Datasets:

xiuhuywh
/

DRIM-VisualReasonHard

Modalities:

Formats:

Size:

ArXiv:

Libraries:

License:

Dataset card Data Studio Files Files and versions

Dataset Viewer

The dataset viewer is not available for this dataset.

Unexpected token '<', "<html> <h"... is not valid JSON

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

This repository contains the RL training datasets used in the paper Deep But Reliable: Advancing Multi-turn Reasoning for Thinking with Images

Downloads last month: 383

Models trained or fine-tuned on xiuhuywh/DRIM-VisualReasonHard

Below-Image/Poralus-Image-1

Text-to-Image • Updated Jan 17 • 9 • 4

Mimi1782/NANSEN

Visual Question Answering • Updated Jan 28

Mimi1782/KNN

Tabular Regression • Updated Feb 1

Mimi1782/SVM

Document Question Answering • Updated Feb 5

Menoviar28/menov99

Paper for xiuhuywh/DRIM-VisualReasonHard

Deep But Reliable: Advancing Multi-turn Reasoning for Thinking with Images

Paper • 2512.17306 • Published Dec 19, 2025 • 3