Shadowgraph
TrendingBuzzAbout
Sign in

StevenHH2000/Fine-R1-Stage2-data

Dataset

This is the official release of the training dataset of Stage 2 (TAPO) for paper Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning. Code is available at https://github.com/PKU-ICST-MIPL/FineR1_ICLR2026.

1

3 commits

1 linked in READMEs

updated Feb 11, 2026

See the code

README

This is the official release of the training dataset of Stage 2 (TAPO) for paper Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning. Code is available at https://github.com/PKU-ICST-MIPL/FineR1_ICLR2026.

Data Source

Training

  • We randomly sample 4-shot data per seen category from the following 6 datasets to construct our TAPO training dataset.
    • FGVC-Aircraft ✈️
    • CaltechUCSD Bird-200 🦤
    • Stanford Car-196 🚙
    • Stanford Dog-120 🐕
    • Flower-102 🌼
    • Oxford-IIIT Pet-37 🐈

Data Fields

  • images: input image(s)
    • data type: List
  • problem: input question or statement
    • data type: String
  • answer: ground-truth answer
    • data type: String
  • image_noisy: positive image(s) from the same sub-category
    • data type: List
  • image_aug: negative image(s) from a different sub-category
    • data type: List
multimodal
reinforcement-learning
vision

Similar repos

StevenHH2000/iNat21-1shot-fewshots

Dataset

This is the official release of the training dataset for paper Taxonomy-Aware Representation…

1

WaltonFuture/Multimodal-RL-Data

Dataset

Multimodal-Cold-Start Dataset

8

PKU-ICST-MIPL/FineR1_ICLR2026

78

cythu/PeBR_R1_dataset

Dataset

📘 PeBR-R1 Dataset

0

WaltonFuture/Multimodal-Cold-Start

Dataset

Multimodal-Cold-Start Dataset: Advancing Multimodal Reasoning via Reinforcement Learning with Cold…

11

GD-ML/UniVG-R1-data

Dataset

UniVG-R1 Model Card

4

WaltonFuture/geometry3k-direct-synthesizing

Dataset

This dataset is used in the paper Unsupervised Post-Training for Multi-Modal LLM Reasoning via GRPO…

1

JefferyZhan/Vision-R1-Data

Dataset

This repository contains the dataset used in the paper Vision-R1: Evolving Human-Free Alignment in…

3

Linked in READMEs

PKU-ICST-MIPL/FineR1_ICLR2026

78

README badge

StevenHH2000/Fine-R1-Stage2-data

Dataset

This is the official release of the training dataset of Stage 2 (TAPO) for paper Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning. Code is available at https://github.com/PKU-ICST-MIPL/FineR1_ICLR2026.

1

3 commits

1 linked in READMEs

updated Feb 11, 2026

See the code

README

This is the official release of the training dataset of Stage 2 (TAPO) for paper Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning. Code is available at https://github.com/PKU-ICST-MIPL/FineR1_ICLR2026.

Data Source

Training

  • We randomly sample 4-shot data per seen category from the following 6 datasets to construct our TAPO training dataset.
    • FGVC-Aircraft ✈️
    • CaltechUCSD Bird-200 🦤
    • Stanford Car-196 🚙
    • Stanford Dog-120 🐕
    • Flower-102 🌼
    • Oxford-IIIT Pet-37 🐈

Data Fields

  • images: input image(s)
    • data type: List
  • problem: input question or statement
    • data type: String
  • answer: ground-truth answer
    • data type: String
  • image_noisy: positive image(s) from the same sub-category
    • data type: List
  • image_aug: negative image(s) from a different sub-category
    • data type: List
multimodal
reinforcement-learning
vision

Similar repos

StevenHH2000/iNat21-1shot-fewshots

Dataset

This is the official release of the training dataset for paper Taxonomy-Aware Representation…

1

WaltonFuture/Multimodal-RL-Data

Dataset

Multimodal-Cold-Start Dataset

8

PKU-ICST-MIPL/FineR1_ICLR2026

78

cythu/PeBR_R1_dataset

Dataset

📘 PeBR-R1 Dataset

0

WaltonFuture/Multimodal-Cold-Start

Dataset

Multimodal-Cold-Start Dataset: Advancing Multimodal Reasoning via Reinforcement Learning with Cold…

11

GD-ML/UniVG-R1-data

Dataset

UniVG-R1 Model Card

4

WaltonFuture/geometry3k-direct-synthesizing

Dataset

This dataset is used in the paper Unsupervised Post-Training for Multi-Modal LLM Reasoning via GRPO…

1

JefferyZhan/Vision-R1-Data

Dataset

This repository contains the dataset used in the paper Vision-R1: Evolving Human-Free Alignment in…

3

Linked in READMEs

PKU-ICST-MIPL/FineR1_ICLR2026

78

README badge