Table · dataset · 2026

MoralTrace

Listed in Hugging Face Datasets

Description

MoralTrace MoralTrace is a controlled English-language benchmark for studying how language models construct affected participants, bind outcomes to causal sources, and route event information into moral judgments. It accompanies the anonymous ICLR submission “Tracing the Causal Transition from Event Relations to Moral Judgment in Language Models.” Dataset composition The release contains 51,177 instances in nine configurations: single_chain_core: 9,900 instances;… See the full description on the dataset page: huggingface.co/datasets/happy-holiday/MoralTrace.

Topics

Stated by source: text · text classification
From keywords: Computer Science & AI

Provenance · 1 source records, 10 field assertions

Source	Key	Last seen	Raw
Hugging Face Datasets	happy-holiday/MoralTrace	15 d ago	JSON v1

Field	Assertion	Extractor	Evidence
access_level	source · Hugging Face	connector:huggingface@1.0.0	/gated
concepts[field].local:field:computer-science-ai	mapping · Hugging Face	connector:huggingface@1.0.0
concepts[modality].hf_modality:text	source · Hugging Face	connector:huggingface@1.0.0
concepts[task].hf_task:text-classification	source · Hugging Face	connector:huggingface@1.0.0	/tags[task_categories:*]
created_date	source · Hugging Face	connector:huggingface@1.0.0
description	source · Hugging Face	connector:huggingface@1.0.0	/description
license	source · Hugging Face	connector:huggingface@1.0.0	/tags[license:*]
publication_date	source · Hugging Face	connector:huggingface@1.0.0
title	source · Hugging Face	connector:huggingface@1.0.0	/id
updated_date	source · Hugging Face	connector:huggingface@1.0.0

MoralTrace

Description

Links

Where it is published

Topics