Skip to content

Update README.md : Add RoboRefer, a 3D spatial reasoning model with advanced dataset & benchmark #254

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Anjingkun wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Update README.md : Add RoboRefer, a 3D spatial reasoning model with advanced dataset & benchmark #254

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -116,6 +116,7 @@ This is the first work to correct hallucination in multimodal large language mod
## Multimodal Instruction Tuning
| Title | Venue | Date | Code | Demo |
|:--------|:--------:|:--------:|:--------:|:--------:|
| ![Star](https://img.shields.io/github/stars/Zhoues/RoboRefer.svg?style=social&label=Star) <br> [**RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics**](https://arxiv.org/abs/2506.04308) <br> | arXiv | 2025-06-04 | [Github](https://github.com/Zhoues/RoboRefer) | [Demo](https://huggingface.co/Zhoues/RoboRefer-2B-SFT) |
| ![Star](https://img.shields.io/github/stars/THUDM/GLM-4.1V-Thinking.svg?style=social&label=Star) <br> [**GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning**](https://arxiv.org/pdf/2507.01006) <br> | arXiv | 2025-07-02 | [Github](https://github.com/THUDM/GLM-4.1V-Thinking) | [Demo](https://huggingface.co/spaces/THUDM/GLM-4.1V-9B-Thinking-API-Demo) |
| ![Star](https://img.shields.io/github/stars/lxtGH/DenseWorld-1M.svg?style=social&label=Star) <br> [**DenseWorld-1M: Towards Detailed Dense Grounded Caption in the Real World**](https://arxiv.org/pdf/2506.24102) <br> | arXiv | 2025-06-30 | [Github](https://github.com/lxtGH/DenseWorld-1M) | - |
| [**Qwen VLo: From "Understanding" the World to "Depicting" It**](https://qwenlm.github.io/blog/qwen-vlo/) | Qwen | 2025-06-26 | - | [Demo](https://chat.qwen.ai/) |
Expand Down Expand Up @@ -441,6 +442,7 @@ This is the first work to correct hallucination in multimodal large language mod
## LLM-Aided Visual Reasoning
| Title | Venue | Date | Code | Demo |
|:--------|:--------:|:--------:|:--------:|:--------:|
| ![Star](https://img.shields.io/github/stars/Zhoues/RoboRefer.svg?style=social&label=Star) <br> [**RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics**](https://arxiv.org/abs/2506.04308) <br> | arXiv | 2025-06-04 | [Github](https://github.com/Zhoues/RoboRefer) | [Demo](https://huggingface.co/Zhoues/RoboRefer-2B-SFT) |
| ![Star](https://img.shields.io/github/stars/yhy-2000/VideoDeepResearch.svg?style=social&label=Star) <br> [**VideoDeepResearch: Long Video Understanding With Agentic Tool Using**](https://arxiv.org/pdf/2506.10821) <br> | arXiv | 2025-06-12 | [Github](https://github.com/yhy-2000/VideoDeepResearch) | Local Demo |
| ![Star](https://img.shields.io/github/stars/LaVi-Lab/Visual-Table.svg?style=social&label=Star) <br> [**Beyond Embeddings: The Promise of Visual Table in Multi-Modal Models**](https://arxiv.org/pdf/2403.18252.pdf) <br> | arXiv | 2024-03-27 | [Github](https://github.com/LaVi-Lab/Visual-Table) | - |
| ![Star](https://img.shields.io/github/stars/penghao-wu/vstar.svg?style=social&label=Star) <br> [**V∗: Guided Visual Search as a Core Mechanism in Multimodal LLMs**](https://arxiv.org/pdf/2312.14135.pdf) <br> | arXiv | 2023-12-21 | [Github](https://github.com/penghao-wu/vstar) | Local Demo |
Expand Down Expand Up @@ -477,6 +479,7 @@ This is the first work to correct hallucination in multimodal large language mod
## Foundation Models
| Title | Venue | Date | Code | Demo |
|:--------|:--------:|:--------:|:--------:|:--------:|
| ![Star](https://img.shields.io/github/stars/Zhoues/RoboRefer.svg?style=social&label=Star) <br> [**RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics**](https://arxiv.org/abs/2506.04308) <br> | arXiv | 2025-06-04 | [Github](https://github.com/Zhoues/RoboRefer) | [Demo](https://huggingface.co/Zhoues/RoboRefer-2B-SFT) |
| ![Star](https://img.shields.io/github/stars/DAMO-NLP-SG/VideoLLaMA3.svg?style=social&label=Star) <br> [**VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding**](https://arxiv.org/pdf/2501.13106) <br> | arXiv | 2025-01-22 | [Github](https://github.com/DAMO-NLP-SG/VideoLLaMA3) | [Demo](https://huggingface.co/spaces/lixin4ever/VideoLLaMA3) |
| ![Star](https://img.shields.io/github/stars/baaivision/Emu3.svg?style=social&label=Star) <br> [**Emu3: Next-Token Prediction is All You Need**](https://arxiv.org/pdf/2409.18869) <br> | arXiv | 2024-09-27 | [Github](https://github.com/baaivision/Emu3) | Local Demo |
| [**Llama 3.2: Revolutionizing edge AI and vision with open, customizable models**](https://ai.meta.com/blog/llama-3-2-connect-2024-vision-edge-mobile-devices/) | Meta | 2024-09-25 | - | [Demo](https://huggingface.co/meta-llama/Llama-3.2-11B-Vision-Instruct) |
Expand Down Expand Up @@ -511,6 +514,7 @@ This is the first work to correct hallucination in multimodal large language mod
## Evaluation
| Title | Venue | Date | Page |
|:--------|:--------:|:--------:|:--------:|
| ![Star](https://img.shields.io/github/stars/Zhoues/RoboRefer.svg?style=social&label=Star) <br> [**RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics**](https://arxiv.org/abs/2506.04308) <br> | arXiv | 2025-06-04 | [Github](https://zhoues.github.io/RoboRefer/) |
| ![Stars](https://img.shields.io/github/stars/lerogo/MMGenBench?style=social&label=Star) <br> [**Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces**](https://arxiv.org/pdf/2412.14171) <br> | arXiv | 2024-12-18 | [Github](https://github.com/vision-x-nyu/thinking-in-space) |
| ![Stars](https://img.shields.io/github/stars/lerogo/MMGenBench?style=social&label=Star) <br> [**MMGenBench: Evaluating the Limits of LMMs from the Text-to-Image Generation Perspective**](https://arxiv.org/pdf/2411.14062) <br> | arXiv | 2024-11-21 | [Github](https://github.com/lerogo/MMGenBench) |
| ![Stars](https://img.shields.io/github/stars/multimodal-art-projection/OmniBench?style=social&label=Star) <br> [**OmniBench: Towards The Future of Universal Omni-Language Models**](https://arxiv.org/pdf/2409.15272) <br> | arXiv | 2024-09-23 | [Github](https://github.com/multimodal-art-projection/OmniBench) |
Expand Down Expand Up @@ -614,6 +618,7 @@ This is the first work to correct hallucination in multimodal large language mod
## Datasets of Multimodal Instruction Tuning
| Name | Paper | Link | Notes |
|:-----|:-----:|:----:|:-----:|
| **RefSpatial** | [RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics](https://arxiv.org/abs/2506.04308)| [Link](https://huggingface.co/datasets/JingkunAn/RefSpatial) | A 3D spatial relationships and object referring dataset, containing 31 distinct spatial relations and 2.5 million samples |
| **Inst-IT Dataset** | [Inst-IT: Boosting Multimodal Instance Understanding via Explicit Visual Prompt Instruction Tuning](https://arxiv.org/pdf/2412.03565) | [Link](https://github.com/inst-it/inst-it) | An instruction-tuning dataset which contains fine-grained multi-level annotations for 21k videos and 51k images |
| **E.T. Instruct 164K** | [E.T. Bench: Towards Open-Ended Event-Level Video-Language Understanding](https://arxiv.org/pdf/2409.18111) | [Link](https://github.com/PolyU-ChenLab/ETBench) | An instruction-tuning dataset for time-sensitive video understanding |
| **MSQA** | [Multi-modal Situated Reasoning in 3D Scenes](https://arxiv.org/pdf/2409.02389) | [Link](https://msr3d.github.io/) | A large scale dataset for multi-modal situated reasoning in 3D scenes |
Expand Down Expand Up @@ -678,6 +683,7 @@ This is the first work to correct hallucination in multimodal large language mod
## Benchmarks for Evaluation
| Name | Paper | Link | Notes |
|:-----|:-----:|:----:|:-----:|
| **RefSpatial-Bench** | [RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics](https://arxiv.org/abs/2506.04308)| [Link](https://huggingface.co/datasets/BAAI/RefSpatial-Bench) | A new benchmark that fills the gap in evaluating spatial referring with multi-step reasoning. Over 70% of the tasks require multi-step reasoning (up to 5 steps) |
| **Inst-IT Bench** | [Inst-IT: Boosting Multimodal Instance Understanding via Explicit Visual Prompt Instruction Tuning](https://arxiv.org/pdf/2412.03565) | [Link](https://github.com/inst-it/inst-it) | A benchmark to evaluate fine-grained instance-level understanding in images and videos |
| **M<sup>3</sup>CoT** | [M<sup>3</sup>CoT: A Novel Benchmark for Multi-Domain Multi-step Multi-modal Chain-of-Thought](https://arxiv.org/pdf/2405.16473) | [Link](https://github.com/LightChen233/M3CoT) | A multi-domain, multi-step benchmark for multimodal CoT |
| **MMGenBench** | [MMGenBench: Evaluating the Limits of LMMs from the Text-to-Image Generation Perspective](https://arxiv.org/pdf/2411.14062) | [Link](https://github.com/lerogo/MMGenBench) | A benchmark that gauges the performance of constructing image-generation prompt given an image |
Expand Down