OmniCoT is a new panoramic reasoning benchmark with 6.7K eval, 1K real, and 14.3K training examples plus a two-stage SFT+GRPO training method to enforce global 360-degree consistency.
arXiv:2509.18905 (2025) 6, 9, 17
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
fields
cs.CV 10years
2026 10roles
background 2representative citing papers
SpatialAct benchmark shows VLMs handle isolated spatial reasoning but fail to maintain coherent spatial beliefs and produce reliable actions in multi-turn 3D interactions, underperforming humans.
Proposes an equation-anchored tool-use method for MLLMs that writes the pinhole back-projection equation in Chain-of-Thought and substitutes retrieved camera intrinsics and depths to achieve robustness in 3D object detection and visual grounding under rescaled intrinsics.
Orientation information is recoverable from MLLM visual encoder embeddings via linear regression, contradicting the hypothesis that failures originate in the encoders.
Backward token warping in ViT-based MLLMs enables reliable reasoning from nearby viewpoints by preserving semantic coherence better than pixel-wise warping or fine-tuning baselines.
SCP defines a new benchmark task for predicting spatial causal outcomes beyond direct observation and shows that 23 leading models lag far behind humans on it.
HSGM structures 3D geometry and semantics into a multi-level map that lets VLMs perform high-level planning in zero-shot VLN, achieving SOTA on R2R-CE and RxR-CE.
SpaceMind++ adds an explicit voxelized allocentric cognitive map and coordinate-guided fusion to video MLLMs, claiming SOTA on VSI-Bench and improved out-of-distribution generalization on three other 3D benchmarks.
Distilling view-consistent future views and action-outcome supervision from a generative world model into a VLM via two-stage post-training improves dynamic spatial reasoning on SAT-Real, VSI-Bench and similar benchmarks while avoiding test-time world-model cost.
Semantic Generative Tuning applies segmentation-based generative proxies during post-training to align and improve both understanding and generation in unified multimodal models.
citing papers explorer
-
Why MLLMs Struggle to Determine Object Orientations
Orientation information is recoverable from MLLM visual encoder embeddings via linear regression, contradicting the hypothesis that failures originate in the encoders.