datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Sci2Pol-BenchSci2Pol-Bench
Data, scripts, and recipes for the benchmark Sci2Pol-Bench, a comprehensive benchmark for evaluating large language models.
About •
Usage•
Authors
About
The data consists of policy briefs obtained from Nature Energy, Nature Climate, Nature Cities, and Journal of Health and Social Behavior Policy Briefs.
Policy briefs originally were introduced in the Nature Energy journal with the goal of:
This format aims to provide… See the full description on the dataset page: https://huggingface.co/datasets/Northwestern-CSSI/Sci2Pol-Bench.landing-pages-styling-css-only-v2v3-merged
merged_v2_v3_dedup
This dataset is the deduplicated merged training set built from the repository's landing_page_v2 and landing_page_v3 pipelines.
It contains chat-format rows for CSS generation on landing pages:
messages + metadata
Each sample keeps:
messages: the training conversation, usually a user prompt plus the assistant CSS response.
metadata: generation and provenance fields such as source pipeline, style phrase, recipe, and other analysis attributes.… See the full description on the dataset page: https://huggingface.co/datasets/kogai/landing-pages-styling-css-only-v2v3-merged.full-html-stying-dataset-generated-css-from-style-plan
Generated CSS From Style Plan
kogai/full-html-stying-dataset-generated-css-from-style-plan contains generated_css_from_style_plan.jsonl, a JSONL dataset with 44458 synthetic examples. Model-generated CSS outputs conditioned on source HTML, user style requests, and structured style plans.
Schema
chat_template_overhead_tokens: field present in the JSONL records.
created_at: field present in the JSONL records.
input_html: source HTML before Tailwind classes are… See the full description on the dataset page: https://huggingface.co/datasets/kogai/full-html-stying-dataset-generated-css-from-style-plan.css-bench
CSS-Bench: Counterfactual Strategic Synthesis Benchmark
CSS-Bench tests whether language models make strategic decisions based on the underlying payoff topology of a game, or on the semantic valence of the words used to narrate it. Every item exists as a matched pair (or triple): a canonical framing where the numerically optimal action is also lexically "nice," and a counterfactual framing with the identical payoff structure but inverted narrative valence -- the numerically… See the full description on the dataset page: https://huggingface.co/datasets/jub-aer/css-bench.