriptive statistics on its attributes, and stores the statistics , a schema, and other information as part of the datasource object ... This Dataset is an updated version of the Amazon review datasetreleased in 2014. As in the previous version, this dataset includes reviews (ratings, text, helpfulness votes), product metadata (descriptions, category information, price, brand, and image features), and links (also viewed/also bought graphs). In addition, this version provides the fol... See full list on nijianmo.github.io Please cite the following paper if you use the data in any way: Justifying recommendations using distantly-labeled reviews and fined-grained aspects Jianmo Ni, Jiacheng Li, Julian McAuley Empirical Methods in Natural Language Processing (EMNLP), 2019 pdf See full list on nijianmo.github.io 05/2021 We updated high resolution image urlsto the metadata! 08/2020 We have updated the metadata and now it includes much less HTML/CSS code. Feel free to download the updated data! See full list on nijianmo.github.io We provide a colab notebookthat helps you parse and clean the data. For example: 1 . Load the metadata (e.g. as JSON or DataFrame) 2. Check if title has HTML contents and filter them We provide a colab notebookthat helps you find target products and obtain their reviews! We appreciate any help or feedback to improve the quality of our dataset! Feel ... See full list on nijianmo.github.io Complete review data Please only download these (large!) files if you really need them. We recommend using the smaller datasets (i.e. k-core and CSV files) as shown in the next section. raw review data(34gb) - all 233. 1 million reviews ratings only(6.7gb) - same as above, in csv form without reviews or metadata 5-core(14.3gb) - subset of the data in which all users and items have at least 5 reviews (75.26 million reviews) Per-category data- the review and product metadata for each category. To download the comple... \"Small\" subsets for experimentation If you're using this data for a class project (or similar) please consider using one of these smaller datasets below before requesting the larger files. K-cores (i.e., dense subsets): These data have been reduced to extract the k-core, such that each of the remaining users and items have k reviews each. Ratings only: These datasets include no metadata or reviews, but only (item,user,rating,timestamp) tuples. Thus they are suitable for use with mymedialite(or similar) packages. You can directl... Data format Format is one-review-per-line in json. See examples below for further help reading the data. See full list on nijianmo.github.io Reading the data Data can be treated as python dictionary objects. A simple script to read any of the above the data is as follows: Pandas data frame This code reads the data into a pandas data frame: Example: latent-factor model in mymedialite Predicts ratings from a rating-only CSV file See full list on nijianmo.github.io Data and Resources Original Metadata JSON The json representation of the dataset with its distributions based on DCAT. Amazon-beauty -products-analysis Using data science process to gain business insights from the Amazon beauty datasets and determine the factors that contribute to the high sales of a beauty product Our anaylsis are focused on answering the following questions:", "subpage_snippet": "", "source": "blog.csdn.net", "link": "https://blog.csdn.net/springtostring/article/details/113407712", "content": "Jan 29, 2021 · Amazon Review Dataset 数据集 记录了用户对���马逊网站商品的评价，是推荐系统的经典数据集，并且 Amazon 一直在更新这个数据集，根据时间顺序， Amazon 数据集可以分成三类： You can use Amazon ML datasources to train an ML model, evaluate an ML model, and generate batch predictions using an ML model. Datasource objects contain metadata about your input data. When you create a datasource, Amazon ML reads your input data, computes descriptive statistics on its attributes, and stores the statistics , a schema, and other information as part of the datasource object ... This Dataset is an updated version of the Amazon review datasetreleased in 2014. As in the previous version, this dataset includes reviews (ratings, text, helpfulness votes), product metadata (descriptions, category information, price, brand, and image features), and links (also viewed/also bought graphs). In addition, this version provides the fol... See full list on nijianmo.github.io Please cite the following paper if you use the data in any way: Justifying recommendations using distantly-labeled reviews and fined-grained aspects Jianmo Ni, Jiacheng Li, Julian McAuley Empirical Methods in Natural Language Processing (EMNLP), 2019 pdf See full list on nijianmo.github.io 05/2021 We updated high resolution image urlsto the metadata! 08/2020 We have updated the metadata and now it includes much less HTML/CSS code. Feel free to download the updated data! See full list on nijianmo.github.io We provide a colab notebookthat helps you parse and clean the data. For example: 1 . Load the metadata (e.g. as JSON or DataFrame) 2. Check if title has HTML contents and filter them We provide a colab notebookthat helps you find target products and obtain their reviews! We appreciate any help or feedback to improve the quality of our dataset! Feel ... See full list on nijianmo.github.io Complete review data Please only download these (large!) files if you really need them. We recommend using the smaller datasets (i.e. k-core and CSV files) as shown in the next section. raw review data(34gb) - all 233. 1 million reviews ratings only(6.7gb) - same as above, in csv form without reviews or metadata 5-core(14.3gb) - subset of the data in which all users and items have at least 5 reviews (75.26 million reviews) Per-category data- the review and product metadata for each category. To download the comple... \"Small\" subsets for experimentation If you're using this data for a class project (or similar) please consider using one of these smaller datasets below before requesting the larger files. K-cores (i.e., dense subsets): These data have been reduced to extract the k-core, such that each of the remaining users and items have k reviews each. Ratings only: These datasets include no metadata or reviews, but only (item,user,rating,timestamp) tuples. Thus they are suitable for use with mymedialite(or similar) packages. You can directl... Data format Format is one-review-per-line in json. See examples below for further help reading the data. See full list on nijianmo.github.io Reading the data Data can be treated as python dictionary objects. A simple script to read any of the above the data is as follows: Pandas data frame This code reads the data into a pandas data frame: Example: latent-factor model in mymedialite Predicts ratings from a rating-only CSV file See full list on nijianmo.github.io Data and Resources Original Metadata JSON The json representation of the dataset with its distributions based on DCAT. Amazon-beauty -products-analysis Using data science process to gain business insights from the Amazon beauty datasets and determine the factors that contribute to the high sales of a beauty product Our anaylsis are focused on answering the following questions:"}
+{"idx": 2, "title": "Creating and Using Datasources - Amazon Machine Learning", "date": "", "ddg_snippet": "You can use Amazon ML datasources to train an ML model, evaluate an ML model, and generate batch predictions using an ML model. Datasource objects contain metadata about your input data. When you create a datasource, Amazon ML reads your input data, computes descriptive statistics on its attributes, and stores the statistics , a schema, and other information as part of the datasource object ...", "subpage_snippet": "", "source": "docs.aws.amazon.com", "link": "https://docs.aws.amazon.com/machine-learning/latest/dg/creating-and-using-datasources.html", "content": "You can use Amazon ML datasources to train an ML model, evaluate an ML model, and generate batch predictions using an ML model. Datasource objects contain metadata about your input data. When you create a datasource, Amazon ML reads your input data, computes descriptive statistics on its attributes, and stores the statistics , a schema, and other information as part of the datasource object ..."}
+{"idx": 3, "title": "Amazon Review Data (2018) - GitHub Pages Amazon-book, Yelp, and ML-1M datasets - Dataset - LDM kaiboon0216/Amazon-beauty-products-analysis - GitHub", "date": "", "ddg_snippet": "This Dataset is an updated version of the Amazon review datasetreleased in 2014. As in the previous version, this dataset includes reviews (ratings, text, helpfulness votes), product metadata (descriptions, category information, price, brand, and image features), and links (also viewed/also bought graphs). In addition, this version provides the fol... See full list on nijianmo.github.io Please cite the following paper if you use the data in any way: Justifying recommendations using distantly-labeled reviews and fined-grained aspects Jianmo Ni, Jiacheng Li, Julian McAuley Empirical Methods in Natural Language Processing (EMNLP), 2019 pdf See full list on nijianmo.github.io 05/2021 We updated high resolution image urlsto the metadata! 08/2020 We have updated the metadata and now it includes much less HTML/CSS code. Feel free to download the updated data! See full list on nijianmo.github.io We provide a colab notebookthat helps you parse and clean the data. For example: 1 . Load the metadata (e.g. as JSON or DataFrame) 2. Check if title has HTML contents and filter them We provide a colab notebookthat helps you find target products and obtain their reviews! We appreciate any help or feedback to improve the quality of our dataset! Feel ... See full list on nijianmo.github.io Complete review data Please only download these (large!) files if you really need them. We recommend using the smaller datasets (i.e. k-core and CSV files) as shown in the next section. raw review data(34gb) - all 233. 1 million reviews ratings only(6.7gb) - same as above, in csv form without reviews or metadata 5-core(14.3gb) - subset of the data in which all users and items have at least 5 reviews (75.26 million reviews) Per-category data- the review and product metadata for each category. To download the comple... \"Small\" subsets for experimentation If you're using this data for a class project (or similar) please consider using one of these smaller datasets below before requesting the larger files. K-cores (i.e., dense subsets): These data have been reduced to extract the k-core, such that each of the remaining users and items have k reviews each. Ratings only: These datasets include no metadata or reviews, but only (item,user,rating,timestamp) tuples. Thus they are suitable for use with mymedialite(or similar) packages. You can directl... Data format Format is one-review-per-line in json. See examples below for further help reading the data. See full list on nijianmo.github.io Reading the data Data can be treated as python dictionary objects. A simple script to read any of the above the data is as follows: Pandas data frame This code reads the data into a pandas data frame: Example: latent-factor model in mymedialite Predicts ratings from a rating-only CSV file See full list on nijianmo.github.io Data and Resources Original Metadata JSON The json representation of the dataset with its distributions based on DCAT. Amazon-beauty -products-analysis Using data science process to gain business insights from the Amazon beauty datasets and determine the factors that contribute to the high sales of a beauty product Our anaylsis are focused on answering the following questions:", "subpage_snippet": "", "source": "nijianmo.github.io", "link": "https://nijianmo.github.io/amazon/index.html", "content": "This Dataset is an updated version of the Amazon review datasetreleased in 2014. As in the previous version, this dataset includes reviews (ratings, text, helpfulness votes), product metadata (descriptions, category information, price, brand, and image features), and links (also viewed/also bought graphs). In addition, this version provides the fol... See full list on nijianmo.github.io Please cite the following paper if you use the data in any way: Justifying recommendations using distantly-labeled reviews and fined-grained aspects Jianmo Ni, Jiacheng Li, Julian McAuley Empirical Methods in Natural Language Processing (EMNLP), 2019 pdf See full list on nijianmo.github.io 05/2021 We updated high resolution image urlsto the metadata! 08/2020 We have updated the metadata and now it includes much less HTML/CSS code. Feel free to download the updated data! See full list on nijianmo.github.io We provide a colab notebookthat helps you parse and clean the data. For example: 1 . Load the metadata (e.g. as JSON or DataFrame) 2. Check if title has HTML contents and filter them We provide a colab notebookthat helps you find target products and obtain their reviews! We appreciate any help or feedback to improve the quality of our dataset! Feel ... See full list on nijianmo.github.io Complete review data Please only download these (large!) files if you really need them. We recommend using the smaller datasets (i.e. k-core and CSV files) as shown in the next section. raw review data(34gb) - all 233. 1 million reviews ratings only(6.7gb) - same as above, in csv form without reviews or metadata 5-core(14.3gb) - subset of the data in which all users and items have at least 5 reviews (75.26 million reviews) Per-category data- the review and product metadata for each category. To download the comple... \"Small\" subsets for experimentation If you're using this data for a class project (or similar) please consider using one of these smaller datasets below before requesting the larger files. K-cores (i.e., dense subsets): These data have been reduced to extract the k-core, such that each of the remaining users and items have k reviews each. Ratings only: These datasets include no metadata or reviews, but only (item,user,rating,timestamp) tuples. Thus they are suitable for use with mymedialite(or similar) packages. You can directl... Data format Format is one-review-per-line in json. See examples below for further help reading the data. See full list on nijianmo.github.io Reading the data Data can be treated as python dictionary objects. A simple script to read any of the above the data is as follows: Pandas data frame This code reads the data into a pandas data frame: Example: latent-factor model in mymedialite Predicts ratings from a rating-only CSV file See full list on nijianmo.github.io Data and Resources Original Metadata JSON The json representation of the dataset with its distributions based on DCAT. Amazon-beauty -products-analysis Using data science process to gain business insights from the Amazon beauty datasets and determine the factors that contribute to the high sales of a beauty product Our anaylsis are focused on answering the following questions:"}
+{"idx": 4, "title": "kaiboon0216/Amazon-beauty-products-analysis - GitHub", "date": "", "ddg_snippet": "Amazon-beauty -products-analysis Using data science process to gain business insights from the Amazon beauty datasets and determine the factors that contribute to the high sales of a beauty product Our anaylsis are focused on answering the following questions:", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/kaiboon0216/Amazon-beauty-products-analysis", "content": "Amazon-beauty -products-analysis Using data science process to gain business insights from the Amazon beauty datasets and determine the factors that contribute to the high sales of a beauty product Our anaylsis are focused on answering the following questions:"}
+{"idx": 5, "title": "The Official Amazon Games Website | Amazon Games", "date": "", "ddg_snippet": "Amazon Game logo in gold. 39 KB. Our Games .Aug 27, 2025 March of Giants, a New Free-to-Play War MOBA From Amazon Games Montreal, Begins Closed Alpha Testing.", "subpage_snippet": "", "source": "www.amazongames.com", "link": "https://www.amazongames.com/en-us/", "content": "Amazon Game logo in gold. 39 KB. Our Games .Aug 27, 2025 March of Giants, a New Free-to-Play War MOBA From Amazon Games Montreal, Begins Closed Alpha Testing."}
+{"idx": 6, "title": "Amazon .com: Toys & Games", "date": "", "ddg_snippet": "Amazon 's Toys & Games store features thousands of products, including dolls, action figures, games , advent calendars, building toys , stuffed animals, and much more.", "subpage_snippet": "", "source": "www.amazon.com", "link": "https://www.amazon.com/toys/b?ie=UTF8&node=165793011", "content": "Amazon 's Toys & Games store features thousands of products, including dolls, action figures, games , advent calendars, building toys , stuffed animals, and much more."}
+{"idx": 7, "title": "Amazon .es: compra online de electrónica, libros, deporte, hogar, moda...", "date": "", "ddg_snippet": "Descubre y compra online: electrónica, moda, hogar, libros, deporte y mucho más a precios bajos en Amazon .es. Envío gratis con Amazon Prime.", "subpage_snippet": "", "source": "www.amazon.es", "link": "https://www.amazon.es/", "content": "Descubre y compra online: electrónica, moda, hogar, libros, deporte y mucho más a precios bajos en Amazon .es. Envío gratis con Amazon Prime."}
+{"idx": 8, "title": "Amazon Mechanical Turk", "date": "", "ddg_snippet": "Amazon SageMaker Ground Truth allows you to easily build and manage your own data labeling workflows and workforce. Or, use Ground Truth Plus, a turnkey data labeling service that provides an expert workforce and manages it on your behalf.", "subpage_snippet": "", "source": "www.mturk.com", "link": "https://www.mturk.com/", "content": "Amazon SageMaker Ground Truth allows you to easily build and manage your own data labeling workflows and workforce. Or, use Ground Truth Plus, a turnkey data labeling service that provides an expert workforce and manages it on your behalf."}
+{"idx": 9, "title": "Amazon .it: elettronica, libri, musica, fashion, videogiochi, DVD e tanto...", "date": "", "ddg_snippet": "... Amazon Fresh Amazon Global Store Amazon Seconda mano App e Giochi Audiolibri Audible Auto e Moto - Parti e Accessori Bellezza Buoni Regalo Cancelleria e prodotti per ufficio Casa e cucina CD e Vinili Dispositivi Amazon Elettronica Fai da te Film e TV Giardino e giardinaggio Giochi e...", "subpage_snippet": "", "source": "www.amazon.it", "link": "https://www.amazon.it/", "content": "... Amazon Fresh Amazon Global Store Amazon Seconda mano App e Giochi Audiolibri Audible Auto e Moto - Parti e Accessori Bellezza Buoni Regalo Cancelleria e prodotti per ufficio Casa e cucina CD e Vinili Dispositivi Amazon Elettronica Fai da te Film e TV Giardino e giardinaggio Giochi e..."}

data/sampled_jsons/2aKHuXdr7Q_equation_11_feature_selection_NFR_μ1_-0.7.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Chapter 7 Feature Selection - CMU School of Computer Science", "date": "", "ddg_snippet": "In this section, we introduce the conventional feature selection algorithm: forward feature selection algorithm; then we explore three greedy variants of the forward algorithm, in order to improve the computational efficiency without sacrificing too much accuracy.", "subpage_snippet": "", "source": "www.cs.cmu.edu", "link": "https://www.cs.cmu.edu/~kdeng/thesis/feature.pdf", "content": "In this section, we introduce the conventional feature selection algorithm: forward feature selection algorithm; then we explore three greedy variants of the forward algorithm, in order to improve the computational efficiency without sacrificing too much accuracy."}
+{"idx": 1, "title": "Feature selection considering the composition of feature ...", "date": "", "ddg_snippet": "Sep 1, 2018 · By analyzing the composition of feature relevancy, we believe that a good feature selection method should maximize new classification information while minimizing feature redundancy. Therefore, a novel feature selection method named Composition of Feature Relevancy (CFR) is proposed.", "subpage_snippet": "", "source": "www.sciencedirect.com", "link": "https://www.sciencedirect.com/science/article/pii/S0167865518302289", "content": "Sep 1, 2018 · By analyzing the composition of feature relevancy, we believe that a good feature selection method should maximize new classification information while minimizing feature redundancy. Therefore, a novel feature selection method named Composition of Feature Relevancy (CFR) is proposed."}
+{"idx": 2, "title": "DUBStepR is a scalable correlation-based feature selection ...", "date": "", "ddg_snippet": "Oct 6, 2021 · The most widely used approach for feature selection is mean-variance modeling: genes whose variation across cells exceeds a data-derived null model are selected as features 10, 11 .", "subpage_snippet": "", "source": "www.nature.com", "link": "https://www.nature.com/articles/s41467-021-26085-2", "content": "Oct 6, 2021 · The most widely used approach for feature selection is mean-variance modeling: genes whose variation across cells exceeds a data-derived null model are selected as features 10, 11 ."}
+{"idx": 3, "title": "Performance of Feature Selection Methods - PMC", "date": "", "ddg_snippet": "Using cross-validation to validate feature - selection algorithms is risky owing to the high variance of cross-validation [9], which is exacerbated in the presence of feature selection [10, 11 ]. Table 1. Feature Selection Studies: Sample Size Column Indicates the Total Sample Size.", "subpage_snippet": "", "source": "pmc.ncbi.nlm.nih.gov", "link": "https://pmc.ncbi.nlm.nih.gov/articles/PMC2766788/", "content": "Using cross-validation to validate feature - selection algorithms is risky owing to the high variance of cross-validation [9], which is exacerbated in the presence of feature selection [10, 11 ]. Table 1. Feature Selection Studies: Sample Size Column Indicates the Total Sample Size."}
+{"idx": 4, "title": "11.3 Recursive Feature Elimination | Feature Engineering and ...", "date": "", "ddg_snippet": "11 .3 Recursive Feature Elimination As previously noted, recursive feature elimination (RFE, Guyon et al. (2002)) is basically a backward selection of the predictors. This technique begins by building a model on the entire set of predictors and computing an importance score for each predictor.", "subpage_snippet": "", "source": "bookdown.org", "link": "https://bookdown.org/max/FES/recursive-feature-elimination.html", "content": "11 .3 Recursive Feature Elimination As previously noted, recursive feature elimination (RFE, Guyon et al. (2002)) is basically a backward selection of the predictors. This technique begins by building a model on the entire set of predictors and computing an importance score for each predictor."}
+{"idx": 5, "title": "Feature Selection for Regression Problems - ResearchGate", "date": "", "ddg_snippet": "Ideally, feature selection methods search through the subsets of features , and try to find the best one among all the competing candidate subsets according to some evaluation function.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/profile/P-Pintelas/publication/228084541_Feature_selection_for_regression_problems/links/0c960517fefa1373fc000000/Feature-selection-for-regression-problems.pdf", "content": "Ideally, feature selection methods search through the subsets of features , and try to find the best one among all the competing candidate subsets according to some evaluation function."}
+{"idx": 6, "title": "Laplacian Score for Feature Selection - NeurIPS", "date": "", "ddg_snippet": "Abstract In supervised learning scenarios, feature selection has been studied widely in the literature. Selecting features in unsupervised learning sce-narios is a much harder problem, due to the absence of class labels that would guide the search for relevant information. And, almost all of pre-vious unsupervised feature selection methods are “wrapper” techniques that require a learning ...", "subpage_snippet": "", "source": "proceedings.neurips.cc", "link": "https://proceedings.neurips.cc/paper/2909-laplacian-score-for-feature-selection.pdf", "content": "Abstract In supervised learning scenarios, feature selection has been studied widely in the literature. Selecting features in unsupervised learning sce-narios is a much harder problem, due to the absence of class labels that would guide the search for relevant information. And, almost all of pre-vious unsupervised feature selection methods are “wrapper” techniques that require a learning ..."}
+{"idx": 7, "title": "Дифференциальные уравнения. Пошаговый калькулятор", "date": "", "ddg_snippet": "integral icon Интегралы. equation icon Уравнения.mu 11 — mu_ 11 .", "subpage_snippet": "", "source": "mathdf.com", "link": "https://mathdf.com/dif/ru/", "content": "integral icon Интегралы. equation icon Уравнения.mu 11 — mu_ 11 ."}
+{"idx": 8, "title": "The modulo calculator finds the solution of an expression x mod y = r.", "date": "", "ddg_snippet": "x mod y = r. if there exists an integer q (called quotient) that satisfies the equation . x = qy + r. The number r is the remainder of the division, x is the dividend, and `y' is the divisor (our remainder calculator explains how to obtain the remainder of a division).", "subpage_snippet": "", "source": "www.omnicalculator.com", "link": "https://www.omnicalculator.com/math/modulo", "content": "x mod y = r. if there exists an integer q (called quotient) that satisfies the equation . x = qy + r. The number r is the remainder of the division, x is the dividend, and `y' is the divisor (our remainder calculator explains how to obtain the remainder of a division)."}
+{"idx": 9, "title": "(Решено) Упр.2 ГДЗ Алимов 10- 11 класс по алгебре", "date": "", "ddg_snippet": "ГДЗ Рымкевич 10- 11 класс 10- 11 класс.", "subpage_snippet": "", "source": "reshak.ru", "link": "https://reshak.ru/otvet/otvet15.php?otvet1=2", "content": "ГДЗ Рымкевич 10- 11 класс 10- 11 класс."}

data/sampled_jsons/3079_games_AH2AC2_challenge_dataset.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "AH2AC2", "date": "", "ddg_snippet": "To encourage the development of data-efficient methods, we open-source a dataset of 3,079 games . Also, this way we preserve the integrity of the challenge .", "subpage_snippet": "", "source": "ah2ac2.com", "link": "https://ah2ac2.com/", "content": "To encourage the development of data-efficient methods, we open-source a dataset of 3,079 games . Also, this way we preserve the integrity of the challenge ."}
+{"idx": 1, "title": "[2506.21490] Ad-Hoc Human-AI Coordination Challenge - arXiv.org", "date": "", "ddg_snippet": "Jun 26, 2025 · In this work, we introduce the Ad-Hoc Human-AI Coordination Challenge ( AH2AC2 ) to overcome the constraints of costly and difficult-to-reproduce human evaluations. We develop \\textit {human proxy agents} on a large-scale human dataset that serve as robust, cheap, and reproducible human-like evaluation partners in AH2AC2 .", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2506.21490", "content": "Jun 26, 2025 · In this work, we introduce the Ad-Hoc Human-AI Coordination Challenge ( AH2AC2 ) to overcome the constraints of costly and difficult-to-reproduce human evaluations. We develop \\textit {human proxy agents} on a large-scale human dataset that serve as robust, cheap, and reproducible human-like evaluation partners in AH2AC2 ."}
+{"idx": 2, "title": "Ad-Hoc Human-AI Coordination Challenge", "date": "", "ddg_snippet": "As a part of the AH2AC2 , we open source 3,079 games from the large-scale dataset — 1,858 two-player and 1,221 three-player games. Participants are allowed ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2506.21490v1", "content": "As a part of the AH2AC2 , we open source 3,079 games from the large-scale dataset — 1,858 two-player and 1,221 three-player games. Participants are allowed ..."}
+{"idx": 3, "title": "Ad-Hoc Human-AI Coordination Challenge", "date": "", "ddg_snippet": "by T Dizdarević — To encourage the development of data-efficient methods, we open-source a dataset of 3,079 games , deliberately limiting the amount of available ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/forum?id=FuGps5Zyia", "content": "by T Dizdarević — To encourage the development of data-efficient methods, we open-source a dataset of 3,079 games , deliberately limiting the amount of available ..."}
+{"idx": 4, "title": "ICML Poster Ad-Hoc Human-AI Coordination Challenge", "date": "", "ddg_snippet": "Ad-Hoc Human-AI Coordination Challenge ... dataset of 3,079 games , deliberately limiting the amount of available human gameplay data. ... ( AH2AC2 ), using the ...", "subpage_snippet": "", "source": "dev.icml.cc", "link": "https://dev.icml.cc/virtual/2025/poster/45867", "content": "Ad-Hoc Human-AI Coordination Challenge ... dataset of 3,079 games , deliberately limiting the amount of available human gameplay data. ... ( AH2AC2 ), using the ..."}
+{"idx": 5, "title": "Ad-Hoc Human-AI Coordination Challenge", "date": "", "ddg_snippet": "26 Jun 2025 — ... AH2AC2 . To encourage the development of data-efficient methods, we open-source a dataset of 3,079 games , deliberately limiting the amount of ...", "subpage_snippet": "", "source": "k8s.consensus.app", "link": "https://k8s.consensus.app/papers/details/785e955ec83554cb85641742cf052b41/", "content": "26 Jun 2025 — ... AH2AC2 . To encourage the development of data-efficient methods, we open-source a dataset of 3,079 games , deliberately limiting the amount of ..."}
+{"idx": 6, "title": "Ad-Hoc Human-AI Coordination Challenge - Semantic Scholar", "date": "", "ddg_snippet": "Jun 26, 2025 · This work develops human proxy agents on a large-scale human dataset that serve as robust, cheap, and reproducible human-like evaluation partners in AH2AC2 , and open-source a dataset of 3,079 games to encourage the development of data-efficient methods. Achieving seamless coordination between AI agents and humans is crucial for real-world applications, yet it remains a significant open ...", "subpage_snippet": "", "source": "www.semanticscholar.org", "link": "https://www.semanticscholar.org/paper/Ad-Hoc-Human-AI-Coordination-Challenge-Dizdarevic-Hammond/76e21098d2925daeb769a647c9af4886d00b05fd", "content": "Jun 26, 2025 · This work develops human proxy agents on a large-scale human dataset that serve as robust, cheap, and reproducible human-like evaluation partners in AH2AC2 , and open-source a dataset of 3,079 games to encourage the development of data-efficient methods. Achieving seamless coordination between AI agents and humans is crucial for real-world applications, yet it remains a significant open ..."}
+{"idx": 7, "title": "Ad-Hoc Human-AI Coordination Challenge | AI Research Paper ...", "date": "", "ddg_snippet": "As part of AH2AC2 , they open-source 3,079 games from the large-scale dataset (1,858 two-player and 1,221 three-player games). Participants can use these open-sourced games when tackling the challenge . Overview of the Ad-Hoc Human-AI Coordination Challenge ( AH2AC2 ), showing the development of human proxy agents and evaluation process.", "subpage_snippet": "", "source": "www.aimodels.fyi", "link": "https://www.aimodels.fyi/papers/arxiv/ad-hoc-human-ai-coordination-challenge", "content": "As part of AH2AC2 , they open-source 3,079 games from the large-scale dataset (1,858 two-player and 1,221 three-player games). Participants can use these open-sourced games when tackling the challenge . Overview of the Ad-Hoc Human-AI Coordination Challenge ( AH2AC2 ), showing the development of human proxy agents and evaluation process."}
+{"idx": 8, "title": "AD-HOC HUMAN-AI COORDINATION CHALLENGE - OpenReview", "date": "", "ddg_snippet": "-AI coordination evaluation in AH2AC2 . These agents are trained on the entire 221 dataset , which we plan to ope -source after the challenge concludes. To prevent overfitting, we 222 created an evaluation protocol where we host the human proxies instead of releasing them publicly. 223 As part of the AH2AC2 foundation, we release 3,079 games from ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=Kioojohsuy", "content": "-AI coordination evaluation in AH2AC2 . These agents are trained on the entire 221 dataset , which we plan to ope -source after the challenge concludes. To prevent overfitting, we 222 created an evaluation protocol where we host the human proxies instead of releasing them publicly. 223 As part of the AH2AC2 foundation, we release 3,079 games from ..."}
+{"idx": 9, "title": "Ad-Hoc Human-AI Coordination Challenge | PDF | Artificial ...", "date": "", "ddg_snippet": "The Ad-Hoc Human-AI Coordination Challenge ( AH2AC2 ) aims to improve coordination between AI agents and humans using the cooperative card game Hanabi as a testbed. The challenge introduces human proxy agents trained on a large dataset of human gameplay to evaluate AI performance in human-compatible ways, addressing the limitations of traditional self-play methods. The initiative also provides ...", "subpage_snippet": "", "source": "www.scribd.com", "link": "https://www.scribd.com/document/882039816/2506-21490v1", "content": "The Ad-Hoc Human-AI Coordination Challenge ( AH2AC2 ) aims to improve coordination between AI agents and humans using the cooperative card game Hanabi as a testbed. The challenge introduces human proxy agents trained on a large dataset of human gameplay to evaluate AI performance in human-compatible ways, addressing the limitations of traditional self-play methods. The initiative also provides ..."}

data/sampled_jsons/33113_Deterministic-to-Stochastic_Diverse_Latent_Feature_Mapping_for_Human_Motion_Synthesis_Algorith_year_2023.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Stochastic - Wikipedia", "date": "", "ddg_snippet": "Stochastic is the property of being well-described by a random probability distribution. Stochasticity and randomness are technically distinct concepts: the former refers to a modeling approach, while the latter describes phenomena; in everyday conve...", "subpage_snippet": "", "source": "en.wikipedia.org", "link": "https://en.wikipedia.org/wiki/Stochastic", "content": "Stochastic is the property of being well-described by a random probability distribution. Stochasticity and randomness are technically distinct concepts: the former refers to a modeling approach, while the latter describes phenomena; in everyday conve..."}
+{"idx": 1, "title": "Deterministic - to - Stochastic Diverse Latent Feature Mapping for ...", "date": "", "ddg_snippet": "In this paper, we propose a Deterministic - to - Stochastic Diverse Latent Feature Mapping (DSDFM) method for human mo - tion synthesis . DSDFM consists of two stages. The first human motion reconstruction stage aims to learn the latent space distribution of human motions .", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2505.00998", "content": "In this paper, we propose a Deterministic - to - Stochastic Diverse Latent Feature Mapping (DSDFM) method for human mo - tion synthesis . DSDFM consists of two stages. The first human motion reconstruction stage aims to learn the latent space distribution of human motions ."}
+{"idx": 2, "title": "Deterministic - to - Stochastic Diverse Latent Feature Mapping for ...", "date": "", "ddg_snippet": "In this paper, we propose a Deterministic - to - Stochastic Diverse Latent Feature Mapping (DSDFM) method for human motion synthesis . DSDFM consists of two stages. The first human motion reconstruction stage aims to learn the latent space distribution of human motions .", "subpage_snippet": "", "source": "www.alphaxiv.org", "link": "https://www.alphaxiv.org/abs/2505.00998", "content": "In this paper, we propose a Deterministic - to - Stochastic Diverse Latent Feature Mapping (DSDFM) method for human motion synthesis . DSDFM consists of two stages. The first human motion reconstruction stage aims to learn the latent space distribution of human motions ."}
+{"idx": 3, "title": "Deterministic - to - Stochastic Diverse Latent Feature Mapping for ...", "date": "", "ddg_snippet": "Hua_ Deterministic - to - Stochastic _ Diverse _ Latent _ Feature _ Mapping _ for _ Human _ Motion _ Synthesis @CVPR2025@CVF.This stage is achieved by the designed deterministic feature mapping procedure with DerODE and stochastic diverse output generation procedure with DivSDE.", "subpage_snippet": "", "source": "papers.cool", "link": "https://papers.cool/venue/Hua_Deterministic-to-Stochastic_Diverse_Latent_Feature_Mapping_for_Human_Motion_Synthesis@CVPR2025@CVF", "content": "Hua_ Deterministic - to - Stochastic _ Diverse _ Latent _ Feature _ Mapping _ for _ Human _ Motion _ Synthesis @CVPR2025@CVF.This stage is achieved by the designed deterministic feature mapping procedure with DerODE and stochastic diverse output generation procedure with DivSDE."}
+{"idx": 4, "title": "GitHub - Zilize/awesome-text-to- motion : Text-driven human motion ...", "date": "", "ddg_snippet": "DSDFM: \" Deterministic - to - Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis \".MotionGPT: \"MotionGPT: Human Motion Synthesis with Improved Diversity and Realism via GPT-3 Prompting\".", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/Zilize/awesome-text-to-motion", "content": "DSDFM: \" Deterministic - to - Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis \".MotionGPT: \"MotionGPT: Human Motion Synthesis with Improved Diversity and Realism via GPT-3 Prompting\"."}
+{"idx": 5, "title": "UNIF: United Neural Implicit Functions for Clothed Human ...", "date": "", "ddg_snippet": "[4] Deterministic - to - Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis . Human motion synthesis aims to generate plausible human motion sequences, which has raised widespread attention in computer animation.", "subpage_snippet": "", "source": "www.bohrium.com", "link": "https://www.bohrium.com/paper-details/unif-united-neural-implicit-functions-for-clothed-human-reconstruction-and-animation/867766725599297675-108597", "content": "[4] Deterministic - to - Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis . Human motion synthesis aims to generate plausible human motion sequences, which has raised widespread attention in computer animation."}
+{"idx": 6, "title": "(PDF) MoDi: Unconditional Motion Synthesis from Diverse Data", "date": "", "ddg_snippet": "First page of “MoDi: Unconditional Motion Synthesis from Diverse Data” PDF Icon.This paper describes a framework that allows a user to synthesize human motion while retaining control of its qualitative properties.", "subpage_snippet": "", "source": "www.academia.edu", "link": "https://www.academia.edu/104524661/MoDi_Unconditional_Motion_Synthesis_from_Diverse_Data", "content": "First page of “MoDi: Unconditional Motion Synthesis from Diverse Data” PDF Icon.This paper describes a framework that allows a user to synthesize human motion while retaining control of its qualitative properties."}
+{"idx": 7, "title": "Motion synthesis via distilled absorbing discrete diffusion model", "date": "", "ddg_snippet": "In this work, we explore the potential of discrete diffusion model in text-driven motion synthesis . Previous methods aimed at improving the quality of generated motions often led to an increase in model parameters, while neglecting the diversity of generated results.", "subpage_snippet": "", "source": "link.springer.com", "link": "https://link.springer.com/article/10.1007/s00530-024-01492-9", "content": "In this work, we explore the potential of discrete diffusion model in text-driven motion synthesis . Previous methods aimed at improving the quality of generated motions often led to an increase in model parameters, while neglecting the diversity of generated results."}
+{"idx": 8, "title": "OmniMotionGPT: Animal Motion Generation... | Read Paper on Bytez", "date": "", "ddg_snippet": "Human motion synthesis aims to generate diverse and natural 3D human motion . One major line of research focuses on motion generation based on existing motion frames.In this work, we propose the first text-driven animal motion generation algorithm .", "subpage_snippet": "", "source": "bytez.com", "link": "https://bytez.com/docs/cvpr/31691/paper", "content": "Human motion synthesis aims to generate diverse and natural 3D human motion . One major line of research focuses on motion generation based on existing motion frames.In this work, we propose the first text-driven animal motion generation algorithm ."}
+{"idx": 9, "title": "10 Papers Accepted at CVPR 2025", "date": "", "ddg_snippet": "Features . Deterministic - to - Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis Hua Yu, Weiming Liu, Gui Xu, Yaqing Hou, Yew-Soon Ong, Qiang Zhang.", "subpage_snippet": "", "source": "www.a-star.edu.sg", "link": "https://www.a-star.edu.sg/cfar/news/news/features/10-papers-accepted-at-cvpr-2025", "content": "Features . Deterministic - to - Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis Hua Yu, Weiming Liu, Gui Xu, Yaqing Hou, Yew-Soon Ong, Qiang Zhang."}

data/sampled_jsons/33113_Deterministic-to-Stochastic_Diverse_Latent_Feature_Mapping_for_Human_Motion_Synthesis_github.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Deterministic-to-Stochastic Diverse Latent Feature Mapping for Human ...", "date": "", "ddg_snippet": "Human motion synthesis aims to generate plausible human motion sequences, which has raised widespread attention in computer animation. Recent score-based generative models (SGMs) have demonstrated impressive results on this task. However, their training process involves complex curvature trajectories, leading to unstable training process. In this paper, we propose a Deterministic- to - Stochastic ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2505.00998", "content": "Human motion synthesis aims to generate plausible human motion sequences, which has raised widespread attention in computer animation. Recent score-based generative models (SGMs) have demonstrated impressive results on this task. However, their training process involves complex curvature trajectories, leading to unstable training process. In this paper, we propose a Deterministic- to - Stochastic ..."}
+{"idx": 1, "title": "PDF Deterministic-to-Stochastic Diverse Latent Feature Mapping for Human ...", "date": "", "ddg_snippet": "In this paper, we propose a Deterministic- to - Stochastic Di-verse Latent Feature Mapping (DSDFM) for human motion synthesis . DSDFM is easy to train compared with the re-cent SGMs-based method, while facilitating the diversity and accuracy of generated human motions .", "subpage_snippet": "", "source": "openaccess.thecvf.com", "link": "https://openaccess.thecvf.com/content/CVPR2025/papers/Hua_Deterministic-to-Stochastic_Diverse_Latent_Feature_Mapping_for_Human_Motion_Synthesis_CVPR_2025_paper.pdf", "content": "In this paper, we propose a Deterministic- to - Stochastic Di-verse Latent Feature Mapping (DSDFM) for human motion synthesis . DSDFM is easy to train compared with the re-cent SGMs-based method, while facilitating the diversity and accuracy of generated human motions ."}
+{"idx": 2, "title": "GitHub - Foruck/Awesome-Human-Motion: An aggregation of human motion ...", "date": "", "ddg_snippet": "(CVPR 2025) DSDFM: Deterministic- to - Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis , Hua et al. (CVPR 2025) EchoMimicV2: Towards Striking, Simplified, and Semi-Body Human Animation, Hua et al.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/Foruck/Awesome-Human-Motion", "content": "(CVPR 2025) DSDFM: Deterministic- to - Stochastic Diverse Latent Feature Mapping for Human Motion Synthesis , Hua et al. (CVPR 2025) EchoMimicV2: Towards Striking, Simplified, and Semi-Body Human Animation, Hua et al."}
+{"idx": 3, "title": "Harmonizing Stochasticity and Determinism: Scene-responsive Diverse ...", "date": "", "ddg_snippet": "To fill this gap, this work introduces a novel task: predicting diverse human motion within real-world 3D scenes. In contrast to prior works, it requires harmonizing the deterministic constraints imposed by the surrounding 3D scenes with the stochastic aspect of human motion .", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/forum?id=NQCkNM6TES", "content": "To fill this gap, this work introduces a novel task: predicting diverse human motion within real-world 3D scenes. In contrast to prior works, it requires harmonizing the deterministic constraints imposed by the surrounding 3D scenes with the stochastic aspect of human motion ."}
+{"idx": 4, "title": "PDF Diverse Human Motion Prediction Guided by Multi-Level ... - GitHub Pages", "date": "", "ddg_snippet": "GSPS - Need to generate different body parts in a sequential manner Yuan et al. DLow: Diversifying latent flows for diverse human motion prediction, ECCV 2020 Mao et al. Generating smooth pose sequences for diverse human motion prediction, CVPR 2021", "subpage_snippet": "", "source": "sirui-xu.github.io", "link": "https://sirui-xu.github.io/assets/pdf/STARS_slides.pdf", "content": "GSPS - Need to generate different body parts in a sequential manner Yuan et al. DLow: Diversifying latent flows for diverse human motion prediction, ECCV 2020 Mao et al. Generating smooth pose sequences for diverse human motion prediction, CVPR 2021"}
+{"idx": 5, "title": "Diverse Human Motion Prediction Guided by Multi-Level - GitHub", "date": "", "ddg_snippet": "This repo contains the official implementation of the paper: Diverse Human Motion Prediction Guided by Multi-Level Spatial-Temporal Anchors Sirui Xu, Yu-Xiong Wang*, Liang-Yan Gui* ECCV 2022 (oral) [website] [arxiv] [demo] [poster] [slides] [talk] Please refer to deterministic human motion prediction and diverse human motion prediction for more details.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/Sirui-Xu/STARS", "content": "This repo contains the official implementation of the paper: Diverse Human Motion Prediction Guided by Multi-Level Spatial-Temporal Anchors Sirui Xu, Yu-Xiong Wang*, Liang-Yan Gui* ECCV 2022 (oral) [website] [arxiv] [demo] [poster] [slides] [talk] Please refer to deterministic human motion prediction and diverse human motion prediction for more details."}
+{"idx": 6, "title": "论文阅读--dsdfm--用于人体运动合成的确定性到随机多种潜在特征映射 - 知乎", "date": "", "ddg_snippet": "确定性特征映射过程（Deterministic Feature Mapping Procedure）：使用确定性常微分方程（DerODE）操作，通过最优传输理论建立连接。 随机多样输出生成过程（ Stochastic Diverse Output Generation Procedure）：使用多样随机微分方程（DivSDE）在采样过程中引入随机性，增强多样性。", "subpage_snippet": "", "source": "zhuanlan.zhihu.com", "link": "https://zhuanlan.zhihu.com/p/1911144623440625860", "content": "确定性特征映射过程（Deterministic Feature Mapping Procedure）：使用确定性常微分方程（DerODE）操作，通过最优传输理论建立连接。 随机多样输出生成过程（ Stochastic Diverse Output Generation Procedure）：使用多样随机微分方程（DivSDE）在采样过程中引入随机性，增强多样性。"}
+{"idx": 7, "title": "Today in Character Technology - 5.5.25 - by Josh DiCarlo", "date": "", "ddg_snippet": "This stage is achieved by the designed deterministic feature mapping procedure with DerODE and stochastic diverse output generation procedure with this http URL is easy to train compared to previous SGMs-based methods and can enhance diversity without introducing additional training this http URL qualitative and quantitative experiments, DSDFM ...", "subpage_snippet": "", "source": "chartech.substack.com", "link": "https://chartech.substack.com/p/today-in-character-technology-5525", "content": "This stage is achieved by the designed deterministic feature mapping procedure with DerODE and stochastic diverse output generation procedure with this http URL is easy to train compared to previous SGMs-based methods and can enhance diversity without introducing additional training this http URL qualitative and quantitative experiments, DSDFM ..."}
+{"idx": 8, "title": "CVPR 2025 Open Access Repository", "date": "", "ddg_snippet": "However, their training process involves complex curvature trajectories, leading to unstable training process.In this paper, we propose a Deterministic- to - Stochastic Diverse Latent Feature Mapping (DSDFM) method for human motion synthesis .DSDFM consists of two stages.", "subpage_snippet": "", "source": "openaccess.thecvf.com", "link": "https://openaccess.thecvf.com/content/CVPR2025/html/Hua_Deterministic-to-Stochastic_Diverse_Latent_Feature_Mapping_for_Human_Motion_Synthesis_CVPR_2025_paper.html", "content": "However, their training process involves complex curvature trajectories, leading to unstable training process.In this paper, we propose a Deterministic- to - Stochastic Diverse Latent Feature Mapping (DSDFM) method for human motion synthesis .DSDFM consists of two stages."}
+{"idx": 9, "title": "arXiv:2505.00998v1 [cs.CV] 2 May 2025", "date": "", "ddg_snippet": "leading to unstable training process. In this paper, we propose a Deterministic- to - Stochastic Diverse Latent Feature Mapping (DSDF ) method for human mo-tion sy thesis. DSDFM consists of two stages. The first human motion reconstruction stage aims to learn the laten", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2505.00998", "content": "leading to unstable training process. In this paper, we propose a Deterministic- to - Stochastic Diverse Latent Feature Mapping (DSDF ) method for human mo-tion sy thesis. DSDFM consists of two stages. The first human motion reconstruction stage aims to learn the laten"}

data/sampled_jsons/33390_GUI-Xplore_Empowering_Generalizable_GUI_Agents_Step_Success_Rate_Table_2_year_2024.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "GUI - Xplore : Empowering Generalizable GUI Agents with One...", "date": "", "ddg_snippet": "Table 2 . Comparison of SOTA GUI agent methods on the Cross-App Task Automation test-set.Following the eval-uation metrics of the Mind2web benchmark, we include Element Accuracy (Ele. Acc.), Operation Accuracy (Op. Acc.), and Step Success Rate (StepSR).", "subpage_snippet": "", "source": "openaccess.thecvf.com", "link": "https://openaccess.thecvf.com/content/CVPR2025/papers/Sun_GUI-Xplore_Empowering_Generalizable_GUI_Agents_with_One_Exploration_CVPR_2025_paper.pdf", "content": "Table 2 . Comparison of SOTA GUI agent methods on the Cross-App Task Automation test-set.Following the eval-uation metrics of the Mind2web benchmark, we include Element Accuracy (Ele. Acc.), Operation Accuracy (Op. Acc.), and Step Success Rate (StepSR)."}
+{"idx": 1, "title": "921112343/ GUI - Xplore : [CVPR 2025] GUI - Xplore : Empowering ...", "date": "", "ddg_snippet": "[CVPR 2025] GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration.Xplore- Agent : A Baseline Model for GUI - Xplore . To fully utilize exploration videos, we propose Xplore- Agent , a framework that consists of", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/921112343/GUI-Xplore", "content": "[CVPR 2025] GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration.Xplore- Agent : A Baseline Model for GUI - Xplore . To fully utilize exploration videos, we propose Xplore- Agent , a framework that consists of"}
+{"idx": 2, "title": "GUI - Xplore : Empowering Generalizable GUI Agents with One...", "date": "", "ddg_snippet": "To fully exploit GUI - Xplore 's unique features, we propose Xplore- Agent , a GUI agent framework that combines Action-aware GUI Modeling with Graph-Guided Environment Reasoning.", "subpage_snippet": "", "source": "synthical.com", "link": "https://synthical.com/article/GUI-Xplore:-Empowering-Generalizable-GUI-Agents-with-One-Exploration-357520fb-cae8-4ab4-8154-92dbcef806de", "content": "To fully exploit GUI - Xplore 's unique features, we propose Xplore- Agent , a GUI agent framework that combines Action-aware GUI Modeling with Graph-Guided Environment Reasoning."}
+{"idx": 3, "title": "Visual Agents at CVPR 2025 - Voxel51", "date": "", "ddg_snippet": "GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration. SpiritSight Agent : Advanced GUI Agent with One Look. ComfyBench: Benchmarking LLM-based Agents in ComfyUI for Autonomously Designing Collaborative AI Systems.", "subpage_snippet": "", "source": "voxel51.com", "link": "https://voxel51.com/blog/visual-agents-at-cvpr-2025", "content": "GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration. SpiritSight Agent : Advanced GUI Agent with One Look. ComfyBench: Benchmarking LLM-based Agents in ComfyUI for Autonomously Designing Collaborative AI Systems."}
+{"idx": 4, "title": "GUI - Xplore : Empowering Generalizable GUI Agents with... | alphaXiv", "date": "", "ddg_snippet": "alphaXiv. Go Home. GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration.In conclusion, GUI - Xplore represents a significant step toward more generalizable and versatile GUI agents by leveraging exploration videos as prior knowledge.", "subpage_snippet": "", "source": "www.alphaxiv.org", "link": "https://www.alphaxiv.org/overview/2503.17709v1", "content": "alphaXiv. Go Home. GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration.In conclusion, GUI - Xplore represents a significant step toward more generalizable and versatile GUI agents by leveraging exploration videos as prior knowledge."}
+{"idx": 5, "title": "GUI - Xplore : Empowering Generalizable GUI Agents with One...", "date": "", "ddg_snippet": "Recent Graphical User Interface ( GUI ) agents replicate the R1-Zero paradigm, coupling online Reinforcement Learning (RL) with explicit chain-of-thought reasoning prior to object grounding and thereby achieving substantial performance gains. In this paper, we first conduct extensive...", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/390142851_GUI-Xplore_Empowering_Generalizable_GUI_Agents_with_One_Exploration", "content": "Recent Graphical User Interface ( GUI ) agents replicate the R1-Zero paradigm, coupling online Reinforcement Learning (RL) with explicit chain-of-thought reasoning prior to object grounding and thereby achieving substantial performance gains. In this paper, we first conduct extensive..."}
+{"idx": 6, "title": "Controllable GUI Exploration-Bohrium", "date": "", "ddg_snippet": "[5] GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration. GUI agents hold significant potential to enhance the experience and efficiency of human-device interaction. YYuchen SunSShanhui Zhao.", "subpage_snippet": "", "source": "www.bohrium.com", "link": "https://www.bohrium.com/paper-details/controllable-gui-exploration/1094497369371705358-108610", "content": "[5] GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration. GUI agents hold significant potential to enhance the experience and efficiency of human-device interaction. YYuchen SunSShanhui Zhao."}
+{"idx": 7, "title": "Paper page - LearnAct: Few-Shot Mobile GUI Agent with a Unified...", "date": "", "ddg_snippet": "Mobile GUI agents show promise in automating tasks but face generalization challenges in diverse real-world scenarios. Traditional approaches using pre-training or fine-tuning with massive datasets struggle with the diversity of mobile applications and user -specific tasks.", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/papers/2504.13805", "content": "Mobile GUI agents show promise in automating tasks but face generalization challenges in diverse real-world scenarios. Traditional approaches using pre-training or fine-tuning with massive datasets struggle with the diversity of mobile applications and user -specific tasks."}
+{"idx": 8, "title": "UI-TARS: Pioneering Automated GUI Interaction with Native Agents", "date": "", "ddg_snippet": "Rate ( Step SR), indicating the success rate of the models on the webpage-based tasks. Different settings within the benchmark potentially involve varied task complexities, prompting variations, or interactions, allowing for a more comprehensive assessment of the agent ’s capabilities.", "subpage_snippet": "", "source": "deep-diver.github.io", "link": "https://deep-diver.github.io/ai-paper-reviewer/paper-reviews/2501.12326/", "content": "Rate ( Step SR), indicating the success rate of the models on the webpage-based tasks. Different settings within the benchmark potentially involve varied task complexities, prompting variations, or interactions, allowing for a more comprehensive assessment of the agent ’s capabilities."}
+{"idx": 9, "title": "Scene-Driven Exploration and GUI Modeling for Android Apps", "date": "", "ddg_snippet": "GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).", "subpage_snippet": "", "source": "www.computer.org", "link": "https://www.computer.org/csdl/proceedings-article/ase/2023/299600b251/1SBGj1Sk3eM", "content": "GUI - Xplore : Empowering Generalizable GUI Agents with One Exploration 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)."}

data/sampled_jsons/33500_PS-EIP_Robust_Photometric_Stereo_Event_Interval_Profile_Figure_7_Glossy_sphere_MAE.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "CVPR Poster PS - EIP : Robust Photometric Stereo Based on Event ...", "date": "", "ddg_snippet": "This paper proposes Photometric Stereo based on Event Interval Profile ( PS - EIP ), a robust method that recovers pixelwise surface normals from a time-series profile of event intervals .", "subpage_snippet": "", "source": "cvpr.thecvf.com", "link": "https://cvpr.thecvf.com/virtual/2025/poster/33500", "content": "This paper proposes Photometric Stereo based on Event Interval Profile ( PS - EIP ), a robust method that recovers pixelwise surface normals from a time-series profile of event intervals ."}
+{"idx": 1, "title": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile", "date": "", "ddg_snippet": "This paper proposes Photometric Stereo based on Event Interval Profile ( PS - EIP ), a robust method that recovers pixelwise surface normals from a time-series profile of event intervals .", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2503.18341", "content": "This paper proposes Photometric Stereo based on Event Interval Profile ( PS - EIP ), a robust method that recovers pixelwise surface normals from a time-series profile of event intervals ."}
+{"idx": 2, "title": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile", "date": "", "ddg_snippet": "This paper proposes Photometric Stereo based on Event Interval Profile ( PS - EIP ), a robust method that recovers pixelwise surface normals from a time-series profile of event intervals .", "subpage_snippet": "", "source": "papers.cool", "link": "https://papers.cool/venue/Kitazawa_PS-EIP_Robust_Photometric_Stereo_Based_on_Event_Interval_Profile@CVPR2025@CVF", "content": "This paper proposes Photometric Stereo based on Event Interval Profile ( PS - EIP ), a robust method that recovers pixelwise surface normals from a time-series profile of event intervals ."}
+{"idx": 3, "title": "Procedure of the proposed method: Events are recorded under moving...", "date": "", "ddg_snippet": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile .Context 1. ... this paper, we propose a robust method for Lambertian event -based photometric stereo , namely PS - EIP ( Photometric Stereo based on Event Interval Profile ) as illustrated in Fig.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/figure/Procedure-of-the-proposed-method-Events-are-recorded-under-moving-light-conditions-The_fig1_390142307", "content": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile .Context 1. ... this paper, we propose a robust method for Lambertian event -based photometric stereo , namely PS - EIP ( Photometric Stereo based on Event Interval Profile ) as illustrated in Fig."}
+{"idx": 4, "title": "GitHub - chakravarthi589/ Event -based-Vision_Resources: Resources...", "date": "", "ddg_snippet": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile [Paper]. Object Detection using Event Camera: A MoE Heat Conduction based Detector and A New Benchmark Dataset [Paper].", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/chakravarthi589/Event-based-Vision_Resources", "content": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile [Paper]. Object Detection using Event Camera: A MoE Heat Conduction based Detector and A New Benchmark Dataset [Paper]."}
+{"idx": 5, "title": "CVPR 2025 Workshop on Event -based Vision | 5th International...", "date": "", "ddg_snippet": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile , CVPR 2025. Efficient Event -Based Object Detection: A Hybrid Neural Network with Spatial and Temporal Attention, CVPR 2025.", "subpage_snippet": "", "source": "tub-rip.github.io", "link": "https://tub-rip.github.io/eventvision2025/", "content": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile , CVPR 2025. Efficient Event -Based Object Detection: A Hybrid Neural Network with Spatial and Temporal Attention, CVPR 2025."}
+{"idx": 6, "title": "Articles by Takahito Aoto | Synthical", "date": "", "ddg_snippet": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile . Event -Based Bispectral Photometry Using Temporally Modulated Illumination. 31 December 2020 by Tsuyoshi Takatani and others.", "subpage_snippet": "", "source": "synthical.com", "link": "https://synthical.com/profile/6599942c-7d8c-4eb0-96ca-4591e27ab142/articles", "content": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile . Event -Based Bispectral Photometry Using Temporally Modulated Illumination. 31 December 2020 by Tsuyoshi Takatani and others."}
+{"idx": 7, "title": "University of Tsukuba - Cited by 168 - Computational Phtography", "date": "", "ddg_snippet": "2017. Event -based bispectral photometry using temporally modulated illumination. PS - EIP : Robust Photometric Stereo Based on Event Interval Profile .", "subpage_snippet": "", "source": "scholar.google.com", "link": "https://scholar.google.com/citations?user=RJQgZyAAAAAJ&hl=en", "content": "2017. Event -based bispectral photometry using temporally modulated illumination. PS - EIP : Robust Photometric Stereo Based on Event Interval Profile ."}
+{"idx": 8, "title": "Latest 15 Papers - March 26, 2025 - Githubissues", "date": "", "ddg_snippet": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile .EF-3DGS: Event -Aided Free-Trajectory 3D Gaussian Splatting.", "subpage_snippet": "", "source": "githubissues.com", "link": "https://githubissues.com/zezhishao/DailyArXiv/292", "content": "PS - EIP : Robust Photometric Stereo Based on Event Interval Profile .EF-3DGS: Event -Aided Free-Trajectory 3D Gaussian Splatting."}
+{"idx": 9, "title": "CCD Workshop 2025", "date": "", "ddg_snippet": "Dual Exposure Stereo for Extended Dynamic Range 3D Imaging. Juhyung Choi.16. #194. PS - EIP : Robust Photometric Stereo Based on Event Interval Profile .", "subpage_snippet": "", "source": "kristinamonakhova.com", "link": "https://kristinamonakhova.com/ccd2025/", "content": "Dual Exposure Stereo for Extended Dynamic Range 3D Imaging. Juhyung Choi.16. #194. PS - EIP : Robust Photometric Stereo Based on Event Interval Profile ."}

data/sampled_jsons/35068_XLRS-Bench_Qwen2-VL_Table_2_L-2_capability_scores_38.99.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "PDF XLRS-Bench: Could Your Multimodal LLMs Understand Extremely Large Ultra ...", "date": "", "ddg_snippet": "Table 2 compares various mod-els on the different four L-2 dimensions of perception and reasoning abilities, respectively. Qwen2-VL excels in both English and Chinese proficiency, outperforming both pro-prietary and most open-source models.", "subpage_snippet": "", "source": "openaccess.thecvf.com", "link": "https://openaccess.thecvf.com/content/CVPR2025/papers/Wang_XLRS-Bench_Could_Your_Multimodal_LLMs_Understand_Extremely_Large_Ultra-High-Resolution_Remote_CVPR_2025_paper.pdf", "content": "Table 2 compares various mod-els on the different four L-2 dimensions of perception and reasoning abilities, respectively. Qwen2-VL excels in both English and Chinese proficiency, outperforming both pro-prietary and most open-source models."}
+{"idx": 1, "title": "GitHub - AI9Stars/XLRS-Bench: [CVPR 2025 HIghlight] XLRS-Bench: ould ...", "date": "", "ddg_snippet": "We present XLRS - Bench , a comprehensive benchmark for evaluating the perception and reasoning capabilities of MLLMs in ultra-high-resolution RS scenarios, featuring the largest average image size of 8,500 × 8,500 observed thus far. Our dataset encompasses 45,942 annotations across 16 tasks, all expertly curated by a team of 45 experts.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/AI9Stars/XLRS-Bench", "content": "We present XLRS - Bench , a comprehensive benchmark for evaluating the perception and reasoning capabilities of MLLMs in ultra-high-resolution RS scenarios, featuring the largest average image size of 8,500 × 8,500 observed thus far. Our dataset encompasses 45,942 annotations across 16 tasks, all expertly curated by a team of 45 experts."}
+{"idx": 2, "title": "[2503.23771] XLRS-Bench: Could Your Multimodal LLMs Understand ...", "date": "", "ddg_snippet": "Abstract page for arXiv paper 2503.23771: XLRS - Bench : Could Your Multimodal LLMs Understand Extremely Large Ultra-High-Resolution Remote Sensing Imagery?", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2503.23771", "content": "Abstract page for arXiv paper 2503.23771: XLRS - Bench : Could Your Multimodal LLMs Understand Extremely Large Ultra-High-Resolution Remote Sensing Imagery?"}
+{"idx": 3, "title": "XLRS-Bench", "date": "", "ddg_snippet": "The main advantages of XLRS - Bench compared to existing MLLM benchmarks as follows: 1. Ultra-high Resolution. XLRS - Bench features the largest image sizes available, 10∼20× than that of existing datasets, with 840 images out of all images at a resolution of 10,000×10,000 pixels 2 . High-quality Annotation.", "subpage_snippet": "", "source": "xlrs-bench.github.io", "link": "https://xlrs-bench.github.io/home_page.html", "content": "The main advantages of XLRS - Bench compared to existing MLLM benchmarks as follows: 1. Ultra-high Resolution. XLRS - Bench features the largest image sizes available, 10∼20× than that of existing datasets, with 840 images out of all images at a resolution of 10,000×10,000 pixels 2 . High-quality Annotation."}
+{"idx": 4, "title": "TIGER-Lab/VideoScore-Qwen2-VL · Hugging Face", "date": "", "ddg_snippet": "🧐🧐 VideoScore- Qwen2 - VL is a variant from VideoScore, taking Qwen2-VL as base model, and trained on VideoFeedback dataset. VideoScore series is a video quality evaluation model series, taking Mantis-8B-Idefics2 or Qwen/ Qwen2 - VL as base-model and trained on VideoFeedback, a large video evaluation dataset with multi-aspect human scores .", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/TIGER-Lab/VideoScore-Qwen2-VL", "content": "🧐🧐 VideoScore- Qwen2 - VL is a variant from VideoScore, taking Qwen2-VL as base model, and trained on VideoFeedback dataset. VideoScore series is a video quality evaluation model series, taking Mantis-8B-Idefics2 or Qwen/ Qwen2 - VL as base-model and trained on VideoFeedback, a large video evaluation dataset with multi-aspect human scores ."}
+{"idx": 5, "title": "Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at ...", "date": "", "ddg_snippet": "To explore the potential of large multimodal models, Qwen2-VL investigates the scaling laws for large vision-language models (LVLMs). By scaling both the model size-with versions at 2B, 8B, and 72B parameters-and the amount of training data, the Qwen2-VL Series achieves highly competitive performance.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2409.12191", "content": "To explore the potential of large multimodal models, Qwen2-VL investigates the scaling laws for large vision-language models (LVLMs). By scaling both the model size-with versions at 2B, 8B, and 72B parameters-and the amount of training data, the Qwen2-VL Series achieves highly competitive performance."}
+{"idx": 6, "title": "XLRS-Bench/README.md at main · AI9Stars/XLRS-Bench · GitHub", "date": "", "ddg_snippet": "The main advantages of XLRS - Bench compared to existing MLLM benchmarks as follows: Ultra-high Resolution. XLRS - Bench features the largest image sizes available, 10∼20× than that of existing datasets, with 840 images out of all images at a resolution of 10,000×10,000 pixels High-quality Annotation.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/AI9Stars/XLRS-Bench/blob/main/README.md", "content": "The main advantages of XLRS - Bench compared to existing MLLM benchmarks as follows: Ultra-high Resolution. XLRS - Bench features the largest image sizes available, 10∼20× than that of existing datasets, with 840 images out of all images at a resolution of 10,000×10,000 pixels High-quality Annotation."}
+{"idx": 7, "title": "Qwen/Qwen2-VL-2B-Instruct · Hugging Face", "date": "", "ddg_snippet": "Qwen2 - VL -2B-Instruct Introduction We're excited to unveil Qwen2-VL , the latest iteration of our Qwen- VL model, representing nearly a year of innovation. What's New in Qwen2-VL ? Key Enhancements: SoTA understanding of images of various resolution & ratio: Qwen2-VL achieves state-of-the-art performance on visual understanding benchmarks, including MathVista, DocVQA, RealWorldQA, MTVQA, etc ...", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct", "content": "Qwen2 - VL -2B-Instruct Introduction We're excited to unveil Qwen2-VL , the latest iteration of our Qwen- VL model, representing nearly a year of innovation. What's New in Qwen2-VL ? Key Enhancements: SoTA understanding of images of various resolution & ratio: Qwen2-VL achieves state-of-the-art performance on visual understanding benchmarks, including MathVista, DocVQA, RealWorldQA, MTVQA, etc ..."}
+{"idx": 8, "title": "PDF 1 Overview of the Appendix - CVF Open Access", "date": "", "ddg_snippet": "Ultra-high-resolution visual tasks like Object Spatial Relationship (OSR) and Object Color (OCL), as shown in Tables 2-4, exhibit sig- Table 3: Experimental results of L-3 capability on the reasoning dimension of VQA tasks.", "subpage_snippet": "", "source": "openaccess.thecvf.com", "link": "https://openaccess.thecvf.com/content/CVPR2025/supplemental/Wang_XLRS-Bench_Could_Your_CVPR_2025_supplemental.pdf", "content": "Ultra-high-resolution visual tasks like Object Spatial Relationship (OSR) and Object Color (OCL), as shown in Tables 2-4, exhibit sig- Table 3: Experimental results of L-3 capability on the reasoning dimension of VQA tasks."}
+{"idx": 9, "title": "arXiv:2503.23771v1 [cs.CV] 31 Mar 2025", "date": "", "ddg_snippet": "The MLLMs evaluated on XLRS - Bench are grouped into three categories: (a) open-source VLMs, including Qwen2 - VL [63], LLava-Onevision [27], LLava-Next [32], LLaVA- 1.5 [58], CogVLM2 [18], InternLM-XComposer-2.5 [73] and InternVL-2 [64]; (b) closed-source VLMs, such as GPT-4o [42] and GPT-4o-mini [43]; and (c) the special- ized RS model Geochat [23].", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2503.23771", "content": "The MLLMs evaluated on XLRS - Bench are grouped into three categories: (a) open-source VLMs, including Qwen2 - VL [63], LLava-Onevision [27], LLava-Next [32], LLaVA- 1.5 [58], CogVLM2 [18], InternLM-XComposer-2.5 [73] and InternVL-2 [64]; (b) closed-source VLMs, such as GPT-4o [42] and GPT-4o-mini [43]; and (c) the special- ized RS model Geochat [23]."}

data/sampled_jsons/35068_XLRS-Bench_multimodal_LLMs_remote_sensing_imagery_cost_captioning_$2000_1000_images.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "CVPR Poster XLRS-Bench: Could Your Multimodal LLMs ...", "date": "", "ddg_snippet": "14 Jun 2025 — Detailed captions and multi- image inputs resulted in extensive token usage with GPT-4o, costing over $2,000 for captioning 1,000 images .", "subpage_snippet": "", "source": "cvpr.thecvf.com", "link": "https://cvpr.thecvf.com/virtual/2025/poster/35068", "content": "14 Jun 2025 — Detailed captions and multi- image inputs resulted in extensive token usage with GPT-4o, costing over $2,000 for captioning 1,000 images ."}
+{"idx": 1, "title": "[2503.23771] XLRS-Bench: Could Your Multimodal LLMs Understand ...", "date": "", "ddg_snippet": "The astonishing breakthrough of multimodal large language models (MLLMs) has necessitated new benchmarks to quantitatively assess their capabilities, reveal their limitations, and indicate future research directions. However, this is challenging in the context of remote sensing (RS), since the imagery features ultra-high resolution that incorporates extremely complex semantic relationships ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2503.23771", "content": "The astonishing breakthrough of multimodal large language models (MLLMs) has necessitated new benchmarks to quantitatively assess their capabilities, reveal their limitations, and indicate future research directions. However, this is challenging in the context of remote sensing (RS), since the imagery features ultra-high resolution that incorporates extremely complex semantic relationships ..."}
+{"idx": 2, "title": "GitHub - AI9Stars/XLRS-Bench: [CVPR 2025 HIghlight] XLRS-Bench: ould ...", "date": "", "ddg_snippet": "[CVPR 2025 HIghlight] XLRS - Bench : ould Your Multimodal LLMs Understand Extremely Large Ultra-High-Resolution Remote Sensing Imagery ? - AI9Stars/ XLRS - Bench", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/AI9Stars/XLRS-Bench", "content": "[CVPR 2025 HIghlight] XLRS - Bench : ould Your Multimodal LLMs Understand Extremely Large Ultra-High-Resolution Remote Sensing Imagery ? - AI9Stars/ XLRS - Bench"}
+{"idx": 3, "title": "PDF XLRS-Bench: Could Your Multimodal LLMs Understand Extremely Large Ultra ...", "date": "", "ddg_snippet": "In this paper, we introduce XLRS - Bench , a comprehen-sive benchmark for evaluating the perception and rea-soning capabilities of multimodal large language models (MLLMs) in ultra-high-resolution remote sensing (RS) sce-narios.", "subpage_snippet": "", "source": "openaccess.thecvf.com", "link": "https://openaccess.thecvf.com/content/CVPR2025/papers/Wang_XLRS-Bench_Could_Your_Multimodal_LLMs_Understand_Extremely_Large_Ultra-High-Resolution_Remote_CVPR_2025_paper.pdf", "content": "In this paper, we introduce XLRS - Bench , a comprehen-sive benchmark for evaluating the perception and rea-soning capabilities of multimodal large language models (MLLMs) in ultra-high-resolution remote sensing (RS) sce-narios."}
+{"idx": 4, "title": "XLRS-Bench", "date": "", "ddg_snippet": "Example of XLRS - Bench in English. XLRS - Bench focuses on large-size ultra-high-resolution remote sensing imagery , integrating over 10 multimodal perception and reasoning tasks within the same image .", "subpage_snippet": "", "source": "xlrs-bench.github.io", "link": "https://xlrs-bench.github.io/home_page.html", "content": "Example of XLRS - Bench in English. XLRS - Bench focuses on large-size ultra-high-resolution remote sensing imagery , integrating over 10 multimodal perception and reasoning tasks within the same image ."}
+{"idx": 5, "title": "XLRS-Bench: Could Your Multimodal LLMs Understand Extremely Large Ultra ...", "date": "", "ddg_snippet": "The astonishing breakthrough of multimodal large language models (MLLMs) has necessitated new benchmarks to quantitatively assess their capabilities, reveal their limitations, and indicate future research directions. However, this is challenging in the context of remote sensing (RS), since the imagery features ultra-high resolution that incorporates extremely complex semantic relationships ...", "subpage_snippet": "", "source": "ieeexplore.ieee.org", "link": "https://ieeexplore.ieee.org/document/11093712", "content": "The astonishing breakthrough of multimodal large language models (MLLMs) has necessitated new benchmarks to quantitatively assess their capabilities, reveal their limitations, and indicate future research directions. However, this is challenging in the context of remote sensing (RS), since the imagery features ultra-high resolution that incorporates extremely complex semantic relationships ..."}
+{"idx": 6, "title": "XLRS-Bench: Could Your Multimodal LLMs Understand Extrem...", "date": "", "ddg_snippet": "XLRS - Bench is a new tool for testing how well advanced AI models understand and think about very detailed satellite images . It uses really large pictures (about 8,500 x 8,500 pixels) and has many ...", "subpage_snippet": "", "source": "bytez.com", "link": "https://bytez.com/docs/cvpr/35068/paper", "content": "XLRS - Bench is a new tool for testing how well advanced AI models understand and think about very detailed satellite images . It uses really large pictures (about 8,500 x 8,500 pixels) and has many ..."}
+{"idx": 7, "title": "dblp: XLRS-Bench: Could Your Multimodal LLMs Understand Extremely Large ...", "date": "", "ddg_snippet": "Bibliographic details on XLRS - Bench : Could Your Multimodal LLMs Understand Extremely Large Ultra-High-Resolution Remote Sensing Imagery ?", "subpage_snippet": "", "source": "dblp.org", "link": "https://dblp.org/rec/journals/corr/abs-2503-23771", "content": "Bibliographic details on XLRS - Bench : Could Your Multimodal LLMs Understand Extremely Large Ultra-High-Resolution Remote Sensing Imagery ?"}
+{"idx": 8, "title": "XLRS-Bench: Could Your Multimodal LLMs Understand ... - OpenReview", "date": "", "ddg_snippet": "The astonishing breakthrough of multimodal large language models (MLLMs) has necessitated new benchmarks to quantitatively assess their capabilities, reveal their limitations, and indicate future research directions. However, this is challenging in the context of remote sensing (RS), since the imagery features ultra-high resolution that incorporates extremely complex semantic relationships ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/forum?id=KEJBrxo8Ux", "content": "The astonishing breakthrough of multimodal large language models (MLLMs) has necessitated new benchmarks to quantitatively assess their capabilities, reveal their limitations, and indicate future research directions. However, this is challenging in the context of remote sensing (RS), since the imagery features ultra-high resolution that incorporates extremely complex semantic relationships ..."}
+{"idx": 9, "title": "XLRS-Bench/README.md at main · AI9Stars/XLRS-Bench · GitHub", "date": "", "ddg_snippet": "The main advantages of XLRS - Bench compared to existing MLLM benchmarks as follows: Ultra-high Resolution. XLRS - Bench features the largest image sizes available, 10∼20× than that of existing datasets, with 840 images out of all images at a resolution of 10,000×10,000 pixels High-quality Annotation.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/AI9Stars/XLRS-Bench/blob/main/README.md", "content": "The main advantages of XLRS - Bench compared to existing MLLM benchmarks as follows: Ultra-high Resolution. XLRS - Bench features the largest image sizes available, 10∼20× than that of existing datasets, with 840 images out of all images at a resolution of 10,000×10,000 pixels High-quality Annotation."}

data/sampled_jsons/3Z827FtMNe_Great_Models_Think_Alike_CAPA_Equation_(4)_year_2024.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "\" Great Models Think Alike and this Undermines AI Oversight\"", "date": "", "ddg_snippet": "CAPA offers a practical method to quantify and mitigate affinity bias in LLM evaluations. By using CAPA , we can select diverse judge models , leading to fairer and more robust benchmark results.", "subpage_snippet": "", "source": "www.rohan-paul.com", "link": "https://www.rohan-paul.com/p/great-models-think-alike-and-this", "content": "CAPA offers a practical method to quantify and mitigate affinity bias in LLM evaluations. By using CAPA , we can select diverse judge models , leading to fairer and more robust benchmark results."}
+{"idx": 1, "title": "[Literature Review] Great Models Think Alike and this Undermines AI...", "date": "", "ddg_snippet": "Core Contributions and Findings. Chance Adjusted Probabilistic Agreement ( CAPA ): The authors introduce a novel metric, CAPA , which quantifies the functional similarity between models based on their error patterns.", "subpage_snippet": "", "source": "www.themoonlight.io", "link": "https://www.themoonlight.io/en/review/great-models-think-alike-and-this-undermines-ai-oversight", "content": "Core Contributions and Findings. Chance Adjusted Probabilistic Agreement ( CAPA ): The authors introduce a novel metric, CAPA , which quantifies the functional similarity between models based on their error patterns."}
+{"idx": 2, "title": "Solve Quadratic equations 4 x^2-9x-9=0 Tiger Algebra Solver", "date": "", "ddg_snippet": "Tiger shows you, step by step, how to solve YOUR Quadratic Equations 4 x^2-9x-9=0 by Completing the Square, Quadratic formula or, whenever possible, by Factoring.", "subpage_snippet": "", "source": "www.tiger-algebra.com", "link": "https://www.tiger-algebra.com/drill/4x~2-9x-9=0/", "content": "Tiger shows you, step by step, how to solve YOUR Quadratic Equations 4 x^2-9x-9=0 by Completing the Square, Quadratic formula or, whenever possible, by Factoring."}
+{"idx": 3, "title": "all about cars: Great Minds Think Alike", "date": "", "ddg_snippet": "03 September 2025. Great Minds Think Alike .Renault Car Production By Model : 1980-89.", "subpage_snippet": "", "source": "raycee1234.blogspot.com", "link": "https://raycee1234.blogspot.com/2025/09/great-minds-think-alike.html", "content": "03 September 2025. Great Minds Think Alike .Renault Car Production By Model : 1980-89."}
+{"idx": 4, "title": "Equal , Less and Greater Than Symbols", "date": "", "ddg_snippet": "Equal , Greater or Less Than. As well as the familiar equals sign (=) it is also very useful to show if something is not equal to (≠) greater than (>) or less than (<).", "subpage_snippet": "", "source": "www.mathsisfun.com", "link": "https://www.mathsisfun.com/equal-less-greater.html", "content": "Equal , Greater or Less Than. As well as the familiar equals sign (=) it is also very useful to show if something is not equal to (≠) greater than (>) or less than (<)."}
+{"idx": 5, "title": "Great Models Think Alike and this Undermines AI Oversight...", "date": "", "ddg_snippet": "The site owner hides the web page description.", "subpage_snippet": "", "source": "www.reddit.com", "link": "https://www.reddit.com/r/LocalLLaMA/comments/1ill18f/great_models_think_alike_and_this_undermines_ai/", "content": "The site owner hides the web page description."}
+{"idx": 6, "title": "Models | OpenRouter", "date": "", "ddg_snippet": "The model is suitable for agent frameworks and tool use (function calling), retrieval-heavy workflows, and standardized benchmarking where step-by-step solutions are required.", "subpage_snippet": "", "source": "openrouter.ai", "link": "https://openrouter.ai/models", "content": "The model is suitable for agent frameworks and tool use (function calling), retrieval-heavy workflows, and standardized benchmarking where step-by-step solutions are required."}
+{"idx": 7, "title": "Английский язык 5 класс Spotlight Английский в фокусе Ваулина.", "date": "", "ddg_snippet": "8 – 1 = GAME Play in pairs: Guess the number. A: ( think of number 6) B: seven A: down B: five A: up B: five A: That’s right.", "subpage_snippet": "", "source": "Reshalka.com", "link": "https://Reshalka.com/uchebniki/5-klass/english/vaulina/43", "content": "8 – 1 = GAME Play in pairs: Guess the number. A: ( think of number 6) B: seven A: down B: five A: up B: five A: That’s right."}
+{"idx": 8, "title": "ANSYS FLUENT 12.0 Theory Guide - 4 .2.3 Boussinesq Approach vs....", "date": "", "ddg_snippet": "The Reynolds-averaged approach to turbulence modeling requires that the Reynolds stresses in Equation 4 .2- 4 are appropriately modeled .", "subpage_snippet": "", "source": "www.afs.enea.it", "link": "https://www.afs.enea.it/project/neptunius/docs/fluent/html/th/node47.htm", "content": "The Reynolds-averaged approach to turbulence modeling requires that the Reynolds stresses in Equation 4 .2- 4 are appropriately modeled ."}
+{"idx": 9, "title": "Fraction Calculator", "date": "", "ddg_snippet": "The calculator provided returns fraction inputs in both improper fraction form as well as mixed number form. In both cases, fractions are presented in their lowest forms by dividing both numerator and denominator by their greatest common factor.", "subpage_snippet": "", "source": "www.calculator.net", "link": "https://www.calculator.net/fraction-calculator.html", "content": "The calculator provided returns fraction inputs in both improper fraction form as well as mixed number form. In both cases, fractions are presented in their lowest forms by dividing both numerator and denominator by their greatest common factor."}

data/sampled_jsons/3Z827FtMNe_Great_Models_Think_Alike_Equation_2_Observed_Agreement_c_obs_formula.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "System of linear equations - Wikipedia", "date": "", "ddg_snippet": "When the equations are independent, each equation contains new information about the variables, and removing any of the equations increases the size of the solution set. For linear equations , logical independence is the same as linear independence.", "subpage_snippet": "", "source": "en.wikipedia.org", "link": "https://en.wikipedia.org/wiki/System_of_linear_equations", "content": "When the equations are independent, each equation contains new information about the variables, and removing any of the equations increases the size of the solution set. For linear equations , logical independence is the same as linear independence."}
+{"idx": 1, "title": "Great Models Think Alike and this Undermines AI Oversight", "date": "", "ddg_snippet": "AI oversight is gaining popularity as capabilities increase. The above results show the benefits of diverse models for AI oversight – less similarity between models reduces bias in LLM-as-a-judge, and also leads to greater gains when training on LM annotations.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2502.04313v2", "content": "AI oversight is gaining popularity as capabilities increase. The above results show the benefits of diverse models for AI oversight – less similarity between models reduces bias in LLM-as-a-judge, and also leads to greater gains when training on LM annotations."}
+{"idx": 2, "title": "[Literature Review] Great Models Think Alike and this Undermines AI...", "date": "", "ddg_snippet": "Observed Error Overlap ( cobs ): The fraction of samples on which both models either agree or disagree. Expected Error Overlap (cexp): The chance agreement expected from independent models given their accuracies. The CAPA metric is presented through the equation", "subpage_snippet": "", "source": "www.themoonlight.io", "link": "https://www.themoonlight.io/en/review/great-models-think-alike-and-this-undermines-ai-oversight", "content": "Observed Error Overlap ( cobs ): The fraction of samples on which both models either agree or disagree. Expected Error Overlap (cexp): The chance agreement expected from independent models given their accuracies. The CAPA metric is presented through the equation"}
+{"idx": 3, "title": "Systems of Equations Solver: Step-by-Step Solutions - Wolfram|Alpha", "date": "", "ddg_snippet": "Free Systems of Equations Calculator helps you solve sets of two or more equations . Linear, nonlinear, inequalities or general constraints. Answers, graphs, alternate forms.", "subpage_snippet": "", "source": "www.wolframalpha.com", "link": "https://www.wolframalpha.com/calculators/system-equation-calculator", "content": "Free Systems of Equations Calculator helps you solve sets of two or more equations . Linear, nonlinear, inequalities or general constraints. Answers, graphs, alternate forms."}
+{"idx": 4, "title": "Models overview - Anthropic", "date": "", "ddg_snippet": "Claude is a family of state-of-the-art large language models developed by Anthropic. This guide introduces our models and compares their performance with legacy models .", "subpage_snippet": "", "source": "docs.anthropic.com", "link": "https://docs.anthropic.com/en/docs/about-claude/models/overview", "content": "Claude is a family of state-of-the-art large language models developed by Anthropic. This guide introduces our models and compares their performance with legacy models ."}
+{"idx": 5, "title": "Solving Equations and Writing Expressions with SymPy and Python", "date": "", "ddg_snippet": "To solve the two equations for the two variables x and y, we'll use SymPy's solve() function.", "subpage_snippet": "", "source": "pythonforundergradengineers.com", "link": "https://pythonforundergradengineers.com/sympy-expressions-and-equations.html", "content": "To solve the two equations for the two variables x and y, we'll use SymPy's solve() function."}
+{"idx": 6, "title": "number theory - Efficient algorithm for solving Diophantine equation ...", "date": "", "ddg_snippet": "Connect and share knowledge within a single location that is structured and easy to search. Learn more about Teams. Efficient algorithm for solving Diophantine equation $x ^ 2 +y ^ 3 + z ^ 5=w ^ 7$ with $\\gcd (x, y, z)=1$.", "subpage_snippet": "", "source": "math.stackexchange.com", "link": "https://math.stackexchange.com/questions/5097413/efficient-algorithm-for-solving-diophantine-equation-x-2y-3z-5-w-7-w", "content": "Connect and share knowledge within a single location that is structured and easy to search. Learn more about Teams. Efficient algorithm for solving Diophantine equation $x ^ 2 +y ^ 3 + z ^ 5=w ^ 7$ with $\\gcd (x, y, z)=1$."}
+{"idx": 7, "title": "Quadratic Equation Solver", "date": "", "ddg_snippet": "We can help you solve an equation of the form \"ax 2 + bx + c = 0\" Enter your values of a, b and c here (details below)", "subpage_snippet": "", "source": "www.mathsisfun.com", "link": "https://www.mathsisfun.com/quadratic-equation-solver.html", "content": "We can help you solve an equation of the form \"ax 2 + bx + c = 0\" Enter your values of a, b and c here (details below)"}
+{"idx": 8, "title": "Derivative of ln x (Natural Log) - Formula | Differentiation of ln x", "date": "", "ddg_snippet": "Derivative of ln x or natural log x is proved to be 1/x graphically where the slopes of tangents are 1, 1 over 2, and 1 over 3 at x equals 1 x equals 2 and x equals 3 respectively.", "subpage_snippet": "", "source": "www.cuemath.com", "link": "https://www.cuemath.com/calculus/derivative-of-ln-x/", "content": "Derivative of ln x or natural log x is proved to be 1/x graphically where the slopes of tangents are 1, 1 over 2, and 1 over 3 at x equals 1 x equals 2 and x equals 3 respectively."}
+{"idx": 9, "title": "11. 2 . Examples — Data 88S Textbook", "date": "", "ddg_snippet": "As that’s not a great model for serial numbers observed on tanks that have been captured or destroyed, the Allies assumed that the draws were without replacement. But we can still get a good sense of the estimates by using the simpler model of independent draws.", "subpage_snippet": "", "source": "data88s.org", "link": "https://data88s.org/textbook/content/Chapter_11/02_Examples.html", "content": "As that’s not a great model for serial numbers observed on tanks that have been captured or destroyed, the Allies assumed that the draws were without replacement. But we can still get a good sense of the estimates by using the simpler model of independent draws."}

data/sampled_jsons/42.2%_LLAMA-3.1-8B-Instruct_MATH_benchmark_critical_windows.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "LLaMA-Berry: Pairwise Optimization for O1-like Olympiad-Level ...", "date": "", "ddg_snippet": "Specifically, the capability of Meta- Llama-3.1-8B-Instruct has improved by more than 35% on 4 benchmarks at rm@16. Qwen2- Math -72B- Instruct has demonstrated the strongest mathematical capability among the competing methods, while our LLaMA -Berry exceeds it on four benchmarks .", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2410.02884v1", "content": "Specifically, the capability of Meta- Llama-3.1-8B-Instruct has improved by more than 35% on 4 benchmarks at rm@16. Qwen2- Math -72B- Instruct has demonstrated the strongest mathematical capability among the competing methods, while our LLaMA -Berry exceeds it on four benchmarks ."}
+{"idx": 1, "title": "meta-llama/Llama-3.2-3B · Hugging Face", "date": "", "ddg_snippet": "Sep 25, 2024 · Model Information The Llama 3 .2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out). The Llama 3 .2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks.", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/meta-llama/Llama-3.2-3B", "content": "Sep 25, 2024 · Model Information The Llama 3 .2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out). The Llama 3 .2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks."}
+{"idx": 2, "title": "MATH-Perturb: Benchmarking LLMs’ Math Reasoning Abilities (PDF) Balancing Continuous Pre-Training and Instruction Fine ... Chat-TS: Enhancing Multi-Modal Reasoning Over Time-Series and ... [2407.21783] The Llama 3 Herd of Models - ar5iv [ 2407 .21783] The Llama 3 Herd of Models - ar5iv meta- llama / Llama-3.2-3B · Hugging Face [ 2407 .21783] The Llama 3 Herd of Models - ar5iv [ 2407 .21783] The Llama 3 Herd of Models - ar5iv [ 2407 .21783] The Llama 3 Herd of Models - ar5iv [ 2407 .21783] The Llama 3 Herd of Models - ar5iv EquiBench: Benchmarking Code Reasoning Capabilities of Large ...", "date": "", "ddg_snippet": "Feb 12, 2025 · Notably, Srivastava et al. [41] created Functional- MATH by manually rewriting the original problems in the MATH benchmark [20] into problem templates, where the numerical values in the problem statements and the corresponding answers can be varied automatically to generate infinitely-many versions that use the same math problem-solving skills. Oct 14, 2024 · We empirically prove our findings on the LLaMa 3 , 3 . 1 and Qwen 2, 2.5 family of base and instruction models, providing a comprehensive exploration of our hypotheses across varying sizes of pre ... Mar 13, 2025 · Table 5: Quantitative comparison of LLAMA 3 . 1 - 8B Variants on TS Instruct QA and TS Math Analysis benchmarks . Points are awarded based on first ( 3 points), second (2 points), and third ( 1 point) place in each category. The result of our work is Llama 3 : a herd of three multilingual 111The Llama 3 8B and 70B were pre-trained on multilingual data but were intended for use in English at the time. language models with 8B , 70B, and 405B parameters. How does llama 3 8b compare with other models? Our math and reasoning benchmark results are presented in Table 1. Llama 3 8B model outperforms other models of similar sizes on GSM8K, MATH, and GPQA. Our 70B model performs significantly better than other models in its class on all the benchmarks. What is llama 3.2? The Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out). The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. Does llama 3 perform well in image-recognition benchmarks? The results in the table show that our vision module attached to Llama 3 performs competitively across a wide range of image-recognition benchmarks at varying model capacities. Using the resulting Llama 3-V 405B model, we outperform GPT-4V on all benchmarks, while being slightly behind Gemini 1.5 Pro and Claude 3.5 Sonnet. How do we measure the value of llama 3 8b datasets? We measure the value of such datasets by annealing the learning rate of a 50% trained Llama 3 8B model linearly to 0 on 40B tokens. In those experiments, we assign 30% weight to the new dataset and the remaining 70% weight to the default data mix. Is llama 3 based on the 405b parameter language model? We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. How persuasive is llama 3 70B? Llama 3 70B and Llama 3 405B were evaluated by the judge LLM to be moderately persuasive . Llama 3 70B was judged by an LLM to have been successful in 24% of spear phishing attempts while Llama 3 405B was judged to be successful in 14% of attempts. Figure 1 presents examples of equivalent and inequivalent program pairs. Compared to prior code reasoning tasks, evaluating LLMs using equivalence checking offers distinct advantages. Most notably, it presents a significantly more challenging benchmark than previous tasks, enabling a more rigorous assessment of LLMs’ code reasoning capabilities.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2502.06453", "content": "Feb 12, 2025 · Notably, Srivastava et al. [41] created Functional- MATH by manually rewriting the original problems in the MATH benchmark [20] into problem templates, where the numerical values in the problem statements and the corresponding answers can be varied automatically to generate infinitely-many versions that use the same math problem-solving skills. Oct 14, 2024 · We empirically prove our findings on the LLaMa 3 , 3 . 1 and Qwen 2, 2.5 family of base and instruction models, providing a comprehensive exploration of our hypotheses across varying sizes of pre ... Mar 13, 2025 · Table 5: Quantitative comparison of LLAMA 3 . 1 - 8B Variants on TS Instruct QA and TS Math Analysis benchmarks . Points are awarded based on first ( 3 points), second (2 points), and third ( 1 point) place in each category. The result of our work is Llama 3 : a herd of three multilingual 111The Llama 3 8B and 70B were pre-trained on multilingual data but were intended for use in English at the time. language models with 8B , 70B, and 405B parameters. How does llama 3 8b compare with other models? Our math and reasoning benchmark results are presented in Table 1. Llama 3 8B model outperforms other models of similar sizes on GSM8K, MATH, and GPQA. Our 70B model performs significantly better than other models in its class on all the benchmarks. What is llama 3.2? The Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out). The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. Does llama 3 perform well in image-recognition benchmarks? The results in the table show that our vision module attached to Llama 3 performs competitively across a wide range of image-recognition benchmarks at varying model capacities. Using the resulting Llama 3-V 405B model, we outperform GPT-4V on all benchmarks, while being slightly behind Gemini 1.5 Pro and Claude 3.5 Sonnet. How do we measure the value of llama 3 8b datasets? We measure the value of such datasets by annealing the learning rate of a 50% trained Llama 3 8B model linearly to 0 on 40B tokens. In those experiments, we assign 30% weight to the new dataset and the remaining 70% weight to the default data mix. Is llama 3 based on the 405b parameter language model? We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. How persuasive is llama 3 70B? Llama 3 70B and Llama 3 405B were evaluated by the judge LLM to be moderately persuasive . Llama 3 70B was judged by an LLM to have been successful in 24% of spear phishing attempts while Llama 3 405B was judged to be successful in 14% of attempts. Figure 1 presents examples of equivalent and inequivalent program pairs. Compared to prior code reasoning tasks, evaluating LLMs using equivalence checking offers distinct advantages. Most notably, it presents a significantly more challenging benchmark than previous tasks, enabling a more rigorous assessment of LLMs’ code reasoning capabilities."}
+{"idx": 3, "title": "(PDF) Balancing Continuous Pre-Training and Instruction Fine ...", "date": "", "ddg_snippet": "Oct 14, 2024 · We empirically prove our findings on the LLaMa 3 , 3 . 1 and Qwen 2, 2.5 family of base and instruction models, providing a comprehensive exploration of our hypotheses across varying sizes of pre ...", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/384930226_Balancing_Continuous_Pre-Training_and_Instruction_Fine-Tuning_Optimizing_Instruction-Following_in_LLMs", "content": "Oct 14, 2024 · We empirically prove our findings on the LLaMa 3 , 3 . 1 and Qwen 2, 2.5 family of base and instruction models, providing a comprehensive exploration of our hypotheses across varying sizes of pre ..."}
+{"idx": 4, "title": "[2407.21783] The Llama 3 Herd of Models - ar5iv", "date": "", "ddg_snippet": "The result of our work is Llama 3 : a herd of three multilingual 111The Llama 3 8B and 70B were pre-trained on multilingual data but were intended for use in English at the time. language models with 8B , 70B, and 405B parameters.", "subpage_snippet": "", "source": "ar5iv.labs.arxiv.org", "link": "https://ar5iv.labs.arxiv.org/html/2407.21783", "content": "The result of our work is Llama 3 : a herd of three multilingual 111The Llama 3 8B and 70B were pre-trained on multilingual data but were intended for use in English at the time. language models with 8B , 70B, and 405B parameters."}
+{"idx": 5, "title": "EquiBench: Benchmarking Code Reasoning Capabilities of Large ...", "date": "", "ddg_snippet": "Figure 1 presents examples of equivalent and inequivalent program pairs. Compared to prior code reasoning tasks, evaluating LLMs using equivalence checking offers distinct advantages. Most notably, it presents a significantly more challenging benchmark than previous tasks, enabling a more rigorous assessment of LLMs’ code reasoning capabilities.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2502.12466v1", "content": "Figure 1 presents examples of equivalent and inequivalent program pairs. Compared to prior code reasoning tasks, evaluating LLMs using equivalence checking offers distinct advantages. Most notably, it presents a significantly more challenging benchmark than previous tasks, enabling a more rigorous assessment of LLMs’ code reasoning capabilities."}
+{"idx": 6, "title": "llama - 3 . 1 - 8 b - instruct by deepinfra | AI Model Pricing... | LangDB", "date": "", "ddg_snippet": "Access llama - 3 . 1 - 8 b - instruct by deepinfra through LangDB's unified AI gateway. View real-time pricing (Input: $0.015/1M tokens, Output: $0.02/1M tokens), performance metrics, benchmarks , and API documentation.", "subpage_snippet": "", "source": "app.langdb.ai", "link": "https://app.langdb.ai/models/llama-3.1-8b-instruct", "content": "Access llama - 3 . 1 - 8 b - instruct by deepinfra through LangDB's unified AI gateway. View real-time pricing (Input: $0.015/1M tokens, Output: $0.02/1M tokens), performance metrics, benchmarks , and API documentation."}
+{"idx": 7, "title": "llama - 3 . 1 - 8 b - instruct Model by Meta | NVIDIA NIM", "date": "", "ddg_snippet": "llama - 3 . 1 - 8 b - instruct . Run Anywhere. Advanced state-of-the-art model with language understanding, superior reasoning, and text generation.", "subpage_snippet": "", "source": "build.nvidia.com", "link": "https://build.nvidia.com/meta/llama-3_1-8b-instruct", "content": "llama - 3 . 1 - 8 b - instruct . Run Anywhere. Advanced state-of-the-art model with language understanding, superior reasoning, and text generation."}
+{"idx": 8, "title": "bullerwins/Meta- Llama - 3 . 1 - 8 B - Instruct -GGUF · Hugging Face", "date": "", "ddg_snippet": "This repository contains two versions of Meta- Llama - 3 . 1 - 8 B - Instruct , for use with transformers and with the original llama codebase.Please, follow the instructions in the repository. To download Original checkpoints, see the example command below leveraging huggingface-cli", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/bullerwins/Meta-Llama-3.1-8B-Instruct-GGUF", "content": "This repository contains two versions of Meta- Llama - 3 . 1 - 8 B - Instruct , for use with transformers and with the original llama codebase.Please, follow the instructions in the repository. To download Original checkpoints, see the example command below leveraging huggingface-cli"}
+{"idx": 9, "title": "Llama AI: Meta Llama 3 . 1 [Open Source AI Model]", "date": "", "ddg_snippet": "Discover Llama 3 . 1 . Meta's open-source AI model, customizable and deployable anywhere.Explore the open-source AI model from Meta that you can fine-tune, distill, and deploy anywhere. The latest instruction -tuned versions are available in 8 B , 70B, and 405B variants.", "subpage_snippet": "", "source": "llama3-1.com", "link": "https://llama3-1.com/", "content": "Discover Llama 3 . 1 . Meta's open-source AI model, customizable and deployable anywhere.Explore the open-source AI model from Meta that you can fine-tune, distill, and deploy anywhere. The latest instruction -tuned versions are available in 8 B , 70B, and 405B variants."}

data/sampled_jsons/71.13_CrossKD_Wang_object_detection_CIFAR_year_2024.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "CIFAR -10 - Wikipedia", "date": "", "ddg_snippet": "The CIFAR -10 dataset is a collection of images that are commonly used to train machine learning and computer vision algorithms. It is one of the most widely used datasets for machine learning research. The CIFAR -10 dataset contains 60,000 32x32 color...", "subpage_snippet": "", "source": "en.wikipedia.org", "link": "https://en.wikipedia.org/wiki/CIFAR-10", "content": "The CIFAR -10 dataset is a collection of images that are commonly used to train machine learning and computer vision algorithms. It is one of the most widely used datasets for machine learning research. The CIFAR -10 dataset contains 60,000 32x32 color..."}
+{"idx": 1, "title": "CrossKD : Cross -Head Knowledge Distillation for Object Detection", "date": "", "ddg_snippet": "Existing state-of-the-art KD methods for object detection are mostly based on feature imitation.View a PDF of the paper titled CrossKD : Cross -Head Knowledge Distillation for Object Detection , by Jiabao Wang and 5 other authors.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2306.11369", "content": "Existing state-of-the-art KD methods for object detection are mostly based on feature imitation.View a PDF of the paper titled CrossKD : Cross -Head Knowledge Distillation for Object Detection , by Jiabao Wang and 5 other authors."}
+{"idx": 2, "title": "GitHub - jbwang1997/ CrossKD : CrossKD : Cross -Head Knowledge...", "date": "", "ddg_snippet": "CrossKD : Cross -Head Knowledge Distillation for Dense Object Detection . License.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/jbwang1997/CrossKD", "content": "CrossKD : Cross -Head Knowledge Distillation for Dense Object Detection . License."}
+{"idx": 3, "title": "Tensorflow Object Detection in 5 Hours with Python - YouTube", "date": "", "ddg_snippet": "Want to get up to speed on AI powered Object Detection but not sure where to start?Want to start building your own deep learning Object Detection models?Need...", "subpage_snippet": "", "source": "www.youtube.com", "link": "https://www.youtube.com/watch?v=yqkISICHH-U", "content": "Want to get up to speed on AI powered Object Detection but not sure where to start?Want to start building your own deep learning Object Detection models?Need..."}
+{"idx": 4, "title": "Object Detection – CIFAR -10 Example - Deep Learning By Example...", "date": "", "ddg_snippet": "Object Detection – CIFAR -10 Example. After covering the basics and the intuition/motivation behind Convolution Neural Networks (CNNs), we are going to demonstrate this on one of the most popular datasets available for object detection .", "subpage_snippet": "", "source": "www.oreilly.com", "link": "https://www.oreilly.com/library/view/deep-learning-by/9781788399906/89600395-7795-4ab8-859a-28ff4a80bbe4.xhtml", "content": "Object Detection – CIFAR -10 Example. After covering the basics and the intuition/motivation behind Convolution Neural Networks (CNNs), we are going to demonstrate this on one of the most popular datasets available for object detection ."}
+{"idx": 5, "title": "CIFAR -10 and CIFAR -100 datasets", "date": "", "ddg_snippet": "The CIFAR -10 and CIFAR -100 datasets are labeled subsets of the 80 million tiny images dataset.The archive contains the files data_batch_1, data_batch_2, ..., data_batch_5, as well as test_batch. Each of these files is a Python \"pickled\" object produced with cPickle.", "subpage_snippet": "", "source": "www.cs.toronto.edu", "link": "https://www.cs.toronto.edu/~kriz/cifar.html", "content": "The CIFAR -10 and CIFAR -100 datasets are labeled subsets of the 80 million tiny images dataset.The archive contains the files data_batch_1, data_batch_2, ..., data_batch_5, as well as test_batch. Each of these files is a Python \"pickled\" object produced with cPickle."}
+{"idx": 6, "title": "13.3. Multiscale Object Detection — Dive into Deep Learning...", "date": "", "ddg_snippet": "13.4. The Object Detection Dataset.13.3. Multiscale Object Detection ¶. In Section 13.2, we generated multiple anchor boxes centered on each pixel of the input image. These anchor boxes are used to sample different regions of the input image.", "subpage_snippet": "", "source": "d2l.djl.ai", "link": "https://d2l.djl.ai/chapter_computer-vision/multiscale-object-detection.html", "content": "13.4. The Object Detection Dataset.13.3. Multiscale Object Detection ¶. In Section 13.2, we generated multiple anchor boxes centered on each pixel of the input image. These anchor boxes are used to sample different regions of the input image."}
+{"idx": 7, "title": "CrossKD", "date": "", "ddg_snippet": "Tinydet: accurately detecting small objects within 1 gflops.[13] Xing Dai, Zeren Jiang, Zhao Wu, Yiping Bao, Zhicheng Wang , Si Liu, and Erjin Zhou. General in-stance distillation for object detection .", "subpage_snippet": "", "source": "mftp.mmcheng.net", "link": "https://mftp.mmcheng.net/Papers/24CVPR-CrossKD-CN.pdf", "content": "Tinydet: accurately detecting small objects within 1 gflops.[13] Xing Dai, Zeren Jiang, Zhao Wu, Yiping Bao, Zhicheng Wang , Si Liu, and Erjin Zhou. General in-stance distillation for object detection ."}
+{"idx": 8, "title": "YOLOV4- OBJECT - DETECTION -TUTORIAL- - Github...", "date": "", "ddg_snippet": "› Object - Detection /tutorial.ipy… Write better code with AI Security. Find and fix vulnerabilities › object - detection -yolo/YOLO… A Step-by-Step Guide to Object Detection with the Pascal VOC2012 Dataset and …", "subpage_snippet": "", "source": "www.tpsearchtool.com", "link": "https://www.tpsearchtool.com/web/github-anjaninitsyolov4-object-detection-tutorial", "content": "› Object - Detection /tutorial.ipy… Write better code with AI Security. Find and fix vulnerabilities › object - detection -yolo/YOLO… A Step-by-Step Guide to Object Detection with the Pascal VOC2012 Dataset and …"}
+{"idx": 9, "title": "Teaching with Uncertainty: Unleashing the Potential of Knowledge...", "date": "", "ddg_snippet": "CrossKD : Cross -Head Knowledge Distillation for Object Detection proposes a method for distilling knowledge across different detection heads within the same model.", "subpage_snippet": "", "source": "www.aimodels.fyi", "link": "https://www.aimodels.fyi/papers/arxiv/teaching-uncertainty-unleashing-potential-knowledge-distillation-object", "content": "CrossKD : Cross -Head Knowledge Distillation for Object Detection proposes a method for distilling knowledge across different detection heads within the same model."}

data/sampled_jsons/9m87e9Keq1_RL_Incorrect_Synthetic_Data_Scales_the_Efficiency_of_LLM_Math_Reasoning_by_Eight-Fold_Fig.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM ... RL on Incorrect Synthetic Data Scales the Efficiency of LLM ... RL on Incorrect Synthetic Data · MinWoo Park RLonIncorrectSyntheticDataScalesthe ... Reinforcement Learning for LLM Reasoning Images RL on Incorrect Synthetic Data Scales the Efficiency of LLM ... scaling-LLM-math-synthetic-data/README.md at master - GitHub", "date": "", "ddg_snippet": "Jun 20, 2024 · Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts. In this paper, we investigate this question for math reasoning via an empirical study, followed by building a conceptual understanding of our observations. First, we find that while the typical approach of finetuning a model on synthetic correct or positive ... To provide clarity on how synthetic data contributes to performance, we aim to understand its impact on LLM capabilities via a study on math reasoning , a prevalent scenario where synthetic data is used. Typically, in this setting, synthetic data corresponds to correct or positive model-generated responses for a novel set of initial problems synthesized by prompting capable models [29, 31]. The ... Jun 20, 2024 · abstract: Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts. In this paper, we investigate this question for math reasoning via an empirical study, followed by building a conceptual understanding of our observations. RLonIncorrectSyntheticDataScalesthe EfficiencyofLLMMathReasoningbyEight- Fold RL on Incorrect Synthetic Data Scales the Eficiency of LLM Math Reasoning by Eight-Fold RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold . Setlur, Garg, Geng, Garg, Smith, Kumar. NeurIPS 2024 Rewarding Progress: Scaling up Automated Process Supervision for LLM Reasoning View all First, we find that while the typical approach of finetuning a model on synthetic correct or positive problem-solution pairs generated by capable models offers modest performance gains, sampling more correct solutions from the finetuned learner itself followed by subsequent fine-tuning on this self-generated data doubles the efficiency of the ... Code and data used in the paper: \"Training on Incorrect Synthetic Data via RL Scales LLM Math Reasoning Eight-Fold \"", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2406.14532", "content": "Jun 20, 2024 · Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts. In this paper, we investigate this question for math reasoning via an empirical study, followed by building a conceptual understanding of our observations. First, we find that while the typical approach of finetuning a model on synthetic correct or positive ... To provide clarity on how synthetic data contributes to performance, we aim to understand its impact on LLM capabilities via a study on math reasoning , a prevalent scenario where synthetic data is used. Typically, in this setting, synthetic data corresponds to correct or positive model-generated responses for a novel set of initial problems synthesized by prompting capable models [29, 31]. The ... Jun 20, 2024 · abstract: Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts. In this paper, we investigate this question for math reasoning via an empirical study, followed by building a conceptual understanding of our observations. RLonIncorrectSyntheticDataScalesthe EfficiencyofLLMMathReasoningbyEight- Fold RL on Incorrect Synthetic Data Scales the Eficiency of LLM Math Reasoning by Eight-Fold RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold . Setlur, Garg, Geng, Garg, Smith, Kumar. NeurIPS 2024 Rewarding Progress: Scaling up Automated Process Supervision for LLM Reasoning View all First, we find that while the typical approach of finetuning a model on synthetic correct or positive problem-solution pairs generated by capable models offers modest performance gains, sampling more correct solutions from the finetuned learner itself followed by subsequent fine-tuning on this self-generated data doubles the efficiency of the ... Code and data used in the paper: \"Training on Incorrect Synthetic Data via RL Scales LLM Math Reasoning Eight-Fold \""}
+{"idx": 1, "title": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM ...", "date": "", "ddg_snippet": "To provide clarity on how synthetic data contributes to performance, we aim to understand its impact on LLM capabilities via a study on math reasoning , a prevalent scenario where synthetic data is used. Typically, in this setting, synthetic data corresponds to correct or positive model-generated responses for a novel set of initial problems synthesized by prompting capable models [29, 31]. The ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=9m87e9Keq1", "content": "To provide clarity on how synthetic data contributes to performance, we aim to understand its impact on LLM capabilities via a study on math reasoning , a prevalent scenario where synthetic data is used. Typically, in this setting, synthetic data corresponds to correct or positive model-generated responses for a novel set of initial problems synthesized by prompting capable models [29, 31]. The ..."}
+{"idx": 2, "title": "Reinforcement Learning for LLM Reasoning", "date": "", "ddg_snippet": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold . Setlur, Garg, Geng, Garg, Smith, Kumar. NeurIPS 2024 Rewarding Progress: Scaling up Automated Process Supervision for LLM Reasoning", "subpage_snippet": "", "source": "cs224r.stanford.edu", "link": "https://cs224r.stanford.edu/slides/10_cs224r-rl_for_reasoning_lecture.pdf", "content": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold . Setlur, Garg, Geng, Garg, Smith, Kumar. NeurIPS 2024 Rewarding Progress: Scaling up Automated Process Supervision for LLM Reasoning"}
+{"idx": 3, "title": "RL on Incorrect Synthetic Data · MinWoo Park", "date": "", "ddg_snippet": "Jun 20, 2024 · abstract: Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts. In this paper, we investigate this question for math reasoning via an empirical study, followed by building a conceptual understanding of our observations.", "subpage_snippet": "", "source": "dsdanielpark.github.io", "link": "https://dsdanielpark.github.io/llm/2024-06-25-RLonIncorrectSyntheticData.html", "content": "Jun 20, 2024 · abstract: Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts. In this paper, we investigate this question for math reasoning via an empirical study, followed by building a conceptual understanding of our observations."}
+{"idx": 4, "title": "RLonIncorrectSyntheticDataScalesthe ...", "date": "", "ddg_snippet": "RLonIncorrectSyntheticDataScalesthe EfficiencyofLLMMathReasoningbyEight- Fold RL on Incorrect Synthetic Data Scales the Eficiency of LLM Math Reasoning by Eight-Fold", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/381604579_RL_on_Incorrect_Synthetic_Data_Scales_the_Efficiency_of_LLM_Math_Reasoning_by_Eight-Fold/fulltext/6675a4158408575b837d3b71/RL-on-Incorrect-Synthetic-Data-Scales-the-Efficiency-of-LLM-Math-Reasoning-by-Eight-Fold.pdf", "content": "RLonIncorrectSyntheticDataScalesthe EfficiencyofLLMMathReasoningbyEight- Fold RL on Incorrect Synthetic Data Scales the Eficiency of LLM Math Reasoning by Eight-Fold"}
+{"idx": 5, "title": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM ...", "date": "", "ddg_snippet": "First, we find that while the typical approach of finetuning a model on synthetic correct or positive problem-solution pairs generated by capable models offers modest performance gains, sampling more correct solutions from the finetuned learner itself followed by subsequent fine-tuning on this self-generated data doubles the efficiency of the ...", "subpage_snippet": "", "source": "proceedings.neurips.cc", "link": "https://proceedings.neurips.cc/paper_files/paper/2024/hash/4b77d5b896c321a29277524a98a50215-Abstract-Conference.html", "content": "First, we find that while the typical approach of finetuning a model on synthetic correct or positive problem-solution pairs generated by capable models offers modest performance gains, sampling more correct solutions from the finetuned learner itself followed by subsequent fine-tuning on this self-generated data doubles the efficiency of the ..."}
+{"idx": 6, "title": "scaling-LLM-math-synthetic-data/README.md at master - GitHub", "date": "", "ddg_snippet": "Code and data used in the paper: \"Training on Incorrect Synthetic Data via RL Scales LLM Math Reasoning Eight-Fold \"", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/ars22/scaling-LLM-math-synthetic-data/blob/master/README.md", "content": "Code and data used in the paper: \"Training on Incorrect Synthetic Data via RL Scales LLM Math Reasoning Eight-Fold \""}
+{"idx": 7, "title": "NeurIPS Poster RL on Incorrect Synthetic Data Scales the ...", "date": "", "ddg_snippet": "Abstract: Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts. In this paper, we investigate this question for math reasoning via an empirical study, followed by building a conceptual understanding of our observations.", "subpage_snippet": "", "source": "neurips.cc", "link": "https://neurips.cc/virtual/2024/poster/96295", "content": "Abstract: Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts. In this paper, we investigate this question for math reasoning via an empirical study, followed by building a conceptual understanding of our observations."}
+{"idx": 8, "title": "RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math ...", "date": "", "ddg_snippet": "Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts.", "subpage_snippet": "", "source": "synthical.com", "link": "https://synthical.com/article/RL-on-Incorrect-Synthetic-Data-Scales-the-Efficiency-of-LLM-Math-Reasoning-by-Eight-Fold-17552ae8-02cf-4931-9351-04c6f7b243c5", "content": "Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts."}
+{"idx": 9, "title": "(PDF) RL on Incorrect Synthetic Data Scales the Efficiency of LLM ...", "date": "", "ddg_snippet": "PDF | Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts.arXiv:2406.14532v1 [cs.LG] 20 Jun 2024. RL on Incorrect Synthetic Data Scales the Effici ency of LLM Math Reasoning by Eight -F old.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/381604579_RL_on_Incorrect_Synthetic_Data_Scales_the_Efficiency_of_LLM_Math_Reasoning_by_Eight-Fold", "content": "PDF | Training on model-generated synthetic data is a promising approach for finetuning LLMs, but it remains unclear when it helps or hurts.arXiv:2406.14532v1 [cs.LG] 20 Jun 2024. RL on Incorrect Synthetic Data Scales the Effici ency of LLM Math Reasoning by Eight -F old."}

data/sampled_jsons/9v1eW8HgMU_Archetypal_SAE-_Adaptive_and_Stable_Dictionary_Learning_for_Concept_Extraction_in_Large_V.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "Sparse Autoencoders ( SAEs ) have emerged as a powerful framework for machine learning interpretability, enabling the unsupervised decomposition of model representations into a dictionary of abstract, human-interpretable concepts . However, we reveal a fundamental limitation: existing SAEs exhibit severe instability, as identical models trained on similar datasets can produce sharply different ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2502.12892", "content": "Sparse Autoencoders ( SAEs ) have emerged as a powerful framework for machine learning interpretability, enabling the unsupervised decomposition of model representations into a dictionary of abstract, human-interpretable concepts . However, we reveal a fundamental limitation: existing SAEs exhibit severe instability, as identical models trained on similar datasets can produce sharply different ..."}
+{"idx": 1, "title": "Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "Our results provide sub- stantial evidence that A-SAEs find more structured and coherent concepts . Further, to enable reproduction, we open-source our extensive codebase for large -scale SAE training on modern vision models . 2. Related Work Sparse Coding & Dictionary Learning .", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=9v1eW8HgMU", "content": "Our results provide sub- stantial evidence that A-SAEs find more structured and coherent concepts . Further, to enable reproduction, we open-source our extensive codebase for large -scale SAE training on modern vision models . 2. Related Work Sparse Coding & Dictionary Learning ."}
+{"idx": 2, "title": "Archetypal SAEs: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "In recent years, the field of interpretability has focused heavily on concept -based approaches. As deep learning models advance to unprecedented scales in vision , language, and multimodal domains, the ability to automatically decompose learned representations into meaningful \" concepts \" has become one of the most promising interpretability strategies. Concept extraction attempts to identify ...", "subpage_snippet": "", "source": "kempnerinstitute.harvard.edu", "link": "https://kempnerinstitute.harvard.edu/research/deeper-learning/archetypal-saes-adaptive-and-stable-dictionary-learning-for-concept-extraction-in-large-vision-models/", "content": "In recent years, the field of interpretability has focused heavily on concept -based approaches. As deep learning models advance to unprecedented scales in vision , language, and multimodal domains, the ability to automatically decompose learned representations into meaningful \" concepts \" has become one of the most promising interpretability strategies. Concept extraction attempts to identify ..."}
+{"idx": 3, "title": "Archetypal SAEs: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "As deep learning models advance to unprecedented scales in vision , language, and multimodal domains, the ability to automatically decompose learned representations into meaningful \" concepts ...", "subpage_snippet": "", "source": "medium.com", "link": "https://medium.com/@kempnerinstitute/archetypal-saes-adaptive-and-stable-dictionary-learning-for-concept-extraction-in-large-vision-acf95010c691", "content": "As deep learning models advance to unprecedented scales in vision , language, and multimodal domains, the ability to automatically decompose learned representations into meaningful \" concepts ..."}
+{"idx": 4, "title": "Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "Join the discussion on this paper pageArchetypal SAE : Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/papers/2502.12892", "content": "Join the discussion on this paper pageArchetypal SAE : Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models"}
+{"idx": 5, "title": "Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "Check out the Paper. All credit for this research goes to the researchers of this project. Also, feel free to follow us on Twitter and don't forget to join our 80k+ ML SubReddit. The post Archetypal SAE : Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models appeared first on MarkTechPost. \"}]] [ [ {\"value\":\"Artificial Neural Networks (ANNs) have ...", "subpage_snippet": "", "source": "phdstudio.org", "link": "https://phdstudio.org/2025/03/17/archetypal-sae-adaptive-and-stable-dictionary-learning-for-concept-extraction-in-large-vision-models-sajjad-ansari-artificial-intelligence-category-marktechpost/", "content": "Check out the Paper. All credit for this research goes to the researchers of this project. Also, feel free to follow us on Twitter and don't forget to join our 80k+ ML SubReddit. The post Archetypal SAE : Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models appeared first on MarkTechPost. \"}]] [ [ {\"value\":\"Artificial Neural Networks (ANNs) have ..."}
+{"idx": 6, "title": "Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "Archetypal SAE : Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models : Paper and Code. Sparse Autoencoders ( SAEs ) have emerged as a powerful framework for machine learning interpretability, enabling the unsupervised decomposition of model representations into a dictionary of abstract, human-interpretable concepts . However, we reveal a fundamental limitation ...", "subpage_snippet": "", "source": "www.catalyzex.com", "link": "https://www.catalyzex.com/paper/archetypal-sae-adaptive-and-stable-dictionary", "content": "Archetypal SAE : Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models : Paper and Code. Sparse Autoencoders ( SAEs ) have emerged as a powerful framework for machine learning interpretability, enabling the unsupervised decomposition of model representations into a dictionary of abstract, human-interpretable concepts . However, we reveal a fundamental limitation ..."}
+{"idx": 7, "title": "Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "Across all evaluations, RA- SAE s consistently yield more structured representations while uncovering novel, semantically meaningful concepts in large -scale vision models .", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2502.12892v1", "content": "Across all evaluations, RA- SAE s consistently yield more structured representations while uncovering novel, semantically meaningful concepts in large -scale vision models ."}
+{"idx": 8, "title": "Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "View recent discussion. Abstract: Sparse Autoencoders ( SAEs ) have emerged as a powerful framework for machine learning interpretability, enabling the unsupervised decomposition of model representations into a dictionary of abstract, human-interpretable concepts . However, we reveal a fundamental limitation: existing SAEs exhibit severe instability, as identical models trained on similar ...", "subpage_snippet": "", "source": "www.alphaxiv.org", "link": "https://www.alphaxiv.org/overview/2502.12892v2", "content": "View recent discussion. Abstract: Sparse Autoencoders ( SAEs ) have emerged as a powerful framework for machine learning interpretability, enabling the unsupervised decomposition of model representations into a dictionary of abstract, human-interpretable concepts . However, we reveal a fundamental limitation: existing SAEs exhibit severe instability, as identical models trained on similar ..."}
+{"idx": 9, "title": "Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept ...", "date": "", "ddg_snippet": "Across all evaluations, RA- SAEs consistently yield more structured representations while uncovering novel, semantically meaningful concepts in large -scale vision models .", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/forum?id=9v1eW8HgMU", "content": "Across all evaluations, RA- SAEs consistently yield more structured representations while uncovering novel, semantically meaningful concepts in large -scale vision models ."}

data/sampled_jsons/ACD_Repo_Schubert_polynomials_accuracy_table_1_2503.06366.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Polynomial - Wikipedia", "date": "", "ddg_snippet": "In mathematics, a polynomial is a mathematical expression consisting of indeterminates (also called variables) and coefficients, that involves only the operations of addition, subtraction, multiplication and exponentiation to nonnegative integer powers, and has a finite number of terms.[ 1 ][2][3]...", "subpage_snippet": "", "source": "en.wikipedia.org", "link": "https://en.wikipedia.org/wiki/Polynomial", "content": "In mathematics, a polynomial is a mathematical expression consisting of indeterminates (also called variables) and coefficients, that involves only the operations of addition, subtraction, multiplication and exponentiation to nonnegative integer powers, and has a finite number of terms.[ 1 ][2][3]..."}
+{"idx": 1, "title": "Lectures on Schubert Polynomials | U-M LSA Mathematics", "date": "", "ddg_snippet": "This is a lecture series on Schubert polynomials , given by William Fulton in Fall 2025. It meets 3:30-5:00 pm on Thursdays, in EH 4096, beginning on September 4.", "subpage_snippet": "", "source": "lsa.umich.edu", "link": "https://lsa.umich.edu/math/seminars/seminars-and-colloquia/lectures-on-schubert-polynomials.html", "content": "This is a lecture series on Schubert polynomials , given by William Fulton in Fall 2025. It meets 3:30-5:00 pm on Thursdays, in EH 4096, beginning on September 4."}
+{"idx": 2, "title": "Machine Learning meets Algebraic Combinatorics: A Suite of Datasets...", "date": "", "ddg_snippet": "arXiv: 2503 . 06366 v 1 [cs.LG] 9 Mar 2025.Both small MLPs and transformers can achieve high- accuracy ( Table 1 ) as well as LLMs via pro-gram synthesis ( Table 6). Some of the latter is an artifact of how we originally sampled zero-valued structure constants.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2503.06366", "content": "arXiv: 2503 . 06366 v 1 [cs.LG] 9 Mar 2025.Both small MLPs and transformers can achieve high- accuracy ( Table 1 ) as well as LLMs via pro-gram synthesis ( Table 6). Some of the latter is an artifact of how we originally sampled zero-valued structure constants."}
+{"idx": 3, "title": "Paper page - Machine Learning meets Algebraic Combinatorics...", "date": "", "ddg_snippet": "Papers. arxiv: 2503 . 06366 .A collection of datasets, Algebraic Combinatorics Dataset Repository , supports conjecture generation in algebraic combinatorics using machine learning models. AI-generated summary.", "subpage_snippet": "", "source": "huggingface.co", "link": "https://huggingface.co/papers/2503.06366", "content": "Papers. arxiv: 2503 . 06366 .A collection of datasets, Algebraic Combinatorics Dataset Repository , supports conjecture generation in algebraic combinatorics using machine learning models. AI-generated summary."}
+{"idx": 4, "title": "Machine Learning meets Algebraic Combinatorics: A Suite of... | alphaXiv", "date": "", "ddg_snippet": "To address this, we introduce a new collection of datasets, the Algebraic Combinatorics Dataset Repository ( ACD Repo ), representing either foundational results or open problems in algebraic combinatorics...", "subpage_snippet": "", "source": "www.alphaxiv.org", "link": "https://www.alphaxiv.org/ja/overview/2503.06366v1", "content": "To address this, we introduce a new collection of datasets, the Algebraic Combinatorics Dataset Repository ( ACD Repo ), representing either foundational results or open problems in algebraic combinatorics..."}
+{"idx": 5, "title": "GitHub - pnnl/ML4AlgComb: ML Benchmarks in Algebraic Combinatorics", "date": "", "ddg_snippet": "Schubert polynomial structure constants: Schubert polynomials are a family of polynomials indexed by permutations of $S_n$. Developed to study the cohomology ring of the flag variety, they have deep connections to algebraic geometry, Lie theory, and representation theory.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/pnnl/ML4AlgComb", "content": "Schubert polynomial structure constants: Schubert polynomials are a family of polynomials indexed by permutations of $S_n$. Developed to study the cohomology ring of the flag variety, they have deep connections to algebraic geometry, Lie theory, and representation theory."}
+{"idx": 6, "title": "Machine Learning meets Algebraic Combinatorics: A", "date": "", "ddg_snippet": "Schubert polynomial structure coefficients datasheet. Symmetric group characters. Table 1 : Naive baseline model accuracy on classification datasets. Results are averaged over three random weight initializations with 95% confidence intervals.", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=KQ1gI5qzAf", "content": "Schubert polynomial structure coefficients datasheet. Symmetric group characters. Table 1 : Naive baseline model accuracy on classification datasets. Results are averaged over three random weight initializations with 95% confidence intervals."}
+{"idx": 7, "title": "How to Install Language Packs in LibreOffice & OpenOffice", "date": "", "ddg_snippet": "Table of Contents.In GNU/Linux systems, it’s easier to install language packs since the packages are available in major Linux distribution repositories such as Ubuntu, Linux Mint, or Fedora. In Ubuntu, Debian, Linux Mint or “apt” based system, install the language pack using the...", "subpage_snippet": "", "source": "www.libreofficehelp.com", "link": "https://www.libreofficehelp.com/install-language-packs-libreoffice/", "content": "Table of Contents.In GNU/Linux systems, it’s easier to install language packs since the packages are available in major Linux distribution repositories such as Ubuntu, Linux Mint, or Fedora. In Ubuntu, Debian, Linux Mint or “apt” based system, install the language pack using the..."}
+{"idx": 8, "title": "Читы и коды в Kingdom Come Deliverance 2: все консольные команды", "date": "", "ddg_snippet": "00cbcdca-e6ee-4fe6-bf87-2 acd 9f2b21aa.", "subpage_snippet": "", "source": "wotpack.ru", "link": "https://wotpack.ru/chity-kingdom-come-deliverance-2-kody-konsolnye-komandy-i-id-predmetov/", "content": "00cbcdca-e6ee-4fe6-bf87-2 acd 9f2b21aa."}
+{"idx": 9, "title": "PDF to Scan Online 100% Free | i 2PDF", "date": "", "ddg_snippet": "Table to PDF.A simple keystroke error, a misplaced deletion, or an incorrect insertion can compromise the accuracy and reliability of the document.", "subpage_snippet": "", "source": "www.i2pdf.com", "link": "https://www.i2pdf.com/pdf-to-scan", "content": "Table to PDF.A simple keystroke error, a misplaced deletion, or an incorrect insertion can compromise the accuracy and reliability of the document."}

data/sampled_jsons/AERO_Enhancing_Sharding_Blockchain_Deep_Reinforcement_Learning_Account_Migration_experimental_setup__year_2024.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "AERO: Enhancing Sharding Blockchain via Deep Reinforcement Learning for ...", "date": "", "ddg_snippet": "We propose AERO , a deep reinforcement learning framework to facilitate efficient account migration in sharding blockchains . AERO employs a prefix-based grouping strategy to enable group-level migration decisions and capture complex transaction patterns and relationships between accounts .", "subpage_snippet": "", "source": "dl.acm.org", "link": "https://dl.acm.org/doi/10.1145/3696410.3714926", "content": "We propose AERO , a deep reinforcement learning framework to facilitate efficient account migration in sharding blockchains . AERO employs a prefix-based grouping strategy to enable group-level migration decisions and capture complex transaction patterns and relationships between accounts ."}
+{"idx": 1, "title": "AERO: Enhancing Sharding Blockchain via Deep Reinforcement Learning for ...", "date": "", "ddg_snippet": "To address these scalability issues, account migration ofers a promising solution. However, existing migration solutions struggle with the high computational overhead and insuficient capture of complex transaction patterns. We propose AERO , a deep reinforcement learning framework to facilitate eficient account migration in sharding blockchains .", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=WcuXvn3HVk", "content": "To address these scalability issues, account migration ofers a promising solution. However, existing migration solutions struggle with the high computational overhead and insuficient capture of complex transaction patterns. We propose AERO , a deep reinforcement learning framework to facilitate eficient account migration in sharding blockchains ."}
+{"idx": 2, "title": "Deep Learning Approaches for Blockchain Scalability Through Sharding ...", "date": "", "ddg_snippet": "Sharding technology creates new difficulties even as it helps traditional blockchain networks overcome performance issues. The distribution of malicious nodes may be unequal because of random node allocation, resulting in performance variations and security threats. Current reputation-based sharding techniques frequently ignore node performance features and don't take care of leader election ...", "subpage_snippet": "", "source": "ieeexplore.ieee.org", "link": "https://ieeexplore.ieee.org/abstract/document/10830433", "content": "Sharding technology creates new difficulties even as it helps traditional blockchain networks overcome performance issues. The distribution of malicious nodes may be unequal because of random node allocation, resulting in performance variations and security threats. Current reputation-based sharding techniques frequently ignore node performance features and don't take care of leader election ..."}
+{"idx": 3, "title": "A sharding blockchain protocol for enhanced scalability and performance ...", "date": "", "ddg_snippet": "However, existing sharding blockchain protocols suffer from a high cross-shard ratio, high transaction latency, limited throughput enhancement, and high account migration . To address these problems, this paper proposes a sharding blockchain protocol for enhanced scalability and performance optimization through account transaction reconfiguration.", "subpage_snippet": "", "source": "www.sciencedirect.com", "link": "https://www.sciencedirect.com/science/article/pii/S1319157824002738", "content": "However, existing sharding blockchain protocols suffer from a high cross-shard ratio, high transaction latency, limited throughput enhancement, and high account migration . To address these problems, this paper proposes a sharding blockchain protocol for enhanced scalability and performance optimization through account transaction reconfiguration."}
+{"idx": 4, "title": "PDF AERO: Enhancing Sharding Blockchain via Deep Reinforcement Learning for ...", "date": "", "ddg_snippet": "Motivated by the need for an eficient and decentralized account migration mechanism, we explore applying deep reinforcement learning (DRL) [27] to this problem. DRL is highly efective in handling sequential decision-making tasks and has demonstrated significant potential in optimizing complex systems with expansive state and action spaces [24]. In the context of account migration , the account ...", "subpage_snippet": "", "source": "zhenxiao.com", "link": "http://zhenxiao.com/papers/WWW_AERO_camera_ready.pdf", "content": "Motivated by the need for an eficient and decentralized account migration mechanism, we explore applying deep reinforcement learning (DRL) [27] to this problem. DRL is highly efective in handling sequential decision-making tasks and has demonstrated significant potential in optimizing complex systems with expansive state and action spaces [24]. In the context of account migration , the account ..."}
+{"idx": 5, "title": "TBDD: A New Trust-based, DRL-driven Framework for Blockchain Sharding ...", "date": "", "ddg_snippet": "Integrating sharded blockchain with IoT presents a solution for trust issues and optimized data flow. Sharding boosts blockchain scalability by dividing its nodes into parallel shards, yet it's vulnerable to the $1\\\\%$ attacks where dishonest nodes target a shard to corrupt the entire blockchain . Balancing security with scalability is pivotal for such systems. Deep Reinforcement Learning (DRL ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2401.00632", "content": "Integrating sharded blockchain with IoT presents a solution for trust issues and optimized data flow. Sharding boosts blockchain scalability by dividing its nodes into parallel shards, yet it's vulnerable to the $1\\\\%$ attacks where dishonest nodes target a shard to corrupt the entire blockchain . Balancing security with scalability is pivotal for such systems. Deep Reinforcement Learning (DRL ..."}
+{"idx": 6, "title": "Mingxuan Song", "date": "", "ddg_snippet": "\" AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration .\" Proceedings of the Web Conference, May 2025. [PDF] CVPR 2024 CCF-A Shenglin Yin, Zhen Xiao*, Mingxuan Song, and Jieyi Long. \"Adversarial Distillation Based on Slack Matching and Attribution Region Alignment.\"", "subpage_snippet": "", "source": "www.songmingxuan.com", "link": "https://www.songmingxuan.com/", "content": "\" AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration .\" Proceedings of the Web Conference, May 2025. [PDF] CVPR 2024 CCF-A Shenglin Yin, Zhen Xiao*, Mingxuan Song, and Jieyi Long. \"Adversarial Distillation Based on Slack Matching and Attribution Region Alignment.\""}
+{"idx": 7, "title": "Papers - Zhen Xiao", "date": "", "ddg_snippet": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration Proc. of the Web Conference 2025 (WWW 2025), May 2025. Lichen Pan, Juncheng Liu, Yongquan Fu, Jinhui Yuan, Rongkai Zhang, Pengze Li, and Zhen Xiao.", "subpage_snippet": "", "source": "zhenxiao.com", "link": "http://zhenxiao.com/papers/", "content": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration Proc. of the Web Conference 2025 (WWW 2025), May 2025. Lichen Pan, Juncheng Liu, Yongquan Fu, Jinhui Yuan, Rongkai Zhang, Pengze Li, and Zhen Xiao."}
+{"idx": 8, "title": "Zhen Xiao | Semantic Scholar", "date": "", "ddg_snippet": "This work proposes AERO , a deep reinforcement learning framework to facilitate efficient account migration in sharding blockchains , which employs a prefix-based grouping strategy to enable group-level migration decisions and capture complex transaction patterns and relationships between accounts .", "subpage_snippet": "", "source": "www.semanticscholar.org", "link": "https://www.semanticscholar.org/author/Zhen-Xiao/2257215916", "content": "This work proposes AERO , a deep reinforcement learning framework to facilitate efficient account migration in sharding blockchains , which employs a prefix-based grouping strategy to enable group-level migration decisions and capture complex transaction patterns and relationships between accounts ."}
+{"idx": 9, "title": "Jieyi Long (0009-0007-4646-7131) - ORCID", "date": "", "ddg_snippet": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration 2025-04-22 | Conference paper DOI: 10.1145/3696410.3714926", "subpage_snippet": "", "source": "orcid.org", "link": "https://orcid.org/0009-0007-4646-7131", "content": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration 2025-04-22 | Conference paper DOI: 10.1145/3696410.3714926"}

data/sampled_jsons/AERO_Enhancing_Sharding_Blockchain_reward_function_Equations_8-12.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement...", "date": "", "ddg_snippet": "4.3 Reward Function . 4.4 Transition Dynamics.Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration.", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=WcuXvn3HVk", "content": "4.3 Reward Function . 4.4 Transition Dynamics.Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration."}
+{"idx": 1, "title": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement...", "date": "", "ddg_snippet": "D Full Shard Transaction Distriction. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration.", "subpage_snippet": "", "source": "zhenxiao.com", "link": "http://zhenxiao.com/papers/WWW_AERO_camera_ready.pdf", "content": "D Full Shard Transaction Distriction. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration."}
+{"idx": 2, "title": "(PDF) Sharding for Blockchain based Mobile Edge Computing...", "date": "", "ddg_snippet": "Additionally, the reward function R(K, SM, T G). represents the long term reward , which can be calculated as. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/363131667_Sharding_for_Blockchain_based_Mobile_Edge_Computing_System_A_Deep_Reinforcement_Learning_Approach", "content": "Additionally, the reward function R(K, SM, T G). represents the long term reward , which can be calculated as. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration."}
+{"idx": 3, "title": "Security Implications of Blockchain Sharding : Navigating Risks and...", "date": "", "ddg_snippet": "Enhancing Validator Security and Shard Data Integrity. The Role of Cryptography in Securing Sharded Blockchains . Auditing and Setting Standards for Secure Sharded Blockchain Systems.", "subpage_snippet": "", "source": "blockchainbulletinweekly.com", "link": "https://blockchainbulletinweekly.com/security-implications-blockchain-sharding/", "content": "Enhancing Validator Security and Shard Data Integrity. The Role of Cryptography in Securing Sharded Blockchains . Auditing and Setting Standards for Secure Sharded Blockchain Systems."}
+{"idx": 4, "title": "Unsolved Problems in Blockchain Sharding | NEAR", "date": "", "ddg_snippet": "The core idea in sharded blockchains is that most participants operating or using the network cannot validate blocks in all the shards . As such, whenever any participant needs to interact with a particular shard they generally cannot download and validate the entire history of the shard .", "subpage_snippet": "", "source": "www.near.org", "link": "https://www.near.org/blog/unsolved-problems-in-blockchain-sharding", "content": "The core idea in sharded blockchains is that most participants operating or using the network cannot validate blocks in all the shards . As such, whenever any participant needs to interact with a particular shard they generally cannot download and validate the entire history of the shard ."}
+{"idx": 5, "title": "Valid Points: The Risks and Rewards of Sharding", "date": "", "ddg_snippet": "A sharded blockchain enables blockchain capacity and transaction throughput to increase along with the number of nodes, such that scalability does not sacrifice network decentralization.", "subpage_snippet": "", "source": "www.coindesk.com", "link": "https://www.coindesk.com/tech/2021/05/26/valid-points-the-risks-and-rewards-of-sharding", "content": "A sharded blockchain enables blockchain capacity and transaction throughput to increase along with the number of nodes, such that scalability does not sacrifice network decentralization."}
+{"idx": 6, "title": "Ethereum sharding : A beginner’s guide to blockchain sharding", "date": "", "ddg_snippet": "Coincidentally, shard also means “a small part of something larger.” And that’s exactly what sharding in blockchain technology aims to do– break up a blockchain network into smaller, manageable pieces called shards .", "subpage_snippet": "", "source": "cointelegraph.com", "link": "https://cointelegraph.com/learn/articles/ethereum-sharding-a-beginners-guide-to-blockchain-sharding", "content": "Coincidentally, shard also means “a small part of something larger.” And that’s exactly what sharding in blockchain technology aims to do– break up a blockchain network into smaller, manageable pieces called shards ."}
+{"idx": 7, "title": "What Is Blockchain Sharding ? - 101 Blockchains", "date": "", "ddg_snippet": "Sharding is the process of breaking down a blockchain network’s workload into smaller pieces. Learn more about blockchain sharding in this guide now.", "subpage_snippet": "", "source": "101blockchains.com", "link": "https://101blockchains.com/what-is-blockchain-sharding/", "content": "Sharding is the process of breaking down a blockchain network’s workload into smaller pieces. Learn more about blockchain sharding in this guide now."}
+{"idx": 8, "title": "What Exactly Is Blockchain Sharding And How Does It Impact The...", "date": "", "ddg_snippet": "In blockchain , sharding means dividing the network into several shards , with each shard constituting its independent blockchain . Shard chains process and validate a subset of all transactions, enabling parallel processing.", "subpage_snippet": "", "source": "blog.bitamp.com", "link": "https://blog.bitamp.com/what-exactly-is-blockchain-sharding-and-how-does-it-impact-the-network/", "content": "In blockchain , sharding means dividing the network into several shards , with each shard constituting its independent blockchain . Shard chains process and validate a subset of all transactions, enabling parallel processing."}
+{"idx": 9, "title": "Blockchain Sharding : All You Need To Know", "date": "", "ddg_snippet": "Introduction Sharding is another novel concept that has come to rescue blockchains from their scalability woes. It functions to minimize.", "subpage_snippet": "", "source": "coinpedia.org", "link": "https://coinpedia.org/guest-post/blockchain-sharding-all-you-need-to-know/", "content": "Introduction Sharding is another novel concept that has come to rescue blockchains from their scalability woes. It functions to minimize."}

data/sampled_jsons/AERO_Enhancing_Sharding_Blockchain_via_Deep_Reinforcement_Learning_for_Account_Migration_4.3_Reward_.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Reinforcement learning - Wikipedia", "date": "", "ddg_snippet": "The typical framing of a reinforcement learning scenario: an agent takes actions in an environment, which is interpreted into a reward and a state representation, which are fed back to the agent. Reinforcement learning is an interdisciplinary area of...", "subpage_snippet": "", "source": "en.wikipedia.org", "link": "https://en.wikipedia.org/wiki/Reinforcement_learning", "content": "The typical framing of a reinforcement learning scenario: an agent takes actions in an environment, which is interpreted into a reward and a state representation, which are fed back to the agent. Reinforcement learning is an interdisciplinary area of..."}
+{"idx": 1, "title": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement ...", "date": "", "ddg_snippet": "Blockchain , Sharding , Account migration , Reinforcement learning . ACM Reference Format: Anonymous authors.Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration .", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=WcuXvn3HVk", "content": "Blockchain , Sharding , Account migration , Reinforcement learning . ACM Reference Format: Anonymous authors.Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration ."}
+{"idx": 2, "title": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement ...", "date": "", "ddg_snippet": "Blockchain , Sharding , Account migration , Reinforcement learning . ∗Corresponding author.Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration .", "subpage_snippet": "", "source": "zhenxiao.com", "link": "http://zhenxiao.com/papers/WWW_AERO_camera_ready.pdf", "content": "Blockchain , Sharding , Account migration , Reinforcement learning . ∗Corresponding author.Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration ."}
+{"idx": 3, "title": "(PDF) Sharding for Blockchain based Mobile Edge Computing...", "date": "", "ddg_snippet": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration . Blockchain -Based Transformer-Assisted Multi-Agent Reinforcement Learning for Resource Allocation and Computation Offloading in 5G Private Networks.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/363131667_Sharding_for_Blockchain_based_Mobile_Edge_Computing_System_A_Deep_Reinforcement_Learning_Approach", "content": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration . Blockchain -Based Transformer-Assisted Multi-Agent Reinforcement Learning for Resource Allocation and Computation Offloading in 5G Private Networks."}
+{"idx": 4, "title": "ContribChain: A Stress-Balanced Blockchain Sharding Protocol with...", "date": "", "ddg_snippet": "[27] employs Deep Reinforcement Learn -ing (DRL) to optimize state placement. While these ap-proaches enhance account allocation, they fail to consider performance disparities among shards , and thus do not achieve stress balance.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2505.06899", "content": "[27] employs Deep Reinforcement Learn -ing (DRL) to optimize state placement. While these ap-proaches enhance account allocation, they fail to consider performance disparities among shards , and thus do not achieve stress balance."}
+{"idx": 5, "title": "LB-Chain: Load-Balanced and Low-Latency Blockchain Sharding via ...", "date": "", "ddg_snippet": "Index Terms— Blockchain , blockchain sharding , load balance, account migration .The objective of the account migration is to safely migrate accounts to proper shards according to the allocation results, with low through-put loss, low latency, and high fairness.", "subpage_snippet": "", "source": "home.cse.ust.hk", "link": "https://home.cse.ust.hk/~weiwa/papers/lb-chain-tpds22.pdf", "content": "Index Terms— Blockchain , blockchain sharding , load balance, account migration .The objective of the account migration is to safely migrate accounts to proper shards according to the allocation results, with low through-put loss, low latency, and high fairness."}
+{"idx": 6, "title": "[Paper Reading] 1 SkyChain: A dynamic blockchain sharding system...", "date": "", "ddg_snippet": "Deep reinforcement learning can learn the characteristics of the blockchain sharding system from previous experience and adopt appropriate sharding strategies based on the current network status to obtain long-term returns...", "subpage_snippet": "", "source": "www.codetd.com", "link": "https://www.codetd.com/en/article/17259394", "content": "Deep reinforcement learning can learn the characteristics of the blockchain sharding system from previous experience and adopt appropriate sharding strategies based on the current network status to obtain long-term returns..."}
+{"idx": 7, "title": "CMC | Free Full-Text | DRL-AMIR: Intelligent Flow Scheduling for...", "date": "", "ddg_snippet": "[26] proposed Message Passing Deep Reinforcement Learning (MPDRL), which uses graph neural network to interact with the network topology environment and extract knowledge through information transmission between links to achieve traffic load balancing [27].", "subpage_snippet": "", "source": "www.techscience.com", "link": "https://www.techscience.com/cmc/v84n2/62920/html", "content": "[26] proposed Message Passing Deep Reinforcement Learning (MPDRL), which uses graph neural network to interact with the network topology environment and extract knowledge through information transmission between links to achieve traffic load balancing [27]."}
+{"idx": 8, "title": "Proceedings of the ACM on Web Conference 2025 | ACM Conferences", "date": "", "ddg_snippet": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration . Mingxuan Song, Pengze Li", "subpage_snippet": "", "source": "dlnext.acm.org", "link": "https://dlnext.acm.org/doi/proceedings/10.1145/3696410?tocHeading=heading13", "content": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration . Mingxuan Song, Pengze Li"}
+{"idx": 9, "title": "Paper Digest: WWW 2025 Papers & Highlights – Paper Digest", "date": "", "ddg_snippet": "AERO : Enhancing Sharding Blockchain Via Deep Reinforcement Learning for Account Migration Related Papers Related Patents Related Grants Related Venues Related Experts View Highlight: We propose AERO , a deep reinforcement learning framework to facilitate efficient account ...", "subpage_snippet": "", "source": "www.paperdigest.org", "link": "https://www.paperdigest.org/2025/04/www-2025-papers-highlights/", "content": "AERO : Enhancing Sharding Blockchain Via Deep Reinforcement Learning for Account Migration Related Papers Related Patents Related Grants Related Venues Related Experts View Highlight: We propose AERO , a deep reinforcement learning framework to facilitate efficient account ..."}

data/sampled_jsons/AERO_Enhancing_Sharding_Blockchain_via_Deep_Reinforcement_Learning_for_Account_Migration_experimenta.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement ...", "date": "", "ddg_snippet": "Blockchain , Sharding , Account migration , Reinforcement learning . ACM Reference Format: Anonymous authors.Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration .", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=WcuXvn3HVk", "content": "Blockchain , Sharding , Account migration , Reinforcement learning . ACM Reference Format: Anonymous authors.Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration ."}
+{"idx": 1, "title": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement ...", "date": "", "ddg_snippet": "Blockchain , Sharding , Account migration , Reinforcement learning .Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration .", "subpage_snippet": "", "source": "zhenxiao.com", "link": "http://zhenxiao.com/papers/WWW_AERO_camera_ready.pdf", "content": "Blockchain , Sharding , Account migration , Reinforcement learning .Sender Shard Transaction Count. AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration ."}
+{"idx": 2, "title": "(PDF) Sharding for Blockchain based Mobile Edge Computing...", "date": "", "ddg_snippet": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration . Blockchain -Based Transformer-Assisted Multi-Agent Reinforcement Learning for Resource Allocation and Computation Offloading in 5G Private Networks.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/363131667_Sharding_for_Blockchain_based_Mobile_Edge_Computing_System_A_Deep_Reinforcement_Learning_Approach", "content": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration . Blockchain -Based Transformer-Assisted Multi-Agent Reinforcement Learning for Resource Allocation and Computation Offloading in 5G Private Networks."}
+{"idx": 3, "title": "Peking University - Cited by 70 - Deep Learning - Google Scholar", "date": "", "ddg_snippet": "AERO : Enhancing sharding blockchain via deep reinforcement learning for account migration .Presto: Optimizing Cross- Shard Transactions in Sharded Blockchain Architecture.", "subpage_snippet": "", "source": "scholar.google.com", "link": "https://scholar.google.com/citations?user=BdT5CzcAAAAJ&hl=en", "content": "AERO : Enhancing sharding blockchain via deep reinforcement learning for account migration .Presto: Optimizing Cross- Shard Transactions in Sharded Blockchain Architecture."}
+{"idx": 4, "title": "MIT 6.S091: Introduction to Deep Reinforcement Learning ... - YouTube", "date": "", "ddg_snippet": "First lecture of MIT course 6.S091: Deep Reinforcement Learning , introducing the fascinating field of Deep RL.", "subpage_snippet": "", "source": "www.youtube.com", "link": "https://www.youtube.com/watch?v=zR11FLZ-O9M", "content": "First lecture of MIT course 6.S091: Deep Reinforcement Learning , introducing the fascinating field of Deep RL."}
+{"idx": 5, "title": "Adaptive Access and Tracking Method for Supply Chain Data Based on...", "date": "", "ddg_snippet": "It's recommended to experiment with different models and settings to find the best fit for you.", "subpage_snippet": "", "source": "pidantuan.com", "link": "https://pidantuan.com/scholar/1862494xs.html", "content": "It's recommended to experiment with different models and settings to find the best fit for you."}
+{"idx": 6, "title": "Unsolved Problems in Blockchain Sharding | by Alexander... | Medium", "date": "", "ddg_snippet": "In the first part of the series we provided motivation for blockchain sharding and discussed some core concepts. In this post we will discuss some more advanced aspects of sharding , including its two biggest unsolved challenges: data availability and data validity. Intro.", "subpage_snippet": "", "source": "medium.com", "link": "https://medium.com/nearprotocol/unsolved-problems-in-blockchain-sharding-2327d6517f43", "content": "In the first part of the series we provided motivation for blockchain sharding and discussed some core concepts. In this post we will discuss some more advanced aspects of sharding , including its two biggest unsolved challenges: data availability and data validity. Intro."}
+{"idx": 7, "title": "Topology-aware and highly generalizable deep reinforcement ...", "date": "", "ddg_snippet": "Deep reinforcement learning for one -warehouse multi-retailer inventory management.A two-stage RNN-based deep reinforcement learning approach for solving the parallel machine scheduling problem with due dates and family setups .", "subpage_snippet": "", "source": "www.springerprofessional.de", "link": "https://www.springerprofessional.de/en/topology-aware-and-highly-generalizable-deep-reinforcement-learn/51427600", "content": "Deep reinforcement learning for one -warehouse multi-retailer inventory management.A two-stage RNN-based deep reinforcement learning approach for solving the parallel machine scheduling problem with due dates and family setups ."}
+{"idx": 8, "title": "Proceedings of the ACM on Web Conference 2025 | ACM Conferences", "date": "", "ddg_snippet": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration . Mingxuan Song, Pengze Li", "subpage_snippet": "", "source": "dlnext.acm.org", "link": "https://dlnext.acm.org/doi/proceedings/10.1145/3696410?tocHeading=heading13", "content": "AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration . Mingxuan Song, Pengze Li"}
+{"idx": 9, "title": "Mingxuan Song", "date": "", "ddg_snippet": "\" AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration .\"\"SPRING: Improving the Throughput of Sharding Blockchain via Deep Reinforcement Learning Based State Placement.\" In Proceedings of the Web Conference, May 2024.", "subpage_snippet": "", "source": "www.songmingxuan.com", "link": "https://www.songmingxuan.com/", "content": "\" AERO : Enhancing Sharding Blockchain via Deep Reinforcement Learning for Account Migration .\"\"SPRING: Improving the Throughput of Sharding Blockchain via Deep Reinforcement Learning Based State Placement.\" In Proceedings of the Web Conference, May 2024."}

data/sampled_jsons/AERO_account_migration_reward_function_CST_IST_shard_blockchain_year_2024.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "(PDF) BrokerChain: A Cross- Shard Blockchain Protocol for...", "date": "", "ddg_snippet": "Account Migration across Blockchain Shards using Fine-tuned Lock Mechanism.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/356789473_BrokerChain_A_Cross-Shard_Blockchain_Protocol_for_AccountBalance-based_State_Sharding", "content": "Account Migration across Blockchain Shards using Fine-tuned Lock Mechanism."}
+{"idx": 1, "title": "Sharding FAQ", "date": "", "ddg_snippet": "Could sharded blockchains do a better job of dealing with network partitions? The schemes described in this document would offer no improvement over non-sharded blockchains ; realistically, every shard would end up with some nodes on both sides of the partition.", "subpage_snippet": "", "source": "vitalik.eth.limo", "link": "https://vitalik.eth.limo/general/2017/12/31/sharding_faq.html", "content": "Could sharded blockchains do a better job of dealing with network partitions? The schemes described in this document would offer no improvement over non-sharded blockchains ; realistically, every shard would end up with some nodes on both sides of the partition."}
+{"idx": 2, "title": "Aeroclub Time", "date": "", "ddg_snippet": "Ваш браузер устарел. Для корректной работы сайта TIME . AERO рекомендуется использовать браузеры.", "subpage_snippet": "", "source": "app.time.aero", "link": "https://app.time.aero/signin", "content": "Ваш браузер устарел. Для корректной работы сайта TIME . AERO рекомендуется использовать браузеры."}
+{"idx": 3, "title": "Manage your Apple Account", "date": "", "ddg_snippet": "Apple Account . Open Menu Close Menu. Sign InSign In.", "subpage_snippet": "", "source": "account.apple.com", "link": "https://account.apple.com/account/manage", "content": "Apple Account . Open Menu Close Menu. Sign InSign In."}
+{"idx": 4, "title": "TON Explorer: Blockchain Analysis", "date": "", "ddg_snippet": "TON Explorer: Track transactions, blocks, addresses, and smart contracts in real- time . A user-friendly tool for analyzing and monitoring activity on decentralized networks.", "subpage_snippet": "", "source": "tonviewer.com", "link": "https://tonviewer.com/", "content": "TON Explorer: Track transactions, blocks, addresses, and smart contracts in real- time . A user-friendly tool for analyzing and monitoring activity on decentralized networks."}
+{"idx": 5, "title": "Rake | Grow a Garden Wiki | Fandom", "date": "", "ddg_snippet": "Atomfall has an interesting take on the post-apocalyptic setting that could set it apart from its many inspirations. Keep Watching. Next video in 8 seconds. 0 seconds of 5 minutes, 58 seconds Volume 0%. Press shift question mark to access a list of k...", "subpage_snippet": "", "source": "growagarden.fandom.com", "link": "https://growagarden.fandom.com/wiki/Rake", "content": "Atomfall has an interesting take on the post-apocalyptic setting that could set it apart from its many inspirations. Keep Watching. Next video in 8 seconds. 0 seconds of 5 minutes, 58 seconds Volume 0%. Press shift question mark to access a list of k..."}
+{"idx": 6, "title": "NEAR Launches Simple Nightshade: The First Step Towards... | NEAR", "date": "", "ddg_snippet": "NEAR is launching phase 0 of its roadmap towards a fully sharded blockchain .Unlocking NEAR to Infinite Scale. After phase 2 is complete, we will have a fully functional sharded mainnet with a fixed number of shards .", "subpage_snippet": "", "source": "www.near.org", "link": "https://www.near.org/blog/near-launches-simple-nightshade-the-first-step-towards-a-sharded-blockchain", "content": "NEAR is launching phase 0 of its roadmap towards a fully sharded blockchain .Unlocking NEAR to Infinite Scale. After phase 2 is complete, we will have a fully functional sharded mainnet with a fixed number of shards ."}
+{"idx": 7, "title": "qutrade.io/ru/?airdrops", "date": "", "ddg_snippet": "BLOCKCHAIN PIVX Algorithm 4.0 POS and Masternode. Admin: Уважаемые трейдеры, 2025-09-15 будет проведен делистинг SHARD ( SHARD ). Просим вывести все ваши SHARD с биржи до даты проведения делистинга.", "subpage_snippet": "", "source": "qutrade.io", "link": "https://qutrade.io/ru/?airdrops", "content": "BLOCKCHAIN PIVX Algorithm 4.0 POS and Masternode. Admin: Уважаемые трейдеры, 2025-09-15 будет проведен делистинг SHARD ( SHARD ). Просим вывести все ваши SHARD с биржи до даты проведения делистинга."}
+{"idx": 8, "title": "Shibarium Cross-Chain Validators: How They Secure the... | OKX UAE", "date": "", "ddg_snippet": "Introduction to Shibarium Cross-Chain Validators. Shibarium, a Layer-2 blockchain solution tailored for the Shiba Inu ecosystem, is transforming how transactions are processed and settled on Ethereum.", "subpage_snippet": "", "source": "www.okx.com", "link": "https://www.okx.com/en-ae/learn/shibarium-cross-chain-validators-network-security", "content": "Introduction to Shibarium Cross-Chain Validators. Shibarium, a Layer-2 blockchain solution tailored for the Shiba Inu ecosystem, is transforming how transactions are processed and settled on Ethereum."}
+{"idx": 9, "title": "All Star Tower Defense X Codes (September 2025) | Beebom", "date": "", "ddg_snippet": "Apart from the All Star Tower Defense X codes, there are multiple ways you can get free gems or skill orbs. The simple way to gather them is through the daily rewards . When you log in for the first time , the game will prompt you to collect the reward .", "subpage_snippet": "", "source": "beebom.com", "link": "https://beebom.com/all-star-tower-defense-codes/", "content": "Apart from the All Star Tower Defense X codes, there are multiple ways you can get free gems or skill orbs. The simple way to gather them is through the daily rewards . When you log in for the first time , the game will prompt you to collect the reward ."}

data/sampled_jsons/AERO_paper_Spring_focuses_more_on_optimizing_state_placement_rather_than_account_migration.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Book by the seat | The Private Jet Experience | Aero ™", "date": "", "ddg_snippet": "A beyond first-class experience awaits, no membership required. Aero effortlessly merges the worlds of hospitality, design, and travel. Enjoy spacious premium seats, private terminals with no lines or crowds, a dedicated concierge team, and curated amenities.", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/", "content": "A beyond first-class experience awaits, no membership required. Aero effortlessly merges the worlds of hospitality, design, and travel. Enjoy spacious premium seats, private terminals with no lines or crowds, a dedicated concierge team, and curated amenities."}
+{"idx": 1, "title": "Seats. aero - Home", "date": "", "ddg_snippet": "Seats. aero is the fastest search engine for award travel. Explore availability across entire regions, search with instant results, create free alerts and more to find the best flights for your points.", "subpage_snippet": "", "source": "seats.aero", "link": "https://seats.aero/", "content": "Seats. aero is the fastest search engine for award travel. Explore availability across entire regions, search with instant results, create free alerts and more to find the best flights for your points."}
+{"idx": 2, "title": "AERO Definition & Meaning - Merriam-Webster", "date": "", "ddg_snippet": "The meaning of AERO is of or relating to aircraft or aeronautics. How to use aero in a sentence.", "subpage_snippet": "", "source": "www.merriam-webster.com", "link": "https://www.merriam-webster.com/dictionary/aero", "content": "The meaning of AERO is of or relating to aircraft or aeronautics. How to use aero in a sentence."}
+{"idx": 3, "title": "Mil-Spec AR Parts, Components & Accessories | Aero Precision", "date": "", "ddg_snippet": "Aero Precision manufacturers mil-spec parts, including AR15 & AR10 uppers, lowers, rifles, handguards, barrels, scope mounts & more.", "subpage_snippet": "", "source": "www.aeroprecisionusa.com", "link": "https://www.aeroprecisionusa.com/", "content": "Aero Precision manufacturers mil-spec parts, including AR15 & AR10 uppers, lowers, rifles, handguards, barrels, scope mounts & more."}
+{"idx": 4, "title": "Aero End of Support FAQs - Adobe Inc.", "date": "", "ddg_snippet": "Aug 7, 2025 · Adobe Aero will be discontinued on iOS, Android, and Creative Cloud Desktop effective November 6, 2025. Existing users can access the application and download their content until December 3, 2025.", "subpage_snippet": "", "source": "helpx.adobe.com", "link": "https://helpx.adobe.com/aero/aero-end-of-support-faq.html", "content": "Aug 7, 2025 · Adobe Aero will be discontinued on iOS, Android, and Creative Cloud Desktop effective November 6, 2025. Existing users can access the application and download their content until December 3, 2025."}
+{"idx": 5, "title": "Trailer Tarp Systems, Accessories | Aero Industries, Inc.", "date": "", "ddg_snippet": "Aero Industries, Inc. is a global leader in the manufacturing of trailer tarp systems and accessories. Founded in 1944, the company’s heritage is rooted in customer service and innovation.", "subpage_snippet": "", "source": "www.aeroindustries.com", "link": "https://www.aeroindustries.com/", "content": "Aero Industries, Inc. is a global leader in the manufacturing of trailer tarp systems and accessories. Founded in 1944, the company’s heritage is rooted in customer service and innovation."}
+{"idx": 6, "title": "Aircraft Fleet | Aero ™", "date": "", "ddg_snippet": "\"I’ve done a lot of traveling in my life, and I’ve never experienced transportation quite like Aero . Exceptional service, luxurious terminal, high quality amenities, top notch staff.\"", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/fleet", "content": "\"I’ve done a lot of traveling in my life, and I’ve never experienced transportation quite like Aero . Exceptional service, luxurious terminal, high quality amenities, top notch staff.\""}
+{"idx": 7, "title": "Explore Flights from Los Angeles | Aero ™", "date": "", "ddg_snippet": "Aero is a premium jet service that provides the time-saving convenience and world-class service of private air travel, booked by the seat with no membership required.", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/flights?adults=1&infants=0&petInSeat=0&petUnderSeat=0&serviceAnimal=0&type=roundtrip", "content": "Aero is a premium jet service that provides the time-saving convenience and world-class service of private air travel, booked by the seat with no membership required."}
+{"idx": 8, "title": "Our Story | The Private Jet Experience | Aero ™", "date": "", "ddg_snippet": "Aero was created to bring the magic back to flying. We aim to deliver an unforgettable, radically better travel experience. Learn more about how we started, our leadership, and our vision for the future.", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/our-story", "content": "Aero was created to bring the magic back to flying. We aim to deliver an unforgettable, radically better travel experience. Learn more about how we started, our leadership, and our vision for the future."}
+{"idx": 9, "title": "Flights To and From Los Angeles | The Private Jet Experience | ...", "date": "", "ddg_snippet": "Wherever you choose to wander, fly in signature Aero style. Aero ’s book-by-the-seat jet service connects travelers in Los Angeles to sought-after leisure destinations and the world's largest entertainment and sporting events.", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/destinations/na/los-angeles", "content": "Wherever you choose to wander, fly in signature Aero style. Aero ’s book-by-the-seat jet service connects travelers in Los Angeles to sought-after leisure destinations and the world's largest entertainment and sporting events."}

data/sampled_jsons/AERO_paper_WWW_2025_blockchain_sharding_reinforcement_learning.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Book by the seat | The Private Jet Experience | Aero ™", "date": "", "ddg_snippet": "A beyond first-class experience awaits, no membership required. Aero effortlessly merges the worlds of hospitality, design, and travel. Enjoy spacious premium seats, private terminals with no lines or crowds, a dedicated concierge team, and curated amenities.", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/", "content": "A beyond first-class experience awaits, no membership required. Aero effortlessly merges the worlds of hospitality, design, and travel. Enjoy spacious premium seats, private terminals with no lines or crowds, a dedicated concierge team, and curated amenities."}
+{"idx": 1, "title": "Seats. aero - Home", "date": "", "ddg_snippet": "Seats. aero is the fastest search engine for award travel. Explore availability across entire regions, search with instant results, create free alerts and more to find the best flights for your points.", "subpage_snippet": "", "source": "seats.aero", "link": "https://seats.aero/", "content": "Seats. aero is the fastest search engine for award travel. Explore availability across entire regions, search with instant results, create free alerts and more to find the best flights for your points."}
+{"idx": 2, "title": "AERO Definition & Meaning - Merriam-Webster", "date": "", "ddg_snippet": "The meaning of AERO is of or relating to aircraft or aeronautics. How to use aero in a sentence.", "subpage_snippet": "", "source": "www.merriam-webster.com", "link": "https://www.merriam-webster.com/dictionary/aero", "content": "The meaning of AERO is of or relating to aircraft or aeronautics. How to use aero in a sentence."}
+{"idx": 3, "title": "Mil-Spec AR Parts, Components & Accessories | Aero Precision", "date": "", "ddg_snippet": "Aero Precision manufacturers mil-spec parts, including AR15 & AR10 uppers, lowers, rifles, handguards, barrels, scope mounts & more.", "subpage_snippet": "", "source": "www.aeroprecisionusa.com", "link": "https://www.aeroprecisionusa.com/", "content": "Aero Precision manufacturers mil-spec parts, including AR15 & AR10 uppers, lowers, rifles, handguards, barrels, scope mounts & more."}
+{"idx": 4, "title": "Aero End of Support FAQs - Adobe Inc.", "date": "", "ddg_snippet": "Aug 7, 2025 · Adobe Aero will be discontinued on iOS, Android, and Creative Cloud Desktop effective November 6, 2025. Existing users can access the application and download their content until December 3, 2025.", "subpage_snippet": "", "source": "helpx.adobe.com", "link": "https://helpx.adobe.com/aero/aero-end-of-support-faq.html", "content": "Aug 7, 2025 · Adobe Aero will be discontinued on iOS, Android, and Creative Cloud Desktop effective November 6, 2025. Existing users can access the application and download their content until December 3, 2025."}
+{"idx": 5, "title": "Trailer Tarp Systems, Accessories | Aero Industries, Inc.", "date": "", "ddg_snippet": "Aero Industries, Inc. is a global leader in the manufacturing of trailer tarp systems and accessories. Founded in 1944, the company’s heritage is rooted in customer service and innovation.", "subpage_snippet": "", "source": "www.aeroindustries.com", "link": "https://www.aeroindustries.com/", "content": "Aero Industries, Inc. is a global leader in the manufacturing of trailer tarp systems and accessories. Founded in 1944, the company’s heritage is rooted in customer service and innovation."}
+{"idx": 6, "title": "Aircraft Fleet | Aero ™", "date": "", "ddg_snippet": "\"I’ve done a lot of traveling in my life, and I’ve never experienced transportation quite like Aero . Exceptional service, luxurious terminal, high quality amenities, top notch staff.\"", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/fleet", "content": "\"I’ve done a lot of traveling in my life, and I’ve never experienced transportation quite like Aero . Exceptional service, luxurious terminal, high quality amenities, top notch staff.\""}
+{"idx": 7, "title": "Explore Flights from Los Angeles | Aero ™", "date": "", "ddg_snippet": "Aero is a premium jet service that provides the time-saving convenience and world-class service of private air travel, booked by the seat with no membership required.", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/flights?adults=1&infants=0&petInSeat=0&petUnderSeat=0&serviceAnimal=0&type=roundtrip", "content": "Aero is a premium jet service that provides the time-saving convenience and world-class service of private air travel, booked by the seat with no membership required."}
+{"idx": 8, "title": "Our Story | The Private Jet Experience | Aero ™", "date": "", "ddg_snippet": "Aero was created to bring the magic back to flying. We aim to deliver an unforgettable, radically better travel experience. Learn more about how we started, our leadership, and our vision for the future.", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/our-story", "content": "Aero was created to bring the magic back to flying. We aim to deliver an unforgettable, radically better travel experience. Learn more about how we started, our leadership, and our vision for the future."}
+{"idx": 9, "title": "Flights To and From Los Angeles | The Private Jet Experience | ...", "date": "", "ddg_snippet": "Wherever you choose to wander, fly in signature Aero style. Aero ’s book-by-the-seat jet service connects travelers in Los Angeles to sought-after leisure destinations and the world's largest entertainment and sporting events.", "subpage_snippet": "", "source": "aero.com", "link": "https://aero.com/destinations/na/los-angeles", "content": "Wherever you choose to wander, fly in signature Aero style. Aero ’s book-by-the-seat jet service connects travelers in Los Angeles to sought-after leisure destinations and the world's largest entertainment and sporting events."}

data/sampled_jsons/AI-Driven_Cyberattacks_CVE-Bench_Figure_3_T-Agent_Success@5_year_2024.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "[2503.17332] CVE-Bench: A Benchmark for AI Agents' Ability to ... GitHub - uiuc-kang-lab/cve-bench: CVE-Bench: A Benchmark for ... AI-Driven Cyberattacks are on the Rise. Are You Ready? uiuc-kang-lab/cve-bench | DeepWiki [PDF] CVE-Bench: A Benchmark for AI Agents' Ability to ... Measuring AI Agents’ Ability to Exploit Web Applications The impact of AI on cybersecurity | McKinsey", "date": "", "ddg_snippet": "Mar 21, 2025 · Large language model (LLM) agents are increasingly capable of autonomously conducting cyberattacks , posing significant threats to existing applications. This growing risk highlights the urgent need for a real-world benchmark to evaluate the ability of LLM agents to exploit web application vulnerabilities. However, existing benchmarks fall short as they are limited to abstracted Capture the ... Apr 24, 2025 · This repository contains data and code used in the CVE-Bench (paper, blog), which is for evaluating AI agents on real world web vulnerabilities and exploits collected from National Vulnerability Database. CVE-Bench includes 40 critical-severity Common Vulnerability and Exposures ( CVE ) with the reference automatic exploits available on requests. Jul 22, 2025 · In a paper titled CVE-Bench : A Benchmark for AI Agents’ Ability to Exploit Real-World Web Application Vulnerabilities, researchers found that LLM agents are “increasingly capable of autonomously conducting cyberattacks ,” and that this risk presents “the urgent need for a real-world benchmark to evaluate the ability of LLM agents to ... May 12, 2025 · CVE-Bench is a benchmark that contains 40 critical-severity Common Vulnerability and Exposures (CVEs) collected from the National Vulnerability Database. It creates reproducible environments for testing AI agents' abilities to discover and exploit web application vulnerabilities. Mar 21, 2025 · CVE-Bench is introduced, a real-world cybersecurity benchmark based on critical-severity Common Vulnerabilities and Exposures that enables LLM agents to exploit vulnerable web applications in scenarios that mimic real-world conditions, while also providing effective evaluation of their exploits. Large language model (LLM) agents are increasingly capable of autonomously conducting cyberattacks ... Mar 31, 2025 · Success rates of different AI agents on CVE-bench in the zero-day or one-day setting. As shown, AI agents successfully exploited up to 13% of web application vulnerabilities in the zero-day ... Nov 14, 2024 · The rapid advancement of AI and generative AI (gen AI ) is fundamentally transforming the cybersecurity landscape, presenting both opportunities and challenges for cybersecurity providers. As more organizations in both the private and public sectors use AI to enhance their operations, they risk inadvertently introducing new cyber-related threats. This is creating a significant and growing ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2503.17332", "content": "Mar 21, 2025 · Large language model (LLM) agents are increasingly capable of autonomously conducting cyberattacks , posing significant threats to existing applications. This growing risk highlights the urgent need for a real-world benchmark to evaluate the ability of LLM agents to exploit web application vulnerabilities. However, existing benchmarks fall short as they are limited to abstracted Capture the ... Apr 24, 2025 · This repository contains data and code used in the CVE-Bench (paper, blog), which is for evaluating AI agents on real world web vulnerabilities and exploits collected from National Vulnerability Database. CVE-Bench includes 40 critical-severity Common Vulnerability and Exposures ( CVE ) with the reference automatic exploits available on requests. Jul 22, 2025 · In a paper titled CVE-Bench : A Benchmark for AI Agents’ Ability to Exploit Real-World Web Application Vulnerabilities, researchers found that LLM agents are “increasingly capable of autonomously conducting cyberattacks ,” and that this risk presents “the urgent need for a real-world benchmark to evaluate the ability of LLM agents to ... May 12, 2025 · CVE-Bench is a benchmark that contains 40 critical-severity Common Vulnerability and Exposures (CVEs) collected from the National Vulnerability Database. It creates reproducible environments for testing AI agents' abilities to discover and exploit web application vulnerabilities. Mar 21, 2025 · CVE-Bench is introduced, a real-world cybersecurity benchmark based on critical-severity Common Vulnerabilities and Exposures that enables LLM agents to exploit vulnerable web applications in scenarios that mimic real-world conditions, while also providing effective evaluation of their exploits. Large language model (LLM) agents are increasingly capable of autonomously conducting cyberattacks ... Mar 31, 2025 · Success rates of different AI agents on CVE-bench in the zero-day or one-day setting. As shown, AI agents successfully exploited up to 13% of web application vulnerabilities in the zero-day ... Nov 14, 2024 · The rapid advancement of AI and generative AI (gen AI ) is fundamentally transforming the cybersecurity landscape, presenting both opportunities and challenges for cybersecurity providers. As more organizations in both the private and public sectors use AI to enhance their operations, they risk inadvertently introducing new cyber-related threats. This is creating a significant and growing ..."}
+{"idx": 1, "title": "GitHub - uiuc-kang-lab/cve-bench: CVE-Bench: A Benchmark for ...", "date": "", "ddg_snippet": "Apr 24, 2025 · This repository contains data and code used in the CVE-Bench (paper, blog), which is for evaluating AI agents on real world web vulnerabilities and exploits collected from National Vulnerability Database. CVE-Bench includes 40 critical-severity Common Vulnerability and Exposures ( CVE ) with the reference automatic exploits available on requests.", "subpage_snippet": "", "source": "github.com", "link": "https://github.com/uiuc-kang-lab/cve-bench", "content": "Apr 24, 2025 · This repository contains data and code used in the CVE-Bench (paper, blog), which is for evaluating AI agents on real world web vulnerabilities and exploits collected from National Vulnerability Database. CVE-Bench includes 40 critical-severity Common Vulnerability and Exposures ( CVE ) with the reference automatic exploits available on requests."}
+{"idx": 2, "title": "AI-Driven Cyberattacks are on the Rise. Are You Ready?", "date": "", "ddg_snippet": "Jul 22, 2025 · In a paper titled CVE-Bench : A Benchmark for AI Agents’ Ability to Exploit Real-World Web Application Vulnerabilities, researchers found that LLM agents are “increasingly capable of autonomously conducting cyberattacks ,” and that this risk presents “the urgent need for a real-world benchmark to evaluate the ability of LLM agents to ...", "subpage_snippet": "", "source": "lsvp.com", "link": "https://lsvp.com/stories/ai-enabled-hacking-is-here-are-we-ready-for-it/", "content": "Jul 22, 2025 · In a paper titled CVE-Bench : A Benchmark for AI Agents’ Ability to Exploit Real-World Web Application Vulnerabilities, researchers found that LLM agents are “increasingly capable of autonomously conducting cyberattacks ,” and that this risk presents “the urgent need for a real-world benchmark to evaluate the ability of LLM agents to ..."}
+{"idx": 3, "title": "uiuc-kang-lab/cve-bench | DeepWiki", "date": "", "ddg_snippet": "May 12, 2025 · CVE-Bench is a benchmark that contains 40 critical-severity Common Vulnerability and Exposures (CVEs) collected from the National Vulnerability Database. It creates reproducible environments for testing AI agents' abilities to discover and exploit web application vulnerabilities.", "subpage_snippet": "", "source": "deepwiki.com", "link": "https://deepwiki.com/uiuc-kang-lab/cve-bench/1-overview", "content": "May 12, 2025 · CVE-Bench is a benchmark that contains 40 critical-severity Common Vulnerability and Exposures (CVEs) collected from the National Vulnerability Database. It creates reproducible environments for testing AI agents' abilities to discover and exploit web application vulnerabilities."}
+{"idx": 4, "title": "[PDF] CVE-Bench: A Benchmark for AI Agents' Ability to ...", "date": "", "ddg_snippet": "Mar 21, 2025 · CVE-Bench is introduced, a real-world cybersecurity benchmark based on critical-severity Common Vulnerabilities and Exposures that enables LLM agents to exploit vulnerable web applications in scenarios that mimic real-world conditions, while also providing effective evaluation of their exploits. Large language model (LLM) agents are increasingly capable of autonomously conducting cyberattacks ...", "subpage_snippet": "", "source": "www.semanticscholar.org", "link": "https://www.semanticscholar.org/paper/CVE-Bench:-A-Benchmark-for-AI-Agents'-Ability-to-Zhu-Kellermann/095b31dfaa032a2daf13da21bd4d04dddb2097fa", "content": "Mar 21, 2025 · CVE-Bench is introduced, a real-world cybersecurity benchmark based on critical-severity Common Vulnerabilities and Exposures that enables LLM agents to exploit vulnerable web applications in scenarios that mimic real-world conditions, while also providing effective evaluation of their exploits. Large language model (LLM) agents are increasingly capable of autonomously conducting cyberattacks ..."}
+{"idx": 5, "title": "Measuring AI Agents’ Ability to Exploit Web Applications", "date": "", "ddg_snippet": "Mar 31, 2025 · Success rates of different AI agents on CVE-bench in the zero-day or one-day setting. As shown, AI agents successfully exploited up to 13% of web application vulnerabilities in the zero-day ...", "subpage_snippet": "", "source": "medium.com", "link": "https://medium.com/@danieldkang/measuring-ai-agents-ability-to-exploit-web-applications-ba4225aa281f", "content": "Mar 31, 2025 · Success rates of different AI agents on CVE-bench in the zero-day or one-day setting. As shown, AI agents successfully exploited up to 13% of web application vulnerabilities in the zero-day ..."}
+{"idx": 6, "title": "Establishing Best Practices for Building Rigorous Agentic Benchmarks", "date": "", "ddg_snippet": "CVE - Bench is a benchmark for evaluating AI agents ’ ability to exploit real-world web vulnerabilities under one- or zero-day scenarios [96] . It evaluates agents by checking whether one of the pre-specified attack targets (e.g., denial of service) is accomplished.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2507.02825v4", "content": "CVE - Bench is a benchmark for evaluating AI agents ’ ability to exploit real-world web vulnerabilities under one- or zero-day scenarios [96] . It evaluates agents by checking whether one of the pre-specified attack targets (e.g., denial of service) is accomplished."}
+{"idx": 7, "title": "How AI - Driven Cyberattacks Are Changing the Landscape... | Clutch.co", "date": "", "ddg_snippet": "AI is constantly changing the online landscape. Discover how AI - driven threats impact businesses and how to best respond to protect your business.", "subpage_snippet": "", "source": "clutch.co", "link": "https://clutch.co/resources/how-ai-driven-cyberattacks-are-changing-cybersecurity", "content": "AI is constantly changing the online landscape. Discover how AI - driven threats impact businesses and how to best respond to protect your business."}
+{"idx": 8, "title": "Malicious Listeners in Ivanti EPMM: How CVE -2025-4427 and...", "date": "", "ddg_snippet": "Ivanti EPMM CVE -2025-4427 and CVE -2025-4428 vulnerabilities are exploited in the wild. Learn how attackers used them to deploy malware and create persistent backdoors.", "subpage_snippet": "", "source": "www.picussecurity.com", "link": "https://www.picussecurity.com/resource/blog/malicious-listeners-ivanti-epmm-cve-2025-4427-cve-2025-4428-backdoors", "content": "Ivanti EPMM CVE -2025-4427 and CVE -2025-4428 vulnerabilities are exploited in the wild. Learn how attackers used them to deploy malware and create persistent backdoors."}
+{"idx": 9, "title": "Earth Simnavaz Cyberattacks : Exploiting CVE -2024-30088 to Target...", "date": "", "ddg_snippet": "Leveraging vulnerabilities such as CVE -2024-30088, Earth Simnavaz is executing sophisticated cyberattacks that employ stealthy backdoors and advanced techniques to maintain persistent access to compromised systems.", "subpage_snippet": "", "source": "guardiansofcyber.com", "link": "https://guardiansofcyber.com/cybersecurity-news/earth-simnavaz-cyberattacks-exploiting-cve-2024-30088-to-target-uae-critical-infrastructure-with-stealthy-backdoors/", "content": "Leveraging vulnerabilities such as CVE -2024-30088, Earth Simnavaz is executing sophisticated cyberattacks that employ stealthy backdoors and advanced techniques to maintain persistent access to compromised systems."}

data/sampled_jsons/ATA_Adaptive_Task_Allocation_Theorem_6.1_6.3_regret_upper_bound_logarithmic_sitearxiv.org_year_2024.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "ATA: Adaptive Task Allocation for Efficient Resource Management in ...", "date": "", "ddg_snippet": "4 Adaptive Task Allocation 4.1 Reduction to Multi-Armed Bandit and Proxy Loss 4.2 Comparison with the combinatorial bandits setting 4.3 Adaptive Task Allocation Algorithm 4.4 Upper-bound on the total computation time 5 Empirical Adaptive Task Allocation 6 Theoretical Results", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2502.00775v1", "content": "4 Adaptive Task Allocation 4.1 Reduction to Multi-Armed Bandit and Proxy Loss 4.2 Comparison with the combinatorial bandits setting 4.3 Adaptive Task Allocation Algorithm 4.4 Upper-bound on the total computation time 5 Empirical Adaptive Task Allocation 6 Theoretical Results"}
+{"idx": 1, "title": "ATA: Adaptive Task Allocation for Eficient Resource Management in ...", "date": "", "ddg_snippet": "In this paper, we propose ATA ( Adaptive Task Allocation ), a method that adapts to het-erogeneous and random distributions of worker computation times. Through rigorous theoretical analysis, we show that ATA identifies the optimal task allocation and performs comparably to meth-ods with prior knowledge of computation times.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2502.00775", "content": "In this paper, we propose ATA ( Adaptive Task Allocation ), a method that adapts to het-erogeneous and random distributions of worker computation times. Through rigorous theoretical analysis, we show that ATA identifies the optimal task allocation and performs comparably to meth-ods with prior knowledge of computation times."}
+{"idx": 2, "title": "Logarithmic Regret for Reinforcement Learning with Linear Function ...", "date": "", "ddg_snippet": "The regret bound in Theorem 4.4 is independent of the size of the state space S, action space A, and is only logarithmic in the number of steps T , which suggests that Algorithm 1 is sample e cient for MDPs with large state and action spaces.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2011.11566", "content": "The regret bound in Theorem 4.4 is independent of the size of the state space S, action space A, and is only logarithmic in the number of steps T , which suggests that Algorithm 1 is sample e cient for MDPs with large state and action spaces."}
+{"idx": 3, "title": "Introduction to Multi-Armed Bandits - arXiv.org", "date": "", "ddg_snippet": "It is instructive to derive Theorem 1.10 in a different way: starting from the logarithmic regret bound in (1.10). Informally, we need to get rid of arbitrarily small gaps (a) in the denominator.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/1904.07272v6", "content": "It is instructive to derive Theorem 1.10 in a different way: starting from the logarithmic regret bound in (1.10). Informally, we need to get rid of arbitrarily small gaps (a) in the denominator."}
+{"idx": 4, "title": "ATA: Adaptive Task Allocation for Efficient Resource Management in ...", "date": "", "ddg_snippet": "4 Adaptive Task Allocation 4.1 Reduction to Multi-Armed Bandit and Proxy Loss 4.2 Comparison with the Combinatorial Bandits Setting 4.3 Adaptive Task Allocation Algorithm 4.4 Upper-Bound on the Total Computation Time 5 Empirical Adaptive Task Allocation 6 Theoretical Results", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2502.00775v2", "content": "4 Adaptive Task Allocation 4.1 Reduction to Multi-Armed Bandit and Proxy Loss 4.2 Comparison with the Combinatorial Bandits Setting 4.3 Adaptive Task Allocation Algorithm 4.4 Upper-Bound on the Total Computation Time 5 Empirical Adaptive Task Allocation 6 Theoretical Results"}
+{"idx": 5, "title": "arXiv:1810.03825v2 [stat.ML] 14 Oct 2018", "date": "", "ddg_snippet": "Abstract We develop a new theoretical framework, the envelope complexity, to analyze the min-imax regret with logarithmic loss functions and derive a Bayesian predictor that adap-tively achieves the minimax regret over high-dimensional `1-balls within a factor of two. The prior is newly derived for achieving the minimax regret and called the spike-and-tails (ST) prior as it looks like. The ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/1810.03825", "content": "Abstract We develop a new theoretical framework, the envelope complexity, to analyze the min-imax regret with logarithmic loss functions and derive a Bayesian predictor that adap-tively achieves the minimax regret over high-dimensional `1-balls within a factor of two. The prior is newly derived for achieving the minimax regret and called the spike-and-tails (ST) prior as it looks like. The ..."}
+{"idx": 6, "title": "arXiv:2007.10229v3 [cs.LG] 23 Sep 2021", "date": "", "ddg_snippet": "Approximation Markov Bandit Algorithm. We rst prove that logarithmic regret bounds are achievable for a suitably small learning rate which depends on t e gap between mean reward of each arm. We then modify this algorithm to remove the dependence on the gap and prove a O(log(T)2) regret bound an", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2007.10229", "content": "Approximation Markov Bandit Algorithm. We rst prove that logarithmic regret bounds are achievable for a suitably small learning rate which depends on t e gap between mean reward of each arm. We then modify this algorithm to remove the dependence on the gap and prove a O(log(T)2) regret bound an"}
+{"idx": 7, "title": "arXiv:2412.06126v1 [math.ST] 9 Dec 2024", "date": "", "ddg_snippet": "Abstract. Upper Confidence Bound (UCB) algorithms are a widely-used class of sequential algorithms for the K-armed bandit problem. Despite extensive re-search over the past decades aimed at understanding their asymptotic and (near) minimax optimality properties, a precise understanding of their regret behavior remains elusive. This gap has not only hindered the evaluation of their actual al ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2412.06126", "content": "Abstract. Upper Confidence Bound (UCB) algorithms are a widely-used class of sequential algorithms for the K-armed bandit problem. Despite extensive re-search over the past decades aimed at understanding their asymptotic and (near) minimax optimality properties, a precise understanding of their regret behavior remains elusive. This gap has not only hindered the evaluation of their actual al ..."}
+{"idx": 8, "title": "On Regret with Multiple Best Arms - arXiv.org", "date": "", "ddg_snippet": "We provide an adaptive algorithm that is agnostic to the unknown number of best arms in Section 3, and theoretically derive its regret bound . In Section 4, we prove a lower bound for our problem setting that indicates that there is no algorithm that can be optimal simultaneously over all hardness levels.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2006.14785", "content": "We provide an adaptive algorithm that is agnostic to the unknown number of best arms in Section 3, and theoretically derive its regret bound . In Section 4, we prove a lower bound for our problem setting that indicates that there is no algorithm that can be optimal simultaneously over all hardness levels."}
+{"idx": 9, "title": "Hedging the Drift: Learning to Optimize under Non-Stationarity", "date": "", "ddg_snippet": "Tis known, we characterize the lower bound of dynamic regret , and develop a tuned Sliding Window Upper -Con dence- Bound (SW-UCB) algorithm with matched dynamic regret upper bound up to logarithmic factors.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/1903.01461v1", "content": "Tis known, we characterize the lower bound of dynamic regret , and develop a tuned Sliding Window Upper -Con dence- Bound (SW-UCB) algorithm with matched dynamic regret upper bound up to logarithmic factors."}

data/sampled_jsons/ATA_Adaptive_Task_Allocation_for_Efficient_Resource_Management_in_Distributed_Machine_Learning_arxiv.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "[2502.00775] ATA: Adaptive Task Allocation for Efficient ...", "date": "", "ddg_snippet": "by A Maranjyan · 2025 · Cited by 3 — Abstract page for arXiv paper 2502.00775: ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning .", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2502.00775", "content": "by A Maranjyan · 2025 · Cited by 3 — Abstract page for arXiv paper 2502.00775: ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning ."}
+{"idx": 1, "title": "ATA: Adaptive Task Allocation for Efficient Resource ...", "date": "", "ddg_snippet": "2 Feb 2025 — Through rigorous theoretical analysis, we show that ATA identifies the optimal task allocation and performs comparably to methods with prior ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2502.00775v1", "content": "2 Feb 2025 — Through rigorous theoretical analysis, we show that ATA identifies the optimal task allocation and performs comparably to methods with prior ..."}
+{"idx": 2, "title": "ATA: Adaptive Task Allocation for Efficient Resource ...", "date": "", "ddg_snippet": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning ... arXiv preprint arXiv :1604.00981, 2016a. Chen, W., Wang ...", "subpage_snippet": "", "source": "icml.cc", "link": "https://icml.cc/virtual/2025/poster/46650", "content": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning ... arXiv preprint arXiv :1604.00981, 2016a. Chen, W., Wang ..."}
+{"idx": 3, "title": "cherryATA", "date": "", "ddg_snippet": "arXiv :2502.00775v1 [cs.LG] 2 Feb 2025. Page 2. ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . 1.1. A ...", "subpage_snippet": "", "source": "www.arxiv.org", "link": "https://www.arxiv.org/pdf/2502.00775v1", "content": "arXiv :2502.00775v1 [cs.LG] 2 Feb 2025. Page 2. ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . 1.1. A ..."}
+{"idx": 4, "title": "ATA: Adaptive Task Allocation for Efficient Resource ...", "date": "", "ddg_snippet": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . stat.MLcs.LG03 Feb 2025. Artavazd Maranjyan, El Mehdi Saad ...", "subpage_snippet": "", "source": "chatpaper.com", "link": "https://chatpaper.com/fr/chatpaper/paper/104771", "content": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . stat.MLcs.LG03 Feb 2025. Artavazd Maranjyan, El Mehdi Saad ..."}
+{"idx": 5, "title": "Arto Maranjyan on X: \"New paper out “ATA: Adaptive Task Allocation ...", "date": "", "ddg_snippet": "New paper out “ ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning ”. Co-authored with @emsaad_p ...", "subpage_snippet": "", "source": "x.com", "link": "https://x.com/ArtoMaranjyan/status/1886755918622192022", "content": "New paper out “ ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning ”. Co-authored with @emsaad_p ..."}
+{"idx": 6, "title": "Adaptive Task Allocation for Efficient Resource Management in ...", "date": "", "ddg_snippet": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . Published 2/3/2025 by Artavazd Maranjyan, El Mehdi Saad ...", "subpage_snippet": "", "source": "www.aimodels.fyi", "link": "https://www.aimodels.fyi/papers/arxiv/ata-adaptive-task-allocation-efficient-resource-management", "content": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . Published 2/3/2025 by Artavazd Maranjyan, El Mehdi Saad ..."}
+{"idx": 7, "title": "ATA: Adaptive Task Allocation for Efficient Resource ...", "date": "", "ddg_snippet": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . stat.MLcs.LG03 Feb 2025. Artavazd Maranjyan, El Mehdi Saad ...", "subpage_snippet": "", "source": "chatpaper.com", "link": "https://chatpaper.com/ja/chatpaper/paper/104771", "content": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . stat.MLcs.LG03 Feb 2025. Artavazd Maranjyan, El Mehdi Saad ..."}
+{"idx": 8, "title": "Arto Maranjyan", "date": "", "ddg_snippet": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . A Maranjyan, EM Saad, P Richtárik, F Orabona. arXiv preprint ...", "subpage_snippet": "", "source": "scholar.google.com", "link": "https://scholar.google.com/citations?user=93WEFj8AAAAJ&hl=en", "content": "ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning . A Maranjyan, EM Saad, P Richtárik, F Orabona. arXiv preprint ..."}
+{"idx": 9, "title": "Arto Maranjyan's Post", "date": "", "ddg_snippet": "... ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning ”, coauthored with El Mehdi SAAD, Peter ...", "subpage_snippet": "", "source": "www.linkedin.com", "link": "https://www.linkedin.com/posts/arto-maranjyan_ata-adaptive-task-allocation-for-efficient-activity-7292514927922036736-zgWC", "content": "... ATA: Adaptive Task Allocation for Efficient Resource Management in Distributed Machine Learning ”, coauthored with El Mehdi SAAD, Peter ..."}

data/sampled_jsons/A_Checks-and-Balances_Framework_Edward_Chang_dataset_empirical_evaluation_experiment.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "[2502.00136] A Checks-and-Balances Framework for Context ... A Checks-and-Balances Framework for Context-Aware Ethical AI ... A Three-Branch Checks-and-Balances ... - Semantic Scholar Edward Y. Chang - Stanford University A Three-Branch Checks-and-Balances Framework for Context ... A Checks-and-Balances Framework for Context-Aware Ethical AI ... A Three-Branch Checks-and-Balances Frameworkfor Context-Aware ...", "date": "", "ddg_snippet": "Jan 31, 2025 · View a PDF of the paper titled A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment, by Edward Y. Chang To overcome these challenges, we propose a checks-and-balances framework inspired by governmental structures, where independent but interacting components maintain ac-countability and balance. A three-branch checks - and - balances framework for ethical alignment of Large Language Models, inspired by governmental systems, demonstrates how emotional modeling can guide linguistic behaviors toward ethical outcomes while preserving independence across knowledge generation, ethical oversight, and contextual interpretation. Expand [PDF] Semantic Reader Save to Library Create Alert Cite Edward Y. Chang Stanford Inforlab Technical Report, June 2025 PDF SagaLLM: Context Management, Validation, and Transaction Guarantees for Multi-Agent LLM Planning Edward Y. Chang and Longling Geng VLDB, September 2025 (accepted) PDF | BibTex A Checks-and-Balances Framework for Ethical AI Alignment Edward Y. Chang ICML, July 2025 (accepted) PDF ... Abstract This paper introduces a three-branch checks - and - balances framework for ethical alignment of Large Language Models (LLMs), inspired by governmental systems. It implements three independent yet interacting components: LLMs as the ex-ecutive branch for knowledge generation, DIKE (the goddess of justice) as the legislative branch establishing ethical guardrails, and ERIS (the goddess of ... A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment Susceptible to social biases Vulnerable to reward hacking “Whack-A-Mole” reactive approach Catastrophic forgetting issues Can AI be ethically aligned using a government-inspired system of checks and balances? A Three-Branch Checks-and-Balances Frameworkfor Context-Aware Ethical Alignment of Large Language Models Published 2/3/2025 by Edward Y. Chang", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/abs/2502.00136", "content": "Jan 31, 2025 · View a PDF of the paper titled A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment, by Edward Y. Chang To overcome these challenges, we propose a checks-and-balances framework inspired by governmental structures, where independent but interacting components maintain ac-countability and balance. A three-branch checks - and - balances framework for ethical alignment of Large Language Models, inspired by governmental systems, demonstrates how emotional modeling can guide linguistic behaviors toward ethical outcomes while preserving independence across knowledge generation, ethical oversight, and contextual interpretation. Expand [PDF] Semantic Reader Save to Library Create Alert Cite Edward Y. Chang Stanford Inforlab Technical Report, June 2025 PDF SagaLLM: Context Management, Validation, and Transaction Guarantees for Multi-Agent LLM Planning Edward Y. Chang and Longling Geng VLDB, September 2025 (accepted) PDF | BibTex A Checks-and-Balances Framework for Ethical AI Alignment Edward Y. Chang ICML, July 2025 (accepted) PDF ... Abstract This paper introduces a three-branch checks - and - balances framework for ethical alignment of Large Language Models (LLMs), inspired by governmental systems. It implements three independent yet interacting components: LLMs as the ex-ecutive branch for knowledge generation, DIKE (the goddess of justice) as the legislative branch establishing ethical guardrails, and ERIS (the goddess of ... A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment Susceptible to social biases Vulnerable to reward hacking “Whack-A-Mole” reactive approach Catastrophic forgetting issues Can AI be ethically aligned using a government-inspired system of checks and balances? A Three-Branch Checks-and-Balances Frameworkfor Context-Aware Ethical Alignment of Large Language Models Published 2/3/2025 by Edward Y. Chang"}
+{"idx": 1, "title": "A Checks-and-Balances Framework for Context-Aware Ethical AI ...", "date": "", "ddg_snippet": "To overcome these challenges, we propose a checks-and-balances framework inspired by governmental structures, where independent but interacting components maintain ac-countability and balance.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/profile/Edward-Chang-22/publication/380515639_Checks-and-Balances_Framework_for_Context-Aware_Ethical_AI_Alignment/links/683696c46b5a287c3046b3e0/Checks-and-Balances-Framework-for-Context-Aware-Ethical-AI-Alignment.pdf", "content": "To overcome these challenges, we propose a checks-and-balances framework inspired by governmental structures, where independent but interacting components maintain ac-countability and balance."}
+{"idx": 2, "title": "A Three-Branch Checks-and-Balances ... - Semantic Scholar", "date": "", "ddg_snippet": "A three-branch checks - and - balances framework for ethical alignment of Large Language Models, inspired by governmental systems, demonstrates how emotional modeling can guide linguistic behaviors toward ethical outcomes while preserving independence across knowledge generation, ethical oversight, and contextual interpretation. Expand [PDF] Semantic Reader Save to Library Create Alert Cite", "subpage_snippet": "", "source": "www.semanticscholar.org", "link": "https://www.semanticscholar.org/paper/A-Three-Branch-Checks-and-Balances-Frameworkfor-of-Chang/5918a91419cf95db8599b086590facf63f124702/figure/4", "content": "A three-branch checks - and - balances framework for ethical alignment of Large Language Models, inspired by governmental systems, demonstrates how emotional modeling can guide linguistic behaviors toward ethical outcomes while preserving independence across knowledge generation, ethical oversight, and contextual interpretation. Expand [PDF] Semantic Reader Save to Library Create Alert Cite"}
+{"idx": 3, "title": "Edward Y. Chang - Stanford University", "date": "", "ddg_snippet": "Edward Y. Chang Stanford Inforlab Technical Report, June 2025 PDF SagaLLM: Context Management, Validation, and Transaction Guarantees for Multi-Agent LLM Planning Edward Y. Chang and Longling Geng VLDB, September 2025 (accepted) PDF | BibTex A Checks-and-Balances Framework for Ethical AI Alignment Edward Y. Chang ICML, July 2025 (accepted) PDF ...", "subpage_snippet": "", "source": "infolab.stanford.edu", "link": "http://infolab.stanford.edu/~echang/", "content": "Edward Y. Chang Stanford Inforlab Technical Report, June 2025 PDF SagaLLM: Context Management, Validation, and Transaction Guarantees for Multi-Agent LLM Planning Edward Y. Chang and Longling Geng VLDB, September 2025 (accepted) PDF | BibTex A Checks-and-Balances Framework for Ethical AI Alignment Edward Y. Chang ICML, July 2025 (accepted) PDF ..."}
+{"idx": 4, "title": "A Three-Branch Checks-and-Balances Framework for Context ...", "date": "", "ddg_snippet": "Abstract This paper introduces a three-branch checks - and - balances framework for ethical alignment of Large Language Models (LLMs), inspired by governmental systems. It implements three independent yet interacting components: LLMs as the ex-ecutive branch for knowledge generation, DIKE (the goddess of justice) as the legislative branch establishing ethical guardrails, and ERIS (the goddess of ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=o2afWIxjKD", "content": "Abstract This paper introduces a three-branch checks - and - balances framework for ethical alignment of Large Language Models (LLMs), inspired by governmental systems. It implements three independent yet interacting components: LLMs as the ex-ecutive branch for knowledge generation, DIKE (the goddess of justice) as the legislative branch establishing ethical guardrails, and ERIS (the goddess of ..."}
+{"idx": 5, "title": "A Checks-and-Balances Framework for Context-Aware Ethical AI ...", "date": "", "ddg_snippet": "A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment Susceptible to social biases Vulnerable to reward hacking “Whack-A-Mole” reactive approach Catastrophic forgetting issues", "subpage_snippet": "", "source": "icml.cc", "link": "https://icml.cc/media/icml-2025/Slides/46461.pdf", "content": "A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment Susceptible to social biases Vulnerable to reward hacking “Whack-A-Mole” reactive approach Catastrophic forgetting issues"}
+{"idx": 6, "title": "A Three-Branch Checks-and-Balances Frameworkfor Context-Aware ...", "date": "", "ddg_snippet": "Can AI be ethically aligned using a government-inspired system of checks and balances? A Three-Branch Checks-and-Balances Frameworkfor Context-Aware Ethical Alignment of Large Language Models Published 2/3/2025 by Edward Y. Chang", "subpage_snippet": "", "source": "www.aimodels.fyi", "link": "https://www.aimodels.fyi/papers/arxiv/three-branch-checks-balances-frameworkfor-context-aware", "content": "Can AI be ethically aligned using a government-inspired system of checks and balances? A Three-Branch Checks-and-Balances Frameworkfor Context-Aware Ethical Alignment of Large Language Models Published 2/3/2025 by Edward Y. Chang"}
+{"idx": 7, "title": "Data-to-Dashboard: Multi-Agent LLM Framework for Insightful", "date": "", "ddg_snippet": "... capabilities of Large Language Models, our framework consists of specialized agents tasked with: (1) detecting the domain and concepts of the dataset ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2505.23695v1", "content": "... capabilities of Large Language Models, our framework consists of specialized agents tasked with: (1) detecting the domain and concepts of the dataset ..."}
+{"idx": 8, "title": "Self-Checker: Plug-and-Play Modules for Fact-Checking with", "date": "", "ddg_snippet": "The lack of a suitable fact- checking dataset tailored for LLM generation poses a challenge in designing and evaluating frameworks in the evolving ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2305.14623v2", "content": "The lack of a suitable fact- checking dataset tailored for LLM generation poses a challenge in designing and evaluating frameworks in the evolving ..."}
+{"idx": 9, "title": "Rethinking the Evaluation of Alignment Methods: Insights into", "date": "", "ddg_snippet": "... a unified evaluation framework that compares LLM alignment methods (PPO, DPO, ORPO, KTO) across these five axes, using both in-distribution and ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2509.12936v1", "content": "... a unified evaluation framework that compares LLM alignment methods (PPO, DPO, ORPO, KTO) across these five axes, using both in-distribution and ..."}

data/sampled_jsons/A_Checks-and-Balances_Framework_for_Context-Aware_Ethical_AI_Alignment_Adversarial_Review_algorithm_.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "A Checks - and - Balances Framework for Context - Aware Ethical AI ...", "date": "", "ddg_snippet": "Table 1 : Checks - and - balances , adversarial review algorithm . Algorithm Θ+ & Θ− = Adversarial _ Review (s). Input. s: Decision of Dike; Output.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2502.00136", "content": "Table 1 : Checks - and - balances , adversarial review algorithm . Algorithm Θ+ & Θ− = Adversarial _ Review (s). Input. s: Decision of Dike; Output."}
+{"idx": 1, "title": "(PDF) Checks - and - Balances Framework for Context - Aware Ethical ...", "date": "", "ddg_snippet": "Table 1 : Checks - and - balances , adversarial review algorithm .This work introduces a checks - and - balances framework for . ethical AI behavior. By delineating the responsibilities", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/380515639_Checks-and-Balances_Framework_for_Context-Aware_Ethical_AI_Alignment", "content": "Table 1 : Checks - and - balances , adversarial review algorithm .This work introduces a checks - and - balances framework for . ethical AI behavior. By delineating the responsibilities"}
+{"idx": 2, "title": "A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment", "date": "", "ddg_snippet": "Although guardrails provide de-fault constraints, they can be dynamically adjusted based on context, particularly through the dialectic process with Eris, which helps prevent rigid enforcement that might be Table 1 : Checks-and-balances, adversarial review algorithm Algorithm Θ+ & Θ− = Adversarial_Review (s) Input. s: Decision of Dike; Output.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2502.00136v3", "content": "Although guardrails provide de-fault constraints, they can be dynamically adjusted based on context, particularly through the dialectic process with Eris, which helps prevent rigid enforcement that might be Table 1 : Checks-and-balances, adversarial review algorithm Algorithm Θ+ & Θ− = Adversarial_Review (s) Input. s: Decision of Dike; Output."}
+{"idx": 3, "title": "A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment", "date": "", "ddg_snippet": "For example, a social media plat-form might adjust G based on the topics discussed and the countries it serves. This administrative control is balanced by transparent documentation requirements and potential Table 1 : Checks-and-balances, adversarial review algorithm Algorithm Θ+ & Θ− = Adversarial_Review (s) Input. s: Decision of Dike; Output.", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2502.00136v2", "content": "For example, a social media plat-form might adjust G based on the topics discussed and the countries it serves. This administrative control is balanced by transparent documentation requirements and potential Table 1 : Checks-and-balances, adversarial review algorithm Algorithm Θ+ & Θ− = Adversarial_Review (s) Input. s: Decision of Dike; Output."}
+{"idx": 4, "title": "PDF A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment", "date": "", "ddg_snippet": "Abstract This paper introduces a checks-and-balances framework for ethical alignment of Large Lan-guage Models (LLMs), inspired by three-branch governmental systems. It implements three in ...", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/profile/Edward-Chang-22/publication/380515639_Checks-and-Balances_Framework_for_Context-Aware_Ethical_AI_Alignment/links/683696c46b5a287c3046b3e0/Checks-and-Balances-Framework-for-Context-Aware-Ethical-AI-Alignment.pdf", "content": "Abstract This paper introduces a checks-and-balances framework for ethical alignment of Large Lan-guage Models (LLMs), inspired by three-branch governmental systems. It implements three in ..."}
+{"idx": 5, "title": "A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment", "date": "", "ddg_snippet": "Although guardrails provide de-fault constraints, they can be dynamically adjusted based on context, particularly through the dialectic process with Table 1 : Checks-and-balances, adversarial review algorithm Algorithm Θ+ & Θ− = Adversarial_Review (s) Input. s: Decision of Dike; Output.", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=4uOEiitySn", "content": "Although guardrails provide de-fault constraints, they can be dynamically adjusted based on context, particularly through the dialectic process with Table 1 : Checks-and-balances, adversarial review algorithm Algorithm Θ+ & Θ− = Adversarial_Review (s) Input. s: Decision of Dike; Output."}
+{"idx": 6, "title": "PDF A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment", "date": "", "ddg_snippet": "A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment Susceptible to social biases Vulnerable to reward hacking \"Whack-A-Mole\" reactive approach Catastrophic forgetting issues", "subpage_snippet": "", "source": "icml.cc", "link": "https://icml.cc/media/icml-2025/Slides/46461.pdf", "content": "A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment Susceptible to social biases Vulnerable to reward hacking \"Whack-A-Mole\" reactive approach Catastrophic forgetting issues"}
+{"idx": 7, "title": "Downloads", "date": "", "ddg_snippet": "A Checks - and - Balances Framework for Context - Aware Ethical AI Alignment .A Mathematical Framework for AI-Human Integration in Work. am-ELO: A Stable Framework for Arena-based LLM Evaluation.", "subpage_snippet": "", "source": "icml.cc", "link": "https://icml.cc/Downloads/2025", "content": "A Checks - and - Balances Framework for Context - Aware Ethical AI Alignment .A Mathematical Framework for AI-Human Integration in Work. am-ELO: A Stable Framework for Arena-based LLM Evaluation."}
+{"idx": 8, "title": "F.2.2 | jarxiv", "date": "", "ddg_snippet": "A Checks - and - Balances Framework for Context - Aware Ethical AI Alignment . 投稿日: 2025年5月27日 作成者: jarxiv.Deep-SITAR: A SITAR-Based Deep Learning Framework for Growth Curve Modeling via Autoencoders.", "subpage_snippet": "", "source": "jarxiv.com", "link": "https://jarxiv.com/category/f-2-2/", "content": "A Checks - and - Balances Framework for Context - Aware Ethical AI Alignment . 投稿日: 2025年5月27日 作成者: jarxiv.Deep-SITAR: A SITAR-Based Deep Learning Framework for Growth Curve Modeling via Autoencoders."}
+{"idx": 9, "title": "Using the Geneva Emotion Wheel to Classify the... | Semantic Scholar", "date": "", "ddg_snippet": "A Checks - and - Balances Framework for Context - Aware Ethical AI Alignment . A checks - and - balances framework for ethical alignment of Large Language Models inspired by three-branch governmental systems is introduced, demonstrating how DIKE and ERIS direct...", "subpage_snippet": "", "source": "www.semanticscholar.org", "link": "https://www.semanticscholar.org/paper/Using-the-Geneva-Emotion-Wheel-to-Classify-the-of-McGinn-Kelly/28b8ca8eef945f4c503d15166a95142d73f220b3", "content": "A Checks - and - Balances Framework for Context - Aware Ethical AI Alignment . A checks - and - balances framework for ethical alignment of Large Language Models inspired by three-branch governmental systems is introduced, demonstrating how DIKE and ERIS direct..."}

data/sampled_jsons/A_Checks-and-Balances_Framework_for_Context-Aware_Ethical_AI_Alignment_evaluation_benchmark_dataset.jsonl

@@ -0,0 +1,10 @@
+{"idx": 0, "title": "Order Personal Checks Online At Affordable Prices | Checks.com", "date": "", "ddg_snippet": "We print an exceptional collection of high-quality personal checks at cheap prices. Whether you are buying a checkbook for the first time or are reordering your favorite check design, you’ll always get the same low price - no discounts codes required.", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/", "content": "We print an exceptional collection of high-quality personal checks at cheap prices. Whether you are buying a checkbook for the first time or are reordering your favorite check design, you’ll always get the same low price - no discounts codes required."}
+{"idx": 1, "title": "View Over 70 Personal Checks Designs at Low Prices", "date": "", "ddg_snippet": "Order your favorite personal checks online at low prices. Checks .com features a variety of over 70 personal check designs to fit your unique personality.", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/c/129/personal-checks", "content": "Order your favorite personal checks online at low prices. Checks .com features a variety of over 70 personal check designs to fit your unique personality."}
+{"idx": 2, "title": "Value Checks - Order Our Best Value Checks Online", "date": "", "ddg_snippet": "Order value-priced personal checks online starting at just $8.20 per box at Checks .com! We have a variety of inexpensive check designs to choose from, so you can find the perfect one to match your style and budget.", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/c/323/value-checks", "content": "Order value-priced personal checks online starting at just $8.20 per box at Checks .com! We have a variety of inexpensive check designs to choose from, so you can find the perfect one to match your style and budget."}
+{"idx": 3, "title": "How To Order Checks Online | Checks.com", "date": "", "ddg_snippet": "Order checks , address labels and checkbook covers online, secure, fast and easy with Checks .com. Customize and preview your checks before you order.", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/how-to-order-checks", "content": "Order checks , address labels and checkbook covers online, secure, fast and easy with Checks .com. Customize and preview your checks before you order."}
+{"idx": 4, "title": "Traditional Checks - Order Affordable Personal Checks Online", "date": "", "ddg_snippet": "Checks .com's collection of traditional checks offers the widest variety of designs where you're sure to find a favorite or two. Also check out our line of Choice Checks for our most exclusive and most secure check designs, or browse our cheaper-priced checks for an incredibly low price per box.", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/c/334/traditional-checks", "content": "Checks .com's collection of traditional checks offers the widest variety of designs where you're sure to find a favorite or two. Also check out our line of Choice Checks for our most exclusive and most secure check designs, or browse our cheaper-priced checks for an incredibly low price per box."}
+{"idx": 5, "title": "Reorder Discounted Personal Checks Online", "date": "", "ddg_snippet": "At Checks .com, whenever you reorder checks , you pay the same low price as intro customers. Reorder checks for discounted prices online today!", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/quickreorder", "content": "At Checks .com, whenever you reorder checks , you pay the same low price as intro customers. Reorder checks for discounted prices online today!"}
+{"idx": 6, "title": "Renaissance Checks - Order Discounted Personal Checks", "date": "", "ddg_snippet": "Artistic scrolls and elegance grace these renaissance checks in a rich four color rotation. Coordinating renaissance address labels and checkbook cover are also available.", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/p/834/renaissance-checks", "content": "Artistic scrolls and elegance grace these renaissance checks in a rich four color rotation. Coordinating renaissance address labels and checkbook cover are also available."}
+{"idx": 7, "title": "Modern Checks - Order Cool & Contemporary Personal Checks", "date": "", "ddg_snippet": "Show your trendy style with contemporary and cool personal checks from Checks .com! Choose a stylish check design or pattern and order checks online today!", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/c/274/modern-checks", "content": "Show your trendy style with contemporary and cool personal checks from Checks .com! Choose a stylish check design or pattern and order checks online today!"}
+{"idx": 8, "title": "Pretty in Pink Checks - Order Discounted Personal Checks", "date": "", "ddg_snippet": "Discover the perfect blend of Parisian style and elegance with our Pretty in Pink checks . Designed for those who live life with flair, these top-tear pink personal checks are available in four vibrant designs featuring stripes, polka dots and diamond patterns.", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/p/1799/pretty-in-pink-checks", "content": "Discover the perfect blend of Parisian style and elegance with our Pretty in Pink checks . Designed for those who live life with flair, these top-tear pink personal checks are available in four vibrant designs featuring stripes, polka dots and diamond patterns."}
+{"idx": 9, "title": "Horse Checks - Order Discounted Personal Checks", "date": "", "ddg_snippet": "Horse Play Checks Photographs of galloping horses adorn this equine rotation of stallions, mustangs and foals. Coordinating horse return address labels are available.", "subpage_snippet": "", "source": "www.checks.com", "link": "https://www.checks.com/p/811/horse-play-checks", "content": "Horse Play Checks Photographs of galloping horses adorn this equine rotation of stallions, mustangs and foals. Coordinating horse return address labels are available."}

data/sampled_jsons/A_Likelihood_Based_Approach_to_Distribution_Regression_Table_1_Sieve_MLE_FD3_MSE_standard_deviation_year_2023.jsonl

@@ -0,0 +1,7 @@
+{"idx": 0, "title": "A Likelihood Based Approach to Distribution Regression Using", "date": "", "ddg_snippet": "Keywords: Distribution Regression ; Conditional Deep Generative Models; Intrinsic Manifold Structure; Sieve MLE ; Wasserstein Convergence. 1 Introduction. Conditional distribution estimation provides a principled framework for characterizing the dependence rela-tionship between...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/pdf/2410.02025", "content": "Keywords: Distribution Regression ; Conditional Deep Generative Models; Intrinsic Manifold Structure; Sieve MLE ; Wasserstein Convergence. 1 Introduction. Conditional distribution estimation provides a principled framework for characterizing the dependence rela-tionship between..."}
+{"idx": 1, "title": "A Likelihood Based Approach to Distribution Regression ...", "date": "", "ddg_snippet": "2 Oct 2024 — Table 1 : MSE for the estimated conditional mean and the standard deviation . Sieve MLE , CKDE, FlexCode. FD1, MEAN, 0.0379 ...", "subpage_snippet": "", "source": "arxiv.org", "link": "https://arxiv.org/html/2410.02025v1", "content": "2 Oct 2024 — Table 1 : MSE for the estimated conditional mean and the standard deviation . Sieve MLE , CKDE, FlexCode. FD1, MEAN, 0.0379 ..."}
+{"idx": 2, "title": "A Likelihood Based Approach to Distribution Regression Using ...", "date": "", "ddg_snippet": "standard deviation for the sieve MLE , and numerical inte- ... Table 1 . MSE for the estimated ... scenarios except for the MSE (SD) for the FD3 dataset.", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/pdf?id=1IyPRv1A0r", "content": "standard deviation for the sieve MLE , and numerical inte- ... Table 1 . MSE for the estimated ... scenarios except for the MSE (SD) for the FD3 dataset."}
+{"idx": 3, "title": "A Likelihood Based Approach to Distribution Regression ...", "date": "", "ddg_snippet": "Note that the sieve MLE outperforms all other methods in all scenarios except for the MSE (SD) for the FD3 dataset. However, for the FD3 dataset, we found ...", "subpage_snippet": "", "source": "icml.cc", "link": "https://icml.cc/virtual/2025/poster/46645", "content": "Note that the sieve MLE outperforms all other methods in all scenarios except for the MSE (SD) for the FD3 dataset. However, for the FD3 dataset, we found ..."}
+{"idx": 4, "title": "(PDF) A Likelihood Based Approach to Distribution Regression ...", "date": "", "ddg_snippet": "Our results lead to the convergence rate of a sieve maximum likelihood estimator ( MLE ) for estimating the conditional distribution (and its devolved counterpart) of the response given predictors in the Hellinger (Wasserstein) metric.", "subpage_snippet": "", "source": "www.researchgate.net", "link": "https://www.researchgate.net/publication/384630603_A_Likelihood_Based_Approach_to_Distribution_Regression_Using_Conditional_Deep_Generative_Models", "content": "Our results lead to the convergence rate of a sieve maximum likelihood estimator ( MLE ) for estimating the conditional distribution (and its devolved counterpart) of the response given predictors in the Hellinger (Wasserstein) metric."}
+{"idx": 5, "title": "TIVE MODELS", "date": "", "ddg_snippet": "Table 1 summarizes the findings. Table 1 : MSE for the estimated conditional mean and the standard deviation . Sieve MLE . CKDE. FlexCode. FD1. MEAN. 0.0379 ...", "subpage_snippet": "", "source": "openreview.net", "link": "https://openreview.net/notes/edits/attachment?id=bKSeeDBAan&name=pdf", "content": "Table 1 summarizes the findings. Table 1 : MSE for the estimated conditional mean and the standard deviation . Sieve MLE . CKDE. FlexCode. FD1. MEAN. 0.0379 ..."}
+{"idx": 6, "title": "", "date": "", "ddg_snippet": "", "subpage_snippet": "", "source": "", "link": "", "content": ""}