datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
SII_self_evovling_02_training_datasetLiME_dataLLaVA-ReCap-676KThis is an integrated version of LLaVA-ReCap, sourced from lmms-lab/LLaVA-ReCap-558K and lmms-lab/LLaVA-ReCap-118K.
In this version, the conversations field has been split into two separate fields: prompt and response. Additionally, the <image> special token has been removed to facilitate customization.
Inspired by the original paper, the prompt field has been further expanded with human-crafted variations. Specifically, each prompt is sampled from one of the following 30 instructions:… See the full description on the dataset page: https://huggingface.co/datasets/LimeryJorge/LLaVA-ReCap-676K.Synthetic_Unanswerable_Math
Dataset Card for Synthetic Unanswerable Math (SUM)
Dataset Summary
Synthetic Unanswerable Math (SUM) is a dataset of high-quality, implicitly unanswerable math problems constructed to probe and improve the refusal behavior of large language models (LLMs). The goal is to teach models to identify when a problem cannot be answered due to incomplete, ambiguous, or contradictory information, and respond with epistemic humility (e.g., \boxed{I don't know}).
Each entry in the… See the full description on the dataset page: https://huggingface.co/datasets/lime-nlp/Synthetic_Unanswerable_Math.russian-handwriting-ocr
Russian Handwritten Text Recognition Dataset
Датасет для распознавания русских рукописных текстов (сочинений).
Описание
Этот датасет содержит изображения рукописных русских текстов с их расшифровкой.
Предназначен для дообучения vision-language моделей (например, Qwen3 VL) на задачу OCR русского рукописного текста.
Статистика
Всего образцов: 13050
Train: 11745
Validation: 1305
Уникальных текстов: 575
Средняя длина текста: 3790 символов
Типы изображений… See the full description on the dataset page: https://huggingface.co/datasets/Limerencii/russian-handwriting-ocr.DeepScaleR_Difficulty
Difficulty Estimation on DeepScaleR
We annotate the entire DeepScaleR dataset with a difficulty score based on the performance of the Qwen 2.5-MATH-7B model. This provides an adaptive signal for curriculum construction and model evaluation.
DeepScaleR is a curated dataset of 40,000 reasoning-intensive problems used to train and evaluate reinforcement learning-based methods for large language models.
Difficulty Scoring Method
Difficulty scores are estimated using the… See the full description on the dataset page: https://huggingface.co/datasets/lime-nlp/DeepScaleR_Difficulty.infovqalimesoda-tha-classification
LimeSoda_tha_Classification
Deduplicated copy of kornwtp/limesoda-tha-classification.
Splits
split
rows
test
2,723
train
2,648
validation
299
GSM8K_Difficulty
Difficulty Estimation on DeepScaleR
We annotate the entire GSM8K dataset with a difficulty score based on the performance of the Qwen 2.5-MATH-7B model. This provides an adaptive signal for curriculum construction and model evaluation.
GSM8K (Grade School Math 8K) is a dataset of 8.5K high quality linguistically diverse grade school math word problems. The dataset was created to support the task of question answering on basic mathematical problems that require multi-step reasoning.… See the full description on the dataset page: https://huggingface.co/datasets/lime-nlp/GSM8K_Difficulty.safer-instruct
Safer-Instruct: Aligning Language Models with Automated Preference Data
This repository contains the dataset for the paper titled "Safer-Instruct: Aligning Language Models with Automated Preference Data". Check out our project website here!
Abstract
Reinforcement learning from human feedback (RLHF) is a vital strategy for enhancing model capability in language models. However, annotating preference data for RLHF is a resource-intensive and creativity-demanding process… See the full description on the dataset page: https://huggingface.co/datasets/lime-nlp/safer-instruct.limesoda-tha-classificationopenhermes-reasoning-231k
🧠 OpenHermes Reasoning 231K
High-quality instruction dataset with chain-of-thought reasoning
🤗 Dataset • 💬 Discussions
📊 Dataset Overview
This dataset contains 231,144 high-quality instruction-response pairs with explicit chain-of-thought reasoning. Each example includes:
Prompt: Original instruction or question
Thinking: Explicit reasoning process and logical steps
Answer: Final comprehensive response
Key Features
✅ Quality Filtered: Rigorous… See the full description on the dataset page: https://huggingface.co/datasets/limeXx/openhermes-reasoning-231k.LimeStory-1.0NEW IN THIS DATASETLimeStory Dataset Version 1.0 is now available in 🤗 Spaces and users can add stories about anything! (Powered by Pollinations.ai)
NOTICELimeStory is not for training NSFW models, and remember to use dataset: kulia-moon/LimeStory-1.0 for you're using this dataset as target training!
Kulia's datasets
The story,
your impossible
Generated by 🤗 Spaces
Protected by 🤗 Scanner
.hf-sanitized.hf-sanitized-yvTjnu7IKxZ1msJf6A22y .cursive { font-family: "Lobster"… See the full description on the dataset page: https://huggingface.co/datasets/kulia-moon/LimeStory-1.0.ChartQAorz_math_difficulty
Difficulty Estimation on Open Reasoner Zero
We annotate the entire Open Reasoner Zero dataset with a difficulty score based on the performance of the Qwen 2.5-MATH-7B model. This provides an adaptive signal for curriculum construction.
Open Reasoner Zero is a curated a dataset of 57,000 reasoning-intensive problems used to train and evaluate reinforcement learning-based methods for large language models.
Difficulty Scoring Method
Difficulty scores are estimated using… See the full description on the dataset page: https://huggingface.co/datasets/lime-nlp/orz_math_difficulty.ag_newsAG's News Topic Classification Dataset
Version 3, Updated 09/09/2015
ORIGIN
AG is a collection of more than 1 million news articles. News articles have been gathered from more than 2000 news sources by ComeToMyHead in more than 1 year of activity. ComeToMyHead is an academic news search engine which has been running since July, 2004. The dataset is provided by the academic comunity for research purposes in data mining (clustering, classification, etc), information retrieval (ranking, search… See the full description on the dataset page: https://huggingface.co/datasets/limen231/ag_news.MATH_Difficulty
Difficulty Estimation on MATH
We annotate the entire MATH dataset with a difficulty score based on the performance of the Qwen 2.5-MATH-7B model. This provides an adaptive signal for curriculum construction and model evaluation.
The Mathematics Aptitude Test of Heuristics (MATH) dataset consists of problems from mathematics competitions, including the AMC 10, AMC 12, AIME, and more. Each problem in MATH has a full step-by-step solution, which can be used to teach models to generate… See the full description on the dataset page: https://huggingface.co/datasets/lime-nlp/MATH_Difficulty.ai2dOK-VQATextCapslimesoda-tha-classificationlime-nlp-difficulty
lime-nlp Difficulty Estimation Math Datasets collection
Unofficial reformatted version of lime-nlp/difficulty-estimation-math-datasets,
which contains math problems and the Qwen 2.5 7B MATH model's success rates at solving those problems.
The combined dataset has been split into 80% training and 20% testing data.
Fields:
row_id: the row number of each dataset entry, starting at 0
input: the math question from the dataset
output: the correct answer (ground truth)… See the full description on the dataset page: https://huggingface.co/datasets/agentlans/lime-nlp-difficulty.Speeding-up-LIME
Deidentified Medical Charts with Human Curated Explanations
About
This dataset is a small sample from the EHR dataset used by experiments described in our paper, "Speeding up LIME with Attention Weights," submitted to CoDS-COMAD 2024.
COCO-Caption2017textvqaLime-v72Language: Korean
Size: 821 examples (약 2MB, JSONL)
Format: OpenAI Chat Format (messages list)
Target Task: Persona Alignment, Multi-turn Context Tracking, Reasoning, State Overwriting
분리된 사고 과정 (Internal Reasoning vs Visible Output)
모델이 즉각적으로 답을 뱉지 않고, 내부적으로 판단을 거친 뒤 정제된 답변만 출력하도록 학습시킵니다.
/think ~ </think> : 사용자의 의도를 파악하고 현재 상태를 점검하는 내부 사고 영역.
<|lime_think_more|> ~ <|lime_end_think|> : 대화가 길어지거나 복잡한 요청일 경우, 스스로 제약 조건(예: 명사구 누락 금지)을 한 번 더 점검하는 추가 사고 영역.
<|lime_final|> : 최종적으로 사용자에게 노출되는 정제된… See the full description on the dataset page: https://huggingface.co/datasets/naksyu/Lime-v72.ocrbenchlime-explanations-finallime-explanations-final-3.0POPE