datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
data_analysis
Dataset Card for "livebench/data_analysis"
LiveBench is a benchmark for LLMs designed with test set contamination and objective evaluation in mind. It has the following properties:
LiveBench is designed to limit potential contamination by releasing new questions monthly, as well as having questions based on recently-released datasets, arXiv papers, news articles, and IMDb movie synopses.
Each question has verifiable, objective ground-truth answers, allowing hard questions to be… See the full description on the dataset page: https://huggingface.co/datasets/livebench/data_analysis.Taur_CoT_Analysis_Project___gpt-4o-2024-08-06NLU-Sentiment-Analysis
SEA Sentiment Analysis
SEA Sentiment Analysis evaluates a model's ability to identify the sentiment polarity of a text. It is sampled from NusaX for Indonesian, Javanese, and Sundanese, IndicSentiment for Tamil, Wisesight Sentiment for Thai, and UIT-VSFC for Vietnamese.
Supported Tasks and Leaderboards
SEA Sentiment Analysis is designed for evaluating chat or instruction-tuned large language models (LLMs). It is part of the SEA-HELM leaderboard from AI Singapore.… See the full description on the dataset page: https://huggingface.co/datasets/aisingapore/NLU-Sentiment-Analysis.Taur_CoT_Analysis_Project___meta-llama__Meta-Llama-3.1-8B-InstructTaur_CoT_Analysis_Project___gpt-4o-mini-2024-07-18Taur_CoT_Analysis_Project___microsoft__Phi-3-small-8k-instructTaur_CoT_Analysis_Project___google__gemini-1.5-flash-001static-analysis-evalA dataset of 76 Python programs taken from real Python open source projects (top 100 on GitHub),
where each program is a file that has exactly 1 vulnerability as detected by a particular static analyzer (Semgrep), used in the paper Patched MOA: optimizing inference for diverse software development tasks.
OpenAI used the synth-vuln-fixes and fine-tuned
a new version of gpt-4o is now the SOTA on this benchmark. More details and code is available from their repo.
More details on the benchmark… See the full description on the dataset page: https://huggingface.co/datasets/patched-codes/static-analysis-eval.browsecomp-plus-selected-tools-analysis-v1
BrowseComp-Plus: Selected Tools Analysis
Side-by-side view of selected tool calls from a reference trajectory alongside the new agent trajectory conditioned on those steps.
Retrieval model: Qwen3-Embedding-8BAgent model: gpt-oss-120bRun: traj_summary_ext_selected_tools_gpt-oss-120b_seed0
Columns
Column
Description
query_id
Query identifier
rationale
GPT rationale for why these k steps were selected from the reference trajectory
selected_indices
Step indices… See the full description on the dataset page: https://huggingface.co/datasets/timchen0618/browsecomp-plus-selected-tools-analysis-v1.Taur_CoT_Analysis_Project___mistralai__Mistral-7B-Instruct-v0.3source-analysis
NuBerea Source Analysis
Source-critical analysis of the Hebrew Bible, Septuagint, New Testament, Vulgate, and Second Temple literature. The dataset carries machine-generated source and tradition annotations at the verse level — the classical concerns of source criticism (documentary strata in the Old Testament, corpus structure in the New Testament, the pathway of Old Testament traditions into New Testament citation) expressed as structured data — together with semantic-domain… See the full description on the dataset page: https://huggingface.co/datasets/NuBerea/source-analysis.trending-models-analysishttps://github.com/pagezyhf/azure-cron/blob/main/trending_models_analysis.py
Taur_CoT_Analysis_Project___google__gemini-1.5-pro-001Taur_CoT_Analysis_Project___Qwen__Qwen2-72B-InstructTaur_CoT_Analysis_Project___claude-3-5-sonnet-20240620translation-analysis
NuBerea Translation Verse Texts
Verse-level texts of historical Bible translations (Clementine Vulgate, Luther Bible 1545, Matthew's Bible 1537). Part of the NuBerea curated corpus estate of biblical and historical texts.
Attribution
Upstream Data Sources
Source
License
Clementine Vulgate, NOCR
Public Domain
Luther Bible 1545, NOCR
Public Domain
Matthew's Bible 1537, Textus Receptus Bibles
Public Domain
NuBerea project. Licensed… See the full description on the dataset page: https://huggingface.co/datasets/NuBerea/translation-analysis.septuagint-analysis
NuBerea Septuagint Textual Analysis
Curated datasets for study of the Septuagint (the ancient Greek translation of the
Hebrew Bible), part of the NuBerea corpus estate of biblical and patristic texts.
It gathers Septuagint verse texts, apparatus notes, and edition-comparison material
into a set of ready-to-load configurations.
Attribution
This dataset derives from the following upstream sources, which require attribution:
Source
License
Rahlfs… See the full description on the dataset page: https://huggingface.co/datasets/NuBerea/septuagint-analysis.pseudepigrapha-analysis
NuBerea Pseudepigrapha Analysis
Derived linguistic datasets over pseudepigraphal literature, part of the NuBerea curated corpus estate. Covers the Greek and Latin witnesses of these texts along with a multilingual view across the available witness languages.
License
CC BY 4.0.
Attribution
Source
Link
License
NuBerea project
https://huggingface.co/NuBerea
CC BY 4.0
ticker_analysis_articleslxx-analysis
NuBerea Research: LXX Translation-Technique Noise Model
Quantitative study of Septuagint translation technique: verse-by-verse measurements of
where the ancient Greek translation (LXX) diverges from the Hebrew Masoretic Text, with
book-level statistical summaries. The material lets researchers distinguish a translator's
habitual working style — free versus literal rendering — from genuine textual anomalies
worth close scholarly attention, putting on a measurable footing what LXX… See the full description on the dataset page: https://huggingface.co/datasets/NuBerea/lxx-analysis.Taur_CoT_Analysis_Project__Symbolic_Solver_ExperimentTaur_CoT_Analysis_Project___google__gemma-2-9b-itticker_analysis_pricesdetails_deep-analysis-research__D2IL-Arabic-Qwen2.5-72B-Instruct-v0.2_v2
Dataset Card for Evaluation run of deep-analysis-research/D2IL-Arabic-Qwen2.5-72B-Instruct-v0.2
Dataset automatically created during the evaluation run of model deep-analysis-research/D2IL-Arabic-Qwen2.5-72B-Instruct-v0.2.
The dataset is composed of 116 configuration, each one corresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the… See the full description on the dataset page: https://huggingface.co/datasets/OALL/details_deep-analysis-research__D2IL-Arabic-Qwen2.5-72B-Instruct-v0.2_v2.gretel-financial-risk-analysis-v1
gretelai/gretel-financial-risk-analysis-v1
This dataset contains synthetic financial risk analysis text generated by fine-tuning Phi-3-mini-128k-instruct on 14,306 SEC filings (10-K, 10-Q, and 8-K) from 2023-2024, utilizing differential privacy. It is designed for training models to extract key risk factors and generate structured summaries from financial documents while demonstrating the application of differential privacy to safeguard sensitive information.
This dataset showcases… See the full description on the dataset page: https://huggingface.co/datasets/gretelai/gretel-financial-risk-analysis-v1.saraiki-sentiment-analysis-datasetGym-Exercise-Video-Analysis
Gym-Exercise-Video-Analysis
Gym-Exercise-Video-Analysis is a specialized multimodal video understanding dataset comprising 500 annotated gym workout and exercise clips. It is designed for fine-tuning and evaluating Video-Language Models (Video-LLMs), visual fitness coaches, and temporal exercise analysis systems. Each entry pairs exercise videos and extracted frame sequences with in-depth textual descriptions, biomechanical observations, form evaluations, and routine tracking.… See the full description on the dataset page: https://huggingface.co/datasets/prithivMLmods/Gym-Exercise-Video-Analysis.QuRatedPajama-1B_tokens_for_analysis
QuRatedPajama
Paper: QuRating: Selecting High-Quality Data for Training Language Models
This dataset is a 1B token subset derived from princeton-nlp/QuRatedPajama-260B, which is a subset of cerebras/SlimPajama-627B annotated by princeton-nlp/QuRater-1.3B with sequence-level quality ratings across 4 criteria:
Educational Value - e.g. the text includes clear explanations, step-by-step reasoning, or questions and answers
Facts & Trivia - how much factual and trivia knowledge the text… See the full description on the dataset page: https://huggingface.co/datasets/princeton-nlp/QuRatedPajama-1B_tokens_for_analysis.bengali_sentiment_analysis
Bengali Sentiment Analysis
Context
The dataset contains 3307 Negative reviews and 8500 Positive reviews collected and manually annotated from Youtube Bengali drama.
Positive_Label=1 and Negative_Label=0
Acknowledgements
Sazzed, Salim (2021), “Bangla ( Bengali ) sentiment analysis classification benchmark dataset corpus”, Mendeley Data, V4, doi: 10.17632/p6zc7krs37.4
isom5240-td-traffic-analysis