3D-RFT: Reinforcement Fine-Tuning for Video-based 3D Scene Understanding

• URL: http://arxiv.org/abs/2603.04976v1
• Date: Thu, 05 Mar 2026 09:15:16 GMT
• Title: 3D-RFT: Reinforcement Fine-Tuning for Video-based 3D Scene Understanding
• Authors: Xiongkun Linghu, Jiangyong Huang, Baoxiong Jia, Siyuan Huang,
• Abstract summary: We present Reinforcement Fine-Tuning for Video-based 3D Scene Understanding (3D-RFT)<n>3D-RFT is first framework to extend RLVR to video-based 3D perception and reasoning.<n>We show that 3D-RFT-4B achieves state-of-the-art performance on various video-based 3D scene understanding tasks.
• Score: 21.70953326671503
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Reinforcement Learning with Verifiable Rewards ( RLVR ) has emerged as a transformative paradigm for enhancing the reasoning capabilities of Large Language Models ( LLMs), yet its potential in 3D scene understanding remains under-explored. Existing approaches largely rely on Supervised Fine-Tuning ( SFT), where the token-level cross-entropy loss acts as an indirect proxy for optimization, leading to a misalignment between training objectives and task performances. To bridge this gap, we present Reinforcement Fine-Tuning for Video-based 3D Scene Understanding (3D-RFT ), the first framework to extend RLVR to video-based 3D perception and reasoning. 3D-RFT shifts the paradigm by directly optimizing the model towards evaluation metrics. 3D-RFT first activates 3D-aware Multi-modal Large Language Models ( MLLM s) via SFT, followed by reinforcement fine-tuning using Group Relative Policy Optimization ( GRPO) with strictly verifiable reward functions. We design task-specific reward functions directly from metrics like 3D IoU and F1-Score to provide more effective signals to guide model training. Extensive experiments demonstrate that 3D-RFT-4B achieves state-of-the-art performance on various video-based 3D scene understanding tasks. Notably, 3D-RFT-4B significantly outperforms larger models (e.g., VG LLM-8B) on 3D video detection, 3D visual grounding, and spatial reasoning benchmarks. We further reveal good properties of 3D-RFT such as robust efficacy, and valuable insights into training strategies and data impact. We hope 3D-RFT can serve as a robust and promising paradigm for future development of 3D scene understanding.

Related papers

• Reasoning Matters for 3D Visual Grounding [39.725360883988515]
  We propose a 3D visual grounding data pipeline, which is capable of automatically synthesizing 3D visual grounding data along with corresponding reasoning process.<n>We also introduce Reason3DVG-8B, a strong 3D visual grounding LLM that outperforms previous LLM-based method 3D-GRAND using only 1.6% of their training data.
  arXiv  Detail & Related papers  (2026-01-13T18:48:41Z)
• D3D-VLP: Dynamic 3D Vision-Language-Planning Model for Embodied Grounding and Navigation [66.7166217399105]
  Embodied agents face a critical dilemma that end-to-end models lack interpretability and explicit 3D reasoning.<n>Our model introduces two key innovations: 1) A Dynamic 3D Chain-of-Thought (3D CoT) that unifies planning, grounding, navigation, and question answering within a single 3D-VLM and CoT pipeline; 2) A Synergistic Learning from Fragmented Supervision (SLFS) strategy, which uses a masked autoregressive loss to learn from massive and partially-annotated hybrid data.
  arXiv  Detail & Related papers  (2025-12-14T09:53:15Z)
• Abstract 3D Perception for Spatial Intelligence in Vision-Language Models [100.13033631690114]
  Vision-language models (VLMs) struggle with 3D-related tasks such as spatial cognition and physical understanding.<n>We introduce SandboxVLM, a framework that leverages abstract bounding boxes to encode geometric structure and physical kinematics for VLM.<n>Our approach consistently improves spatial intelligence, achieving an 8.3% gain on SAT Real compared with baseline methods.
  arXiv  Detail & Related papers  (2025-11-14T04:16:09Z)
• Vid-LLM: A Compact Video-based 3D Multimodal LLM with Reconstruction-Reasoning Synergy [4.1703677379815565]
  We propose Vid-LLM, a video-based 3D-MLLM that directly processes video inputs without requiring external 3D data.<n>In our method, the geometric prior are directly used to improve the performance of the sceen perception.<n>Experiments across diverse benchmarks verified the effectiveness of our method on 3D Question Answering, 3D Captioning and 3D Visual Grounding tasks.
  arXiv  Detail & Related papers  (2025-09-29T07:34:18Z)
• 3D-R1: Enhancing Reasoning in 3D VLMs for Unified Scene Understanding [11.069512983766783]
  Large vision-language models (VLMs) have made significant strides in 2D visual understanding tasks.<n>We propose 3D-R1, a foundation model that enhances the reasoning capabilities of 3D VLMs.<n>Extensive experiments demonstrate that 3D-R1 delivers an average improvement of 10% across various 3D scene benchmarks.
  arXiv  Detail & Related papers  (2025-07-31T11:59:06Z)
• TriCLIP-3D: A Unified Parameter-Efficient Framework for Tri-Modal 3D Visual Grounding based on CLIP [52.79100775328595]
  3D visual grounding allows an embodied agent to understand visual information in real-world 3D environments based on human instructions.<n>Existing 3D visual grounding methods rely on separate encoders for different modalities.<n>We propose a unified 2D pre-trained multi-modal network to process all three modalities.
  arXiv  Detail & Related papers  (2025-07-20T10:28:06Z)
• MLLMs Need 3D-Aware Representation Supervision for Scene Understanding [14.083262551714133]
  3DRS is a framework that enhances MLLM 3D representation learning by introducing supervision from pretrained 3D foundation models.<n>Our approach aligns MLLM visual features with rich 3D knowledge distilled from 3D models, effectively improving scene understanding.
  arXiv  Detail & Related papers  (2025-06-02T17:58:24Z)
• LLMI3D: MLLM-based 3D Perception from a Single 2D Image [77.13869413871028]
  multimodal large language models (MLLMs) excel in general capacity but underperform in 3D tasks.<n>In this paper, we propose solutions for weak 3D local spatial object perception, poor text-based geometric numerical output, and inability to handle camera focal variations.<n>We employ parameter-efficient fine-tuning for a pre-trained MLLM and develop LLMI3D, a powerful 3D perception MLLM.
  arXiv  Detail & Related papers  (2024-08-14T10:00:16Z)
• PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm [111.16358607889609]
  We introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation.<n>For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness.
  arXiv  Detail & Related papers  (2023-10-12T17:59:57Z)
• Exemplar Fine-Tuning for 3D Human Model Fitting Towards In-the-Wild 3D Human Pose Estimation [107.07047303858664]
  Large-scale human datasets with 3D ground-truth annotations are difficult to obtain in the wild. We address this problem by augmenting existing 2D datasets with high-quality 3D pose fits. The resulting annotations are sufficient to train from scratch 3D pose regressor networks that outperform the current state-of-the-art on in-the-wild benchmarks.
  arXiv  Detail & Related papers  (2020-04-07T20:21:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.