SceneReVis: A Self-Reflective Vision-Grounded Framework for 3D Indoor Scene Synthesis via Multi-Turn RL

Yang Zhao; Shizhao Sun; Meisheng Zhang; Yingdong Shi; Xubo Yang; Jiang Bian

cs.CVFeb 10, 2026

SceneReVis: A Self-Reflective Vision-Grounded Framework for 3D Indoor Scene Synthesis via Multi-Turn RL

Authors: Yang Zhao, Shizhao Sun, Meisheng Zhang, Yingdong Shi, Xubo Yang, Jiang Bian

Organizations: Shanghai Jiao Tong University, Shanghai, China · Microsoft Research Asia, Beijing, China · Peking University, Beijing, China · ShanghaiTech University, Shanghai, China

Abstract

Current one-pass 3D scene synthesis methods often suffer from spatial hallucinations, such as collisions, due to a lack of deliberative reasoning. To bridge this gap, we introduce SceneReVis, a vision-grounded self-reflection framework that employs an iterative "diagnose-and-act" loop to explicitly intercept and resolve spatial conflicts using multi-modal feedback. To support this step-wise paradigm, we construct SceneChain-12k, a large-scale dataset of causal construction trajectories derived through a novel reverse engineering pipeline. We further propose a two-stage training recipe that transitions from Supervised Fine-Tuning to Agentic Reinforcement Learning, evolving the model into an active spatial planner. Extensive experiments demonstrate that SceneReVis achieves state-of-the-art performance in high-fidelity generation and goal-oriented optimization, with improved transfer to underrepresented room categories in the dataset and promising potential for extension to out-of-distribution room types.

Figures & tables

Figure 1. Comparison of generation paradigms. (a) One-Pass Generation lacks intermediate reasoning, leading to severe physical violations. (b) Post-Processing Generation tends to be trapped in local optima, yielding suboptimal visual and semantic quality. (c) SceneReVis (Ours) employs an self-reflection paradigm to ensure physical plausibility and aesthetic coherence. Comparison of one-pass, post-processing, and SceneReVis generation paradigms with corresponding scene quality outcomes.

Paradigm	Method	Input	Data-driven	Self-reflection	Visual Feedback	Collision Detection	Irregular Floor Support	Goal-oriented Optimization
One-Pass	LayoutGPT ( Feng et al., 2023 )	Text	✗	✗	✗	✗	✗	✗
	DiffuScene ( Tang et al., 2024 )	Text / Scene	✓	✗	✗	✗	✗	✗
	PhyScene ( Yang et al., 2024a )	Floor Plan	✓	✗	✗	✗	✓	✗
	ReSpace ( Bucher and Armeni, 2025 )	Text (+ Scene)	✓	✗	✗	✗	✓	✗
	OptiScene ( Yang et al., 2025b )	Text	✓	✗	✗	✗	✗	✗
Post-Processing	Holodeck ( Yang et al., 2024b )	Text	✗	✗	✗	✓	✗	✗

Table 1. High-level capability comparison in text-conditioned 3D indoor scene synthesis.

Figure 2. Overview of the SceneReVis learning framework. The pipeline consists of three stages: (a) Data Construction: We employ a Reverse Engineering strategy to decompose static 3D scenes into causal construction trajectories (SceneChain-12k). (b) Cold Start: A Supervised Fine-Tuning (SFT) stage initializes the agent with basic tool usage capabilities. (c) Legend: Legend explaining the contents of various parts. (d) Agentic RL: The core stage utilizes Group Relative Policy Optimization (GRPO). The agent interacts with a physics-enabled simulator via a "diagnose-and-act" loop, receiving multi-modal feedback and optimizing against non-differentiable objectives. Training pipeline of SceneReVis with data construction, SFT cold start, and agentic RL optimization.

Method	Bedroom						Living Room						Avg. (Bed + Living)
	Physics		Visual & Semantics				Physics		Visual & Semantics				Physics		Visual & Semantics
	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$
DiffuScene ( Tang et al., 2024 )	46.2	33.8	8.2	6.2	6.9	7.1	28.0	41.5	7.3	7.5	8.2	7.7	37.1	37.7	7.8	6.9	7.6	7.4
ReSpace ( Bucher and Armeni, 2025 )	14.7	41.9	6.8	6.1	7.0	6.6	11.5	40.3	6.9	6.6	7.3	6.9	13.1	41.1	6.9	6.4	7.2	6.8
LayoutGPT ( Feng et al., 2023 )	34.3	45.9	7.8	5.6	7.6	7.0	15.0	35.7	7.5	7.3	8.7	7.8	24.7	40.8	7.7	6.5	8.2	7.5
I-Design ( Çelen et al., 2024 )	15.5	18.2	8.3	6.6	8.1	7.7	17.8	14.1	8.3	6.5	8.6	7.8	16.7	16.2	8.3	6.6	8.4	7.8

Table 2. Quantitative comparison on Bedroom and Living Room. ‘Avg.‘ is the mean of Ra, Spa, and Ac. Best results are bolded , and second-best results are underlined .

Method	Dining Room						Study Room						Avg. (Dining + Study)
	Physics		Visual & Semantics				Physics		Visual & Semantics				Physics		Visual & Semantics
	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$
LayoutGPT ( Feng et al., 2023 )	6.2	26.2	6.9	7.7	8.8	7.8	16.6	52.1	8.5	7.2	8.3	8.0	11.4	39.2	7.7	7.5	8.6	7.9
I-Design ( Çelen et al., 2024 )	17.6	0.7	7.8	5.7	8.8	7.4	16.3	4.5	8.2	6.9	8.9	8.0	17.0	2.6	8.0	6.3	8.9	7.7
Holodeck ( Yang et al., 2024b )	11.9	3.3	6.4	8.0	8.5	7.6	13.7	5.1	6.7	8.4	8.9	8.0	12.8	4.2	6.6	8.2	8.7	7.8
LayoutVLM ( Sun et al., 2025 )	19.2	27.0	6.5	7.8	8.8	7.7	12.3	36.6	6.9	7.6	8.6	7.7	15.8	31.8	6.7	7.7	8.7	7.7

Table 3. Quantitative comparison on Dining Room and Study Room. ‘Avg.‘ is the mean of Ra, Spa, and Ac. Best results are bolded , and second-best results are underlined .

Panel of Figure 3 — Figure 3. Qualitative comparison of standard scenes. We visualize the 3D scenes generated by different methods for Bedroom (top) and Living Room (bottom) scenarios based on the given text prompts. Qualitative comparison of generated bedroom and living-room scenes across methods. Qualitative comparison of generated dining-room and study-room scenes across methods.

Method	Cond 1: Chaotic & Missing						Cond 2: Chaotic Only						Cond 3: Missing Only						Overall Average
	Physics		Visual & Semantics				Physics		Visual & Semantics				Physics		Visual & Semantics				Physics		Visual & Semantics
	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$	OBR $\downarrow$	CNR $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$
DiffuScene ( Tang et al., 2024 )	50.8	30.3	7.5	7.3	8.5	7.8	62.8	24.1	7.4	7.0	8.7	7.7	27.8	30.2	7.3	7.0	7.7	7.3	47.1	28.2	7.4	7.1	8.3	7.6
LayoutVLM ( Sun et al., 2025 )	10.0	39.7	6.7	6.3	8.7	7.2	8.1	63.1	7.2	6.6	8.3	7.4	6.5	31.0	6.5	6.4	8.7	7.2	8.2	44.6	6.8	6.4	8.6	7.3
Ours	1.1	1.8	8.8	8.1	8.9	8.6	2.7	1.3	7.7	7.9	8.5	8.0	1.1	1.0	7.9	7.8	8.7	8.1	1.6	1.4	8.1	7.9	8.7	8.2

Table 4. Quantitative comparison on Goal-oriented Scene Optimization tasks. ‘Avg.‘ is the mean of Ra, Spa, and Ac. Best results are bolded , and second-best results are underlined .

Figure 5. Qualitative comparisons on additional OOD categories. Additional qualitative comparisons of generated scenes across methods.

Figure 6. Qualitative results for goal-oriented scene optimization. We show trajectories under Chaotic & Missing , Chaotic Only , and Missing Only conditions. Optimization trajectories for three goal-oriented scene optimization settings.

Components				Metrics
Components				Physics		Visual & Semantics
SFT	Final Reward	Trajectory Reward	Physical Opt.	VBL $\downarrow$	Phy. $\downarrow$	Ra. $\uparrow$	Spa. $\uparrow$	Ac. $\uparrow$	Avg. $\uparrow$
✓	-	-	-	544.5	30.4	8.7	7.1	7.8	7.9
✓	✓	-	-	154.0	15.8	7.7	7.1	8.9	7.9
✓	✓	✓	-	210.9	21.9	9.1	7.7	8.7	8.5
✓	✓	✓	✓	25.6	3.3	9.4	8.0	8.9	8.8

Table 5. Ablation study of key components. VBL measures the volume of physical violations, and Phy. is the mean of OBR and CNR. Final Reward is the terminal reward, while Trajectory Reward is accumulated over multiple turns.

Method	Physical Plausibility $\downarrow$	Visual & Semantic Quality $\downarrow$
I-Design ( Çelen et al., 2024 )	4.63 $\pm$ 1.25	4.73 $\pm$ 1.35
LayoutVLM ( Sun et al., 2025 )	4.30 $\pm$ 1.39	4.15 $\pm$ 1.51
LayoutGPT ( Feng et al., 2023 )	3.97 $\pm$ 1.59	3.98 $\pm$ 1.54
Holodeck ( Yang et al., 2024b )	3.18 $\pm$ 1.71	3.21 $\pm$ 1.60
SceneWeaver ( Yang et al., 2025a )	2.71 $\pm$ 1.37	2.92 $\pm$ 1.52
SceneReVis (Ours)	2.20 $\pm$ 1.46	2.01 $\pm$ 1.15

Table 6. Human-ranking user study (lower is better).

3D Scene Generation

3D scene generation creates structured, explorable environments from text, images, maps, or other inputs.

See all 189 papers

ThinkBLOX: 3D Indoor Scene Generation with Progressive Reasoning
Jul 15, 2026Yuan Xiao, Can Wang, Xiangyu Kong +13D Scene Editing 3D Scene Generation
Department of Computer Science, City University of Hong Kong, Hong Kong · School of Computing and Data Science, University of Hong Kong · School of Computer Science, Beijing Information Science and Technology University
ReSpace: Text-Driven Autoregressive 3D Indoor Scene Synthesis and Editing
Jun 3, 2025Martin JJ. Bucher, Iro Armeni3D Scene Editing 3D Scene Representation
Stanford University
iARCS: Iterative Agentic RL for Controllable 3D Scene Generation
Aug 6, 2026Saugat Adhikari, Ashok Prasad Neupane, Pramish Paudel +2Constrained Generative Modeling Agentic RL
Pulchowk Campus, IOE, Tribhuvan University, Lalitpur, Nepal · INSAIT, Sofia University “St. Kliment Ohridski”, Bulgaria · NAAMII, Kathmandu, Nepal

Jul 15, 2026J/K move · Enter open · S save

ThinkBLOX: 3D Indoor Scene Generation with Progressive Reasoning

Yuan Xiao, Can Wang, Xiangyu Kong, Jing Liao

Department of Computer Science, City University of Hong Kong, Hong Kong · School of Computing and Data Science, University of Hong Kong · School of Computer Science, Beijing Information Science and Technology University

3D Scene Editing 3D Scene Generation 3D Scene Layout Generation Indoor Scene Generation

Open paper AskPDF arXiv

While traditional graphics methods often synthesize 3D indoor scenes autoregressively or hierarchically, recent vision-language model (VLM)-based generators predominantly adopt a one-shot paradigm where the full layout is planned at once. This one-shot approach often requires global re-optimization or complete reconstruction during interactive editing (e.g., inserting or moving objects) and can lead to physically or semantically poorly organized arrangements. To address these challenges, we propose ThinkBLOX, a VLM-based progressive reasoning framework that iteratively designs and refines 3D scenes. ThinkBLOX treats layout generation as a state-conditioned, step-by-step reasoningand-action process. To power this, we construct the ThinkBLOX-Data-200K dataset, containing 224,757 procedural placement pairs annotated with multi-view scene context, explicit Chain-of-Thought (CoT) rationales, and structured JSON layouts. Through supervised fine-tuning (SFT) on this dataset, the VLM learns to bridge the reasoning-action gap under incremental updates. Furthermore, recognizing that scene synthesis is inherently a multisolution task where SFT suffers from reward conflict, we introduce Tier-Decoupled GDPO. This reinforcement learning scheme organizes heterogeneous rewards into distinct tiers, stabilizing policy optimization across physical validity, semantic plausibility, and reasoning-action consistency. Extensive experiments show that ThinkBLOX significantly outperforms recent one-shot and iterative baselines in physical plausibility, semantic alignment, and interactive editability. Additionally, we show that it supports diverse applications, including both global and local generation and rearrangement of 3D scenes.

Abstract

3D Scene Generation

Explore similar work

ThinkBLOX: 3D Indoor Scene Generation with Progressive Reasoning

ReSpace: Text-Driven Autoregressive 3D Indoor Scene Synthesis and Editing

iARCS: Iterative Agentic RL for Controllable 3D Scene Generation