datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ch-pit
Swiss Point-in-Time Law (CH-PiT)
Version v2026.09. Items per language: de: 5000 (core 500), fr: 5000 (core 500), it: 5000 (core 500). Oracle: 1.000 on every language. Load with load_dataset("overthelex/ch-pit", "de", split="core"); scorer and runners: https://github.com/overthelex/ch-pit.
Results on core
system
n
correct %
wrong version %
ungrounded %
date-sensitive: correct %
de / fr / it
recite-easy: correct %
errors
recite/recite
1500
48.7
41.1… See the full description on the dataset page: https://huggingface.co/datasets/overthelex/ch-pit.ch-pilot-rollouts-qwen3.5-9b
C&H Pilot Rollouts — Qwen/Qwen3.5-9B
20 agentic exploration rollouts over the Calderwood & Harkness (C&H) synthetic law-firm
corpus (the open-sourced world from harvey-labs
tasks/firm-knowledge/, MIT), generated by Qwen/Qwen3.5-9B served with vLLM.
Part of an actor-selection pilot for a world-internalization research project: the goal is to
mine agent trajectories into verified fact stores and rewritten likelihood-training targets.
Companion dataset (same seeds/tasks, different… See the full description on the dataset page: https://huggingface.co/datasets/violetxi/ch-pilot-rollouts-qwen3.5-9b.fssai-regulations-chp2-qaturkey-chp-gpt4oHandlabeled dataset to finetune GPT-4o model to classify whether a sentence depicts CHP as "villain" according to the below codebook:
CHP
CHP as the villain
Illegitimacy Component
For the illegitimacy component of polarizing rhetoric, we focus on sentences which portray the main opposition party, CHP, as an illegitimate political actor. More specifically, sentences containing similar claims to categories below, are coded as instances where CHP is shown as illegitimate.… See the full description on the dataset page: https://huggingface.co/datasets/Politics/turkey-chp-gpt4o.ch_poems_for_classificationCHP2015_2026turkey-chp-finetuneLanguage: Turkish.
Finetuning dataset for detecting delegimitizing language towards Turkish political party CHP.
Sentence-level dataset. Sentences extracted from news articles and filtered down to include CHP mentions through party abbrevations and leader surnames during their tenure.
Sentences were labeled by a fine-tuned GPT-4o model containing ~800 manually labeled sentences which was evaluated on the held-out set of ~800 sentences, resulting in an F1 score of 84%. Then this fine-tuned… See the full description on the dataset page: https://huggingface.co/datasets/Politics/turkey-chp-finetune.finetuning_demoturkey-chp-heldoutHeldout dataset to evaluate fine-tuned Politics/turkey-chp-villain model that classifies a given sentence based on whether it depicts CHP as a "villain" according to the below codebook.
CHP
CHP as the villain
Illegitimacy Component
For the illegitimacy component of polarizing rhetoric, we focus on sentences which portray the main opposition party, CHP, as an illegitimate political actor. More specifically, sentences containing similar claims to categories below, are coded… See the full description on the dataset page: https://huggingface.co/datasets/Politics/turkey-chp-heldout.