riddles
Datasets
All datasets matching “riddles”riddle_senseriddles_v1
Riddle Processing with GPT-4
Buy me Ko-fi
Credits
All credit for the original riddles goes to crawsome's GitHub repository.
Project Overview
This project involves processing each riddle using GPT-4. The correct answers were provided to the model to generate a desirable output focused on reasoning and logical breakdown.
riddles.json (riddles_1) — 386 samples, sourced from crawsome's GitHub repository.
riddles_2.json — 83 samples, sourced from various Google… See the full description on the dataset page: https://huggingface.co/datasets/Hypersniper/riddles_v1.riddle_senseAnswering such a riddle-style question is a challenging cognitive process, in that it requires
complex commonsense reasoning abilities, an understanding of figurative language, and counterfactual reasoning
skills, which are all important abilities for advanced natural language understanding (NLU). However,
there is currently no dedicated datasets aiming to test these abilities. Herein, we present RiddleSense,
a new multiple-choice question answering task, which comes with the first large dataset (5.7k examples) for answering
riddle-style commonsense questions. We systematically evaluate a wide range of models over the challenge,
and point out that there is a large gap between the best-supervised model and human performance — suggesting
intriguing future research in the direction of higher-order commonsense reasoning and linguistic creativity towards
building advanced NLU systems.riddles-qa
Riddles QA
This dataset contains 500 riddles and their answers.
Credits
The riddles were downloaded from https://github.com/nkilm/riddles-api
riddles-dedupedriddlesense_plusplus
RiddleSense ++: Evaluating LLMs' Riddling abilities
Some cleaning and other modifications to make the riddle_sense dataset more suitable for training language models to generate riddles via standard text generation
Notable changes
reformatting to use special tags indicating the question and answer components
normalization: whitespace normalization, etc with clean-text
spell correction the original dataset has numerous spelling errors; these are fixed using BertChecker… See the full description on the dataset page: https://huggingface.co/datasets/pszemraj/riddlesense_plusplus.