VSI-Bench
PulseAugur coverage of VSI-Bench — every cluster mentioning VSI-Bench across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
Tsinghua, Tencent, NTU unveil Spatial-TTT for AI spatial memory
Researchers from Tsinghua University, Tencent Hunyuan, and Nanyang Technological University have developed Spatial-TTT, a novel approach to endow AI models with "streaming spatial memory." This method addresses the limi…
-
World Labs unveils Atlas, an omni world model for spatial intelligence · 8 sources tracked
World Labs has introduced Atlas, a new AI model designed for spatial intelligence that can generate, reconstruct, and simulate 3D worlds from various inputs including images, video, text, and depth data. Unlike speciali…
-
New research probes LLM spatial reasoning with 3D scene graphs and 2D benchmarks
Two new research papers explore spatial reasoning capabilities in large language models. The first, "GraFT," introduces a training-free framework that uses 3D scene graphs to enhance multimodal LLMs' geometric understan…
-
OraRL framework boosts video MLLM training efficiency
Researchers have introduced OraRL, a novel reinforcement learning framework designed to enhance the training of video multimodal large language models (MLLMs). This method improves sample efficiency and scalability by t…
-
New frameworks tackle video reasoning challenges in large vision-language models
Researchers have developed new frameworks to address the challenges in video reasoning for large vision-language models (LVLMs). One approach, the Chain of Evidence (CoE), decouples grounding and reasoning to improve ef…
-
New Spa3R framework boosts 3D spatial reasoning in vision-language models
Researchers have developed Spa3R, a novel self-supervised framework designed to enhance 3D spatial reasoning in vision-language models. Unlike existing methods that rely on explicit 3D data or partial geometric priors, …
-
New framework GUIDE enhances MLLMs with progressive geometric integration
Researchers have developed GUIDE (Geometric Unrolling Inside MLLM Early-layers), a novel framework designed to enhance Multimodal Large Language Models (MLLMs) in understanding physical space and 3D scenes. Unlike previ…
-
ConsiSpace framework boosts video spatial reasoning in LLMs
Researchers have introduced ConsiSpace, a novel framework designed to enhance video spatial reasoning capabilities in multimodal large language models (MLLMs). This framework addresses the current semantic-centric limit…
-
RynnBrain 1.1 embodied models outperform existing systems on benchmarks
Researchers have introduced RynnBrain 1.1, a new family of embodied foundation models available in 2B, 9B, and 122B-A10B scales. This model family is designed for robots, supporting perception, spatial reasoning, locali…
-
New framework enhances LLMs' 3D spatial reasoning from sparse inputs
Researchers have developed SpaR3D-MoE, a novel framework designed to enhance the 3D spatial reasoning capabilities of Multimodal Large Language Models (MLLMs) using only sparse RGB inputs. The system employs an adaptive…
-
New framework DR-MV3D enhances 3D visual question answering with dense rewards
Researchers have introduced DR-MV3D, a novel framework designed to enhance multi-view 3D visual question answering (MV3D-VQA). This approach utilizes dense, verifiable rewards to supervise the reasoning process, moving …
-
OneCanvas simplifies 3D scene understanding for VLMs with panoramic reprojection
Researchers have developed OneCanvas, a novel approach to 3D scene understanding for vision-language models (VLMs). Instead of complex geometry encoders or extensive training, OneCanvas projects patch features onto a si…
-
New framework AlloSpatial boosts foundation model spatial reasoning
Researchers have introduced AlloSpatial, a new framework designed to enhance the spatial reasoning capabilities of foundation models. This framework converts egocentric observations into structured allocentric represent…
-
New AlloSpatial Framework Boosts AI Spatial Reasoning
Researchers have developed AlloSpatial, a new framework designed to improve the spatial reasoning capabilities of foundation models. This framework addresses the limitation of current models by converting egocentric obs…
-
Cambrian-P video model uses camera pose for improved spatial reasoning
Researchers have introduced Cambrian-P, a novel video multimodal large language model (MLLM) that incorporates camera pose information. This approach treats video frames not as isolated images but as part of a continuou…
-
GeoThinker framework actively integrates geometry for advanced spatial reasoning
Researchers have developed GeoThinker, a novel framework that enhances spatial reasoning in multimodal large language models (MLLMs) by actively integrating geometric information. Unlike previous passive fusion methods,…
-
VLMs tackle visual illusions, spatial reasoning, and evaluation benchmarks
Researchers are developing new methods to improve the robustness and reasoning capabilities of Vision-Language Models (VLMs). One approach, Structured Qualitative Inference (SQI), aims to mitigate visual illusions by en…
-
New frameworks enhance VLM spatial reasoning with world models and multi-agent systems
Researchers have developed World2VLM, a novel training framework that distills spatial reasoning capabilities from generative world models into vision-language models (VLMs). This approach synthesizes future views to pr…