datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
mlir-benchmarks
MLIR Benchmarks
Multilingual Information Retrieval (MLIR) evaluation benchmarks used in:
MIMO: Multilingual Information Retrieval via Monolingual Objectives (EMNLP 2026)
SHIFT: Semantic Harmonization via Index-side Feature Transformation for Multilingual Information Retrieval (EMNLP 2026 Findings)
Four multilingual QA/retrieval datasets (Belebele, MLQA, XQuAD-R, MultiEuP-v2) adapted to the MLIR setting: all language versions of the context passages are pooled into one mixed… See the full description on the dataset page: https://huggingface.co/datasets/yjoonjang/mlir-benchmarks.MLIRBench
Dataset Card for MLIRBench
MLIRBench is a benchmark dataset for evaluating semantic reasoning, semantic equivalence analysis, execution-aware validation, and compiler-aware reasoning over programs represented in the Multi-Level Intermediate Representation (MLIR).
Dataset Details
Dataset Description
MLIRBench is a benchmark dataset for evaluating semantic reasoning, equivalence analysis, and execution-aware understanding of programs represented in… See the full description on the dataset page: https://huggingface.co/datasets/mlirbench/MLIRBench.MLIR-Spec-150
MLIR-Spec-150
Hand-authored NL→MLIR pairs for arith+func+memref dialects (n=150).
Composition
Instances: 150
Format: one JSON record per line in data/test.jsonl
Schema: fields = dialect, difficulty, id, mlir, nl, notes
Verifier: mlir-opt --verify-diagnostics against pinned LLVM 19.1.7
License: Apache-2.0 (SPDX: Apache-2.0). No third-party IP restrictions.
Loading
from datasets import load_dataset
ds = load_dataset("plawanrath/MLIR-Spec-150", split="test")… See the full description on the dataset page: https://huggingface.co/datasets/plawanrath/MLIR-Spec-150.MLIR-Functional-Reference-30
MLIR-Functional-Reference-30
Hand-authored functional-correctness reference set for arith, linalg+memref, and stablehlo (n=30, 10 per dialect).
Composition
Instances: 30
Format: one JSON record per line in data/test.jsonl
Schema: fields = canonical_fn_name, canonical_signature, dialect, expected_output, expected_output_pattern, expected_stdout_regex, id, inputs, iree_inputs, memref_inputs, memref_print, nl, result_type, scalar_inputs, source_benchmark, source_id… See the full description on the dataset page: https://huggingface.co/datasets/plawanrath/MLIR-Functional-Reference-30.