AI Paper Atlas

Research map / Benchmarks / CV-Bench-3D

CV-Bench-3D leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on CV-Bench-3D. Best accuracy: 97.58 (HiSpatial-3B, CVPR 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

CV-Bench-3D · accuracy

10 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
197.58HiSpatial-3B
PaliGemma2-3B
HiSpatial: Taming Hierarchical 3D Spatial Understanding in Vision-Language Models
CVPR 2026 · Table 2
Multimodal LLMs
291.7FactoSR-8B-RL
Qwen3-VL-8B-Instruct
Unfold The World: Factorize 4D Properties in Reinforcing Spatial Understanding
ECCV 2026 · Table 2
Multimodal LLMs
386.4AnchorSim-7B
Qwen2.5-VL-7B
Mental Imagery That Matters: Curing Latent Collapse in Multimodal Reasoning
NeurIPS 2026 · Table 1
Multimodal LLMs
482.34VERDICT
Qwen2.5-VL-7B-Instruct
VERDICT: Training-Free Step-Wise Verification of Multimodal Reasoning via Disagreement-Aware Consensus
ECCV 2026 · Table 1
Multimodal LLMs
581.7Multi-SpatialMLLM
InternVL2-8B
Multi-SpatialMLLM: Multi-Frame Spatial Understanding with Multi-Modal Large Language Models
CVPR 2026 · Appendix Table
Multimodal LLMs
681.1iVGR
Qwen2.5-VL-7B
iVGR: Internalizing Visually Grounded Reasoning for MLLMs with Reinforcement Learning
ICML 2026 · Table 2
Multimodal LLMs
778.8vanilla RFT w/ Entropy-Regularization
Mini-o3 (Qwen2.5-VL-7B)
Diversity Over Frequency: Rethinking Tool Use in Visual Chain-of-Thought Agents
ICML 2026 · Table 3
Multimodal LLMs
870.8+ PGT
Llama-3-8B (instruction tuning, CLIP-ViT-L/14)
PGT: Procedurally Generated Tasks for improving visual grounding in MLLMs
ICML 2026 · Table 1
Multimodal LLMs
969.2R2-T2 (MoAI-7B, NGD)
MoAI-7B
R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts
ICML 2025 · Table: comparison of three R2-
Human and embodied modeling
1063.2GenHancer
LLaVA-1.5 Vicuna-7B
GenHancer: Imperfect Generative Models are Secretly Strong Vision-Centric Enhancers
ICCV 2025 · Table 2
Multimodal LLMs

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.