glue
DatasetFreeDataset by nyu-mll. 3,94,564 downloads.
Capabilities8 decomposed
multi-task nlu benchmark dataset loading and evaluation
Medium confidenceProvides a curated collection of 9 diverse NLU tasks (CoLA, SST-2, MRPC, QQP, STS-B, MNLI, QNLI, RTE, WNLI) with standardized train/validation/test splits, enabling researchers to evaluate language models across acceptability classification, semantic similarity, natural language inference, and sentiment analysis in a single unified framework. Integrates with HuggingFace Datasets library for streaming, caching, and batch loading with automatic schema validation and format conversion (parquet, CSV, Arrow).
Aggregates 9 heterogeneous NLU tasks under a single standardized interface with consistent schema mapping, enabling single-pass evaluation across grammaticality, entailment, paraphrase, and sentiment tasks — unlike task-specific datasets that require separate loading pipelines. Uses HuggingFace Datasets' columnar Arrow format for efficient streaming and zero-copy access to 394K+ examples.
Provides unified multi-task evaluation framework with standardized splits (unlike SuperGLUE which focuses on harder tasks), lower computational barrier than custom benchmark construction, and native integration with modern NLP frameworks (Hugging Face Transformers, PyTorch Lightning) for immediate fine-tuning workflows.
task-specific train/validation/test split provisioning
Medium confidenceDelivers pre-defined, non-overlapping data splits for each of the 9 GLUE tasks with fixed random seeds ensuring reproducibility across research groups. Splits are accessible via HuggingFace Datasets' split selection API (e.g., dataset['train'], dataset['validation']) and include balanced class distributions where applicable, with metadata tracking original source corpus provenance and annotation guidelines.
Implements fixed, peer-reviewed splits across 9 tasks with documented random seeds and class balance constraints, enabling exact reproduction of published results — unlike ad-hoc dataset splits that vary across implementations. Integrates with HuggingFace Datasets' lazy-loading architecture to avoid materializing full splits in memory until needed.
Eliminates split variance that plagues custom benchmarks by providing official, immutable partitions used in 1000+ published papers, reducing experimental variance from data leakage and enabling fair cross-paper comparisons unlike task-specific datasets with inconsistent split definitions.
heterogeneous task schema mapping and normalization
Medium confidenceAbstracts away task-specific column naming and label encoding schemes (e.g., CoLA uses binary acceptability labels, MRPC uses paraphrase binary labels, STS-B uses continuous 0-5 scores) into a unified interface through HuggingFace Datasets' feature schema system. Automatically handles type conversion (string labels to integers, float scores to normalized ranges) and provides task metadata (number of classes, label names, task type) for downstream model configuration.
Implements Arrow-based columnar schema mapping that preserves task semantics while enabling unified iteration — unlike manual task-specific loaders that require conditional branches. Uses HuggingFace Features API to declare expected types upfront, enabling type validation and automatic casting without runtime overhead.
Eliminates boilerplate task-specific data loading code by providing unified schema across 9 diverse tasks (binary classification, multi-class, regression), reducing implementation complexity vs building separate loaders for each task and enabling true multi-task training without task-specific branches.
efficient streaming and batch loading with caching
Medium confidenceLeverages HuggingFace Datasets' streaming architecture to load GLUE data on-demand without materializing full datasets in memory, using memory-mapped Parquet files and Arrow IPC format for zero-copy access. Implements automatic caching to disk (configurable location) after first download, enabling subsequent loads in <1 second without network I/O. Supports batch iteration with configurable batch sizes and prefetching for GPU-efficient training pipelines.
Implements Arrow-native columnar caching with memory-mapped access, enabling zero-copy iteration over 394K+ examples without materializing in RAM — unlike CSV-based datasets that require full deserialization. Uses HuggingFace's distributed cache management to support multi-GPU training with shared cache across workers.
Provides streaming + caching hybrid that eliminates download bottleneck for initial runs while maintaining fast subsequent access, vs alternatives like raw CSV downloads (slow, memory-intensive) or cloud-only datasets (requires API keys, network latency). Native PyTorch integration enables single-line DataLoader wrapping without custom collate functions.
task-specific metric computation and leaderboard submission support
Medium confidenceProvides task-specific evaluation metrics (accuracy for CoLA/SST-2/MRPC/QQP/QNLI/RTE/WNLI, Pearson/Spearman correlation for STS-B, Matthews correlation for MNLI) through integration with HuggingFace Evaluate library. Metrics are pre-configured with task-appropriate aggregation (macro vs micro averaging, handling of missing predictions) and support leaderboard submission format validation (e.g., ensuring predictions match test set size and label space).
Integrates task-specific metric definitions (accuracy, Matthews correlation, Pearson correlation) with HuggingFace Evaluate's caching system, enabling reproducible metric computation across runs without reimplementation. Provides leaderboard submission format validation to catch common errors (mismatched prediction counts, out-of-range labels) before upload.
Eliminates manual metric implementation by providing pre-validated, task-specific metrics matching official leaderboard evaluation, vs alternatives like scikit-learn (requires task-specific metric selection logic) or custom implementations (prone to bugs, inconsistent with published results). Native integration with HuggingFace Transformers enables single-line evaluation after fine-tuning.
source corpus provenance tracking and annotation metadata
Medium confidenceIncludes structured metadata for each task documenting original source corpus (e.g., SST-2 from Stanford Sentiment Treebank, MRPC from Microsoft Research Paraphrase Corpus), annotation guidelines, inter-annotator agreement scores, and data collection methodology. Metadata is accessible via dataset.info property and includes links to original papers, enabling researchers to understand data quality and potential biases without external documentation lookup.
Embeds structured provenance metadata (source corpus, annotation guidelines, IAA scores) directly in dataset objects, enabling programmatic access to data quality signals without external documentation lookup — unlike standalone benchmark papers that require manual cross-referencing. Includes links to original papers for full methodological transparency.
Provides machine-readable data quality metadata integrated with dataset objects, vs alternatives like separate documentation files (requires manual lookup) or leaderboard websites (limited metadata). Enables automated data quality assessment and bias analysis without external tools.
multi-task learning and transfer learning dataset composition
Medium confidenceEnables researchers to combine multiple GLUE tasks into unified training datasets for multi-task learning experiments through HuggingFace Datasets' concatenation and interleaving APIs. Supports task-weighted sampling (e.g., oversample small tasks like RTE to balance training) and task-specific loss weighting for joint optimization. Provides utilities for task-aware batch construction (e.g., grouping examples by task type to minimize padding overhead).
Provides task-aware dataset composition through HuggingFace Datasets' interleaving API, enabling weighted sampling of heterogeneous tasks (e.g., oversample RTE's 2.5K examples to match QQP's 364K) without manual replication logic. Preserves task identity through metadata columns for downstream loss weighting.
Enables multi-task training without custom dataset construction by providing task-aware composition utilities, vs alternatives like manual concatenation (loses task identity) or separate task-specific models (no transfer learning). Native integration with HuggingFace Transformers enables multi-task fine-tuning with minimal code changes.
cross-task linguistic phenomenon analysis and error categorization
Medium confidenceEnables systematic analysis of model behavior across tasks by providing consistent text representations and label semantics, allowing researchers to identify which linguistic phenomena (grammaticality, entailment, paraphrase, sentiment) models struggle with. Supports error analysis workflows by enabling filtering and grouping of examples by task type, label, and text properties (length, complexity) without custom parsing logic.
Provides consistent text and label representations across 9 diverse linguistic tasks, enabling systematic cross-task error analysis without task-specific parsing — unlike single-task datasets that isolate phenomena. Preserves task identity metadata for grouping and filtering without external annotation.
Enables unified error analysis across diverse linguistic phenomena (grammaticality, entailment, sentiment) by providing consistent task interface, vs alternatives like separate task-specific analysis (fragmented insights) or custom benchmark construction (time-consuming). Native integration with HuggingFace Datasets enables filtering and grouping without custom code.
Capabilities are decomposed by AI analysis. Each maps to specific user intents and improves with match feedback.
Related Artifactssharing capabilities
Artifacts that share capabilities with glue, ranked by overlap. Discovered automatically through the match graph.
FLAN Collection
Google's 1,836-task instruction mixture for broad generalization.
MTEB
Embedding model benchmark — 8 tasks, 112 languages, the standard for comparing embeddings.
Flair
PyTorch NLP framework with contextual embeddings.
PromptBench
Microsoft's unified LLM evaluation and prompt robustness benchmark.
Multiagent Debate
Implementation of a paper on Multiagent Debate
promptbench
PromptBench is a powerful tool designed to scrutinize and analyze the interaction of large language models with various prompts. It provides a convenient infrastructure to simulate **black-box** adversarial **prompt attacks** on the models and evaluate their performances.
Best For
- ✓NLP researchers evaluating language model generalization
- ✓Teams fine-tuning BERT/RoBERTa/T5 variants on standard benchmarks
- ✓Practitioners building production NLU systems requiring baseline performance validation
- ✓Academic researchers publishing results requiring reproducibility
- ✓Benchmark leaderboard submissions (e.g., GLUE leaderboard) requiring exact split compliance
- ✓Teams implementing baseline models for comparison studies
- ✓Researchers building multi-task learning systems that train on multiple GLUE tasks simultaneously
- ✓Framework developers implementing task-agnostic fine-tuning pipelines
Known Limitations
- ⚠English-only (monolingual) — no cross-lingual or multilingual variants included
- ⚠Fixed task definitions and splits — cannot customize task formulations or data augmentation within the dataset itself
- ⚠Some tasks have small test sets (e.g., RTE ~277 examples) limiting statistical significance for low-resource evaluation
- ⚠No built-in handling of class imbalance (e.g., QQP is heavily skewed toward duplicate pairs)
- ⚠Requires external metric computation libraries (scikit-learn, scipy) for detailed evaluation beyond accuracy
- ⚠Splits are immutable — cannot customize train/val/test ratios for specific use cases
Requirements
Input / Output
UnfragileRank
UnfragileRank is computed from adoption signals, documentation quality, ecosystem connectivity, match graph feedback, and freshness. No artifact can pay for a higher rank.
About
glue — a dataset on HuggingFace with 3,94,564 downloads
Categories
Alternatives to glue
Are you the builder of glue?
Claim this artifact to get a verified badge, access match analytics, see which intents users search for, and manage your listing.
Get the weekly brief
New tools, rising stars, and what's actually worth your time. No spam.
Data Sources
Looking for something else?
Search →