11 papers
MemDuet: Learning Dual-Process Memory for Long-Context Reasoning
Jiangcheng Song, Haodong Jing, Yulin Mu +7
Reasoning over infinitely long documents is becoming a reality with recurrent memory agents. However, increasing context length exacerbates the trade-off between reasoning performa…
StrataVLA: Hierarchical and Efficient 3D Geometric Grounding for Vision-Language-Action Models
Jin Cui, Zhaoyu Pu, Botao Cai +4
Vision-Language-Action (VLA) models inherit strong semantic priors from large-scale vision-language pretraining, yet remain limited in robotic manipulation by insufficient 3D spati…
GLOBE: Trajectory-Aligned Gradient Matching with Structured SparseOptimization for Coreset Selection
Hetian Liu, Jin Cui, Mengcheng Shi +4
On-device training of deep neural networks is fundamentally constrained by the computational and memory costs of large-scale datasets. Coreset selection offers a practical solution…
Look Where It Matters: Adaptive Visual Refinement for Vision-Language-Action Models
Jin Cui, Yanbin Hu, Xinyue Long +3
Visual representations of VLA models remain unreliable for spatially precise robotic manipulation. We uncover that vision encoders in VLAs also exhibit attention artifacts previous…
HAFI-VLM: A Frequency Perspective for Diagnosing and Enhancing Visual Perception in Vision-Language Models
Jin Cui, Chuanchang Su, Jiayi Lu +3
Vision-language models (VLMs) remain unreliable when predictions require fine-grained visual evidence. We identify a previously overlooked cause: spectral response rigidity. Despit…
Distill What RGB Can Recover: Privileged 3D Evidence for RGB-Only Vision-Language Models
Yanbin Hu, Jin Cui, Jun Ye +4
3D scene understanding requires reasoning about entity existence, spatial layout, and object relations, yet RGB images alone often provide insufficient 3D cues. Existing 3D-VLMs co…
ECHO: Continuous Hierarchical Memory for Vision-Language-Action Models
Yanbin Hu, Jin Cui, Jiayi Lu +6
Memory capacity is a critical factor determining the performance of Vision-Language-Action (VLA) models in long-horizon manipulation tasks. Existing memory-augmented architectures…
Retrieve, Integrate, and Synthesize: Spatial-Semantic Grounded Latent Visual Reasoning
Jin Cui, Xinyue Long, Xunyong Zhang +5
Multimodal Large Language Models (MLLMs) have made remarkable progress on vision-language reasoning, yet most methods still compress visual evidence into discrete textual thoughts,…
MIND: From Passive Mimicry to Active Reasoning through Capability-Aware Multi-Perspective CoT Distillation
Jin Cui, Jiaqi Guo, Jiepeng Zhou +6
While Large Language Models (LLMs) have emerged with remarkable capabilities in complex tasks through Chain-of-Thought reasoning, practical resource constraints have sparked intere…
"The Whole Is Greater Than the Sum of Its Parts": A Compatibility-Aware Multi-Teacher CoT Distillation Framework
Jin Cui, Jiaqi Guo, Ruixuan Yang +6
Chain-of-Thought (CoT) reasoning empowers Large Language Models (LLMs) with remarkable capabilities but typically requires prohibitive parameter scales. CoT distillation has emerge…
FAST: Topology-Aware Frequency-Domain Distribution Matching for Coreset Selection
Jin Cui, Boran Zhao, Jiajun Xu +3
Coreset selection compresses large datasets into compact, representative subsets, reducing the energy and computational burden of training deep neural networks. Existing methods ar…