VoxAfford fuses multi-scale voxel features into MLLM output tokens using cross-attention with a learned compatibility gate to achieve SOTA open-vocabulary 3D affordance detection with ~8% mIoU gain and zero-shot robot transfer.
Lisa: Reasoning segmentation via large language model
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
A channel attention-guided cross-modal knowledge distillation method transfers high-order vision-language and channel-wise semantic correlations from teacher to student networks, yielding significant performance gains for the student on referring image segmentation without inference-time parameter b
citing papers explorer
-
VoxAfford: Multi-Scale Voxel-Token Fusion for Open-Vocabulary 3D Affordance Detection
VoxAfford fuses multi-scale voxel features into MLLM output tokens using cross-attention with a learned compatibility gate to achieve SOTA open-vocabulary 3D affordance detection with ~8% mIoU gain and zero-shot robot transfer.
-
Channel Attention-Guided Cross-Modal Knowledge Distillation for Referring Image Segmentation
A channel attention-guided cross-modal knowledge distillation method transfers high-order vision-language and channel-wise semantic correlations from teacher to student networks, yielding significant performance gains for the student on referring image segmentation without inference-time parameter b