datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Mantis-Eval
Overview
This is a newly curated dataset to evaluate multimodal language models' capability to reason over multiple images. More details are shown in https://tiger-ai-lab.github.io/Mantis/.
Statistics
This evaluation dataset contains 217 human-annotated challenging multi-image reasoning problems.
Leaderboard
We list the current results as follows:
Models
Size
Mantis-Eval
LLaVA OneVision
72B
77.60
LLaVA OneVision
7B
64.20
GPT-4V
-
62.67… See the full description on the dataset page: https://huggingface.co/datasets/TIGER-Lab/Mantis-Eval.Manalyzer
Manalyzer: End-to-end Automated Meta-analysis with Multi-agent System
🆕 Updates/News
🚩 Updates (2026-02-09) Code has been released.
🚩 Updates (2025-05-22) Initial upload to arXiv [PDF]. The code will be released soon.
🎯 Abstract
Meta-analysis is a systematic research methodology that synthesizes data from multiple existing studies to derive comprehensive conclusions. This approach not only mitigates limitations inherent in individual studies but also… See the full description on the dataset page: https://huggingface.co/datasets/black-yt/Manalyzer.MANBench
MANBench
This repo contains evaluation code for the paper "MANBench: Is Your Multimodal Model Smarter than Human?" [ACL'25 Findings]
🌐 Homepage | 🤗 Dataset | 📑arXiv
🔔News
🔥[2025-05-16]: MANBench is accepted to [ACL'25 Findings]
[2025-04-29]: We added the scores and outputs for GPT-o1 and InternVL2.5-78B-MPO
[2025-02-09]: MANBench is now available on 🤗 Hugging Face
Introduction
We introduce MANBench (Multimodal Ability Norms Benchmark), a comprehensive… See the full description on the dataset page: https://huggingface.co/datasets/MANBench/MANBench.bandham-dataset-exploration
Bandham Dataset Exploration
Using the Dataset
Load the Dataset
from datasets import load_dataset
from huggingface_hub import hf_hub_download
from PIL import Image
# Download the dataset
ds = load_dataset("bandham-manikanta/bandham-dataset-exploration")
df = ds['train']
print(f"Dataset loaded with {len(df)} records")
Download Associated Figures
# To download and view an image for any row:row_index = 0 # Change this to any row index you want
#… See the full description on the dataset page: https://huggingface.co/datasets/bandham-manikanta/bandham-dataset-exploration.mande-ancient-treasures-de-grunne-van-dyke-2016
mande-ancient-treasures-de-grunne-van-dyke-2016
Dataset created with PDF2Dataset -- OCR + structure-aware chunking pipeline.
Dataset Summary
Metric
Value
Total chunks
268
Avg chars/chunk
722
Avg images/chunk
0.14
Source files
1
Duplicates removed
0
Quality filtered
6
Schema
Column
Type
Description
chunk_id
string
Unique identifier: filename_chunk_N
text
string
Raw markdown chunk with image refs
text_clean… See the full description on the dataset page: https://huggingface.co/datasets/Svngoku/mande-ancient-treasures-de-grunne-van-dyke-2016.