span-detection
task858_inquisitive_span_detection
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task858_inquisitive_span_detection
Additional Information
Citation Information
The following paper introduces the corpus in detail. If you use the corpus in published work, please cite it:
@misc{wang2022supernaturalinstructionsgeneralizationdeclarativeinstructions,
title={Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks}… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task858_inquisitive_span_detection.cantonese-span-detection
Cantonese Span Detection dataset
This dataset contains writings in (possibly) a mixture of Standard Chinese and Cantonese, derived from the NLPTEA 2017 Chinese Spelling Check Shared Task (Fung et al., NLP-TEA 2017).
This dataset is intended for text classification or token classification (span detection) tasks.
Columns:
id: Identifier, in the format of ASTRI0XXX, EVAXXX or ADDXXX.
sentence: A Chinese sentence that may contain spelling mistakes and/or Cantonese colloqialisms.… See the full description on the dataset page: https://huggingface.co/datasets/Swithord/cantonese-span-detection.span_absinth_german_faithfulness_detection_dataset
Dataset Card for "Span Absinth - Hallucination Detection Dataset of German News Summarization"
Dataset Description
Span Absinth is an extension of the Absinth dataset, where each hallucinated summary-sentence has been augmented with span annotations, that define which part of the sentence is hallucinated. Span annotations have the advantage of
effectively isolating hallucinations at the token level.
Please refer to our paper and Absinth, for more details about the… See the full description on the dataset page: https://huggingface.co/datasets/mtc/span_absinth_german_faithfulness_detection_dataset.toolace-hallucination-span-detection
ToolACE Hallucination Span Detection
This dataset was generated for span-level hallucination detection in tool-calling dialogues.
It is derived from Team-ACE/ToolACE.
Main file:
generated_data/toolace_ragtruth_all_with_negatives_and_splits.jsonl
Each example contains a query, tool context, final answer, hallucination labels,
hallucination type, split, and tool metadata. Labels are character-level spans
in the answer.
Construction
We extract completed ToolACE… See the full description on the dataset page: https://huggingface.co/datasets/katsubakirill/toolace-hallucination-span-detection.span-marker-base-model-detection
SpanMarker Base Model Detection
It is relative simply to determine base model of a fine-tuned SpanMarker model:
import os
from huggingface_hub import login, HfApi
hf_token = os.environ.get("HF_TOKEN")
login(token=hf_token, add_to_git_credential=True)
api = HfApi()
Please make sure that HF_TOKEN is set as environment variable.
After that, list of all SpanMarker models can be retrieved and configuration file is parsed.
Please make sure that span-marker library is installed:
from… See the full description on the dataset page: https://huggingface.co/datasets/stefan-it/span-marker-base-model-detection.flan_combined_task858_inquisitive_span_detection