# REALM: An MLLM-Agent Framework for Open World 3D Reasoning Segmentation and Editing on Gaussian Splatting

arXiv: https://arxiv.org/abs/2510.16410 · HTML: https://mlsanity.com/paper/2510.16410/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 39 of its 54 references that are in the map.
- First posted on arXiv 18 Oct 2025.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)
- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)

## Builds on (most cited first, 30 of 39)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [Instant Neural Graphics Primitives with a Multiresolution Hash Encoding](https://mlsanity.com/paper/2201.05989/) (2022)
- [Plenoxels: Radiance Fields without Neural Networks](https://mlsanity.com/paper/2112.05131/) (2021)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [TensoRF: Tensorial Radiance Fields](https://mlsanity.com/paper/2203.09517/) (2022)
- [Grounded SAM: Assembling Open-World Models for Diverse Visual Tasks](https://mlsanity.com/paper/2401.14159/) (2024)
- [LISA: Reasoning Segmentation via Large Language Model](https://mlsanity.com/paper/2308.00692/) (2023)
- [LERF: Language Embedded Radiance Fields](https://mlsanity.com/paper/2303.09553/) (2023)
- [LangSplat: 3D Language Gaussian Splatting](https://mlsanity.com/paper/2312.16084/) (2023)
- [Gaussian Grouping: Segment and Edit Anything in 3D Scenes](https://mlsanity.com/paper/2312.00732/) (2023)
- [Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields](https://mlsanity.com/paper/2312.03203/) (2023)
- [FSGS: Real-Time Few-shot View Synthesis using Gaussian Splatting](https://mlsanity.com/paper/2312.00451/) (2023)
- [Language Embedded 3D Gaussians for Open-Vocabulary Scene Understanding](https://mlsanity.com/paper/2311.18482/) (2023)
- [Tracking Anything with Decoupled Video Segmentation](https://mlsanity.com/paper/2309.03903/) (2023)
- [Guiding Instruction-based Image Editing via Multimodal Large Language Models](https://mlsanity.com/paper/2309.17102/) (2023)
- [Segment Any 3D Gaussians](https://mlsanity.com/paper/2312.00860/) (2023)
- [Weakly Supervised 3D Open-vocabulary Segmentation](https://mlsanity.com/paper/2305.14093/) (2023)
- [Is Out-of-Distribution Detection Learnable?](https://mlsanity.com/paper/2210.14707/) (2022)
- [GARField: Group Anything with Radiance Fields](https://mlsanity.com/paper/2401.09419/) (2024)
- [GaussianGrasper: 3D Language Gaussian Splatting for Open-vocabulary Robotic Grasping](https://mlsanity.com/paper/2403.09637/) (2024)
- [Robot See Robot Do: Imitating Articulated Object Manipulation with Monocular 4D Reconstruction](https://mlsanity.com/paper/2409.18121/) (2024)

## Built on by (most cited first, 6 of 6)

- [Sparse4DGS: 4D Gaussian Splatting for Sparse-Frame Dynamic Scene Reconstruction](https://mlsanity.com/paper/2511.07122/) (2025)
- [SRSplat: Feed-Forward Super-Resolution Gaussian Splatting from Sparse Multi-View Images](https://mlsanity.com/paper/2511.12040/) (2025)
- [DenoiseSplat: Feed-Forward Gaussian Splatting for Noisy 3D Scene Reconstruction](https://mlsanity.com/paper/2603.09291/) (2026)
- [DeblurNVS: Geometric Latent Diffusion for Novel View Synthesis from Sparse Motion-Blurred Images](https://mlsanity.com/paper/2606.01315/) (2026)
- [CausalSplat: Towards Comprehensive Hierarchical Reasoning in 3D Gaussian Splatting](https://mlsanity.com/paper/2608.11150/) (2026)
- [GroupForward: Building Referable 3D Scenes via Instance-Grouped Feed-Forward Gaussian Splatting](https://mlsanity.com/paper/2608.17535/) (2026)

## Abstract (arXiv)

Bridging the gap between complex human instructions and precise 3D object grounding remains a significant challenge in vision and robotics. Existing 3D segmentation methods often struggle to interpret ambiguous, reasoning-based instructions, while 2D vision-language models that excel at such reasoning lack intrinsic 3D spatial understanding. In this paper, we introduce REALM, an innovative MLLM-agent framework that enables open-world reasoning-based segmentation without requiring extensive 3D-specific post-training. We perform segmentation directly on 3D Gaussian Splatting representations, capitalizing on their ability to render photorealistic novel views that are highly suitable for MLLM comprehension. As directly feeding one or more rendered views to the MLLM can lead to high sensitivity to viewpoint selection, we propose a novel Global-to-Local Spatial Grounding strategy. Specifically, multiple global views are first fed into the MLLM agent in parallel for coarse-level localization, aggregating responses to robustly identify the target object. Then, several close-up novel views of the object are synthesized to perform fine-grained local segmentation, yielding accurate and consistent 3D masks. Extensive experiments show that REALM achieves remarkable performance in interpreting both explicit and implicit instructions across LERF, 3D-OVS, and our newly introduced REALM3D benchmarks. Furthermore, our agent framework seamlessly supports a range of 3D interaction tasks, including object removal, replacement, and style transfer, demonstrating its practical utility and versatility. Project page: https://ChangyueShi.github.io/REALM.