SpaR3D-MoE: Breaking New Ground in 3D Spatial Reasoning
SpaR3D-MoE introduces a novel approach to spatial reasoning in MLLMs by utilizing sparse RGB inputs and a geometry-aware graph. It outperforms existing models significantly.
Read the full articleYou might also wanna read
Group3D: MLLM-Driven Semantic Grouping for Open-Vocabulary 3D Object Detection (ECCV 2026) - GitHub Repository
[ECCV 2026] Group3D: MLLM-Driven Semantic Grouping for Open-Vocabulary 3D Object Detection - Ubin108/Group3D
github.com·7d ago
Nemotron-Labs-3-Puzzle-75B-A9B: A Compressed Hybrid MoE LLM for Efficient Interactive Deployment
We present Nemotron-Labs-3-Puzzle-75B-A9B, a compressed variant of Nemotron-3-Super optimized for interactive deployment. We designed the mo
arxiv.org·2d agoNemotron-Labs-3-Puzzle-75B-A9B: A Compressed Hybrid MoE LLM for Efficient Interactive Deployment
We present Nemotron-Labs-3-Puzzle-75B-A9B, a compressed variant of Nemotron-3-Super optimized for interactive deployment. We designed the mo
arxiv.org·2d ago
BilliardPhys-Bench: New Benchmark Reveals Physical Reasoning Gaps in Multimodal AI Models
Current multimodal models handle static image recognition well, but intuitive physical reasoning remains a weakness. Predicting how objects
arxiv.org·1mo agoELSA3D: Structured Text-3D Reasoning
ELSA3D revolutionizes unified 3D models with elastic semantic anchoring, achieving SOTA performance in 3D generation and captioning while ha
Geometric and Stochastic Analysis of Discontinuities in Sparse Mixture-of-Experts Architectures
Sparse Mixture-of-Experts (SMoE) architectures are now widely deployed in state-of-the-art language and vision models, where conditional rou
arxiv.org·20d ago
Building high-performance expert-parallel dispatch and combine kernels for MoE LLM inference
How expert-parallel dispatch and combine kernels work, built up from scratch: the high-throughput shape and the low-latency one.
fergusfinn.com·1mo ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.