8 papers
MemPrism: Task-Conditioned Relational Memory Views for Long-Horizon Agents
Zhisheng Chen, Bingfan Zeng, Bangde Cao +8
Long-horizon agents performing multi-step Web tasks must reuse earlier page observations and interactions to support subsequent actions. However, preserving relevant history does n…
ProtoAda: Prototype-Guided Adaptive Adapter Expansion and Geometric Consolidation for Multimodal Continual Instruction Tuning
Yu-Cheng Shi, Zhen-Hao Xie, Jun-Tao Tang +1
Multimodal Large Language Models (MLLMs) achieve strong performance through instruction tuning, but real-world deployment requires them to continually acquire new vision-language c…
CRAM: Centroid-Routing and Adaptive MoE for Multimodal Continual Instruction Tuning
Jun-Tao Tang, Zhen-Hao Xie, Yu-Cheng Shi +1
Multimodal Large Language Models (MLLMs) unify heterogeneous vision-language tasks under a shared generative framework via instruction tuning, yet real-world deployment demands con…
MineEvolve: Self-Evolution with Accumulated Knowledge for Long-Horizon Embodied Minecraft Agents
Zhengwei Xie, Zhisheng Chen, Ziyan Weng +7
Long-horizon embodied intelligence requires agents to improve through interaction, not merely to execute plans generated from static goals. A central challenge is therefore to tran…
SAME: Stabilized Mixture-of-Experts for Multimodal Continual Instruction Tuning
Zhen-Hao Xie, Jun-Tao Tang, Yu-Cheng Shi +3
Multimodal Large Language Models (MLLMs) achieve strong performance through instruction tuning, but real-world deployment requires them to continually expand their capabilities, ma…
PolarMem: A Training-Free Polarized Latent Graph Memory for Verifiable Vision-Language Models
Zhisheng Chen, Tingyu Wu, Zijie Zhou +7
Memory is not merely a storage mechanism for intelligent systems, but a structure for organizing evidence and constraining belief. This is especially important for multimodal reaso…
Hierarchical Semantic Tree Anchoring for CLIP-Based Class-Incremental Learning
Tao Hu, Lan Li, Zhen-Hao Xie +1
Class-Incremental Learning (CIL) enables models to learn new classes continually while preserving past knowledge. Recently, vision-language models like CLIP offer transferable feat…
AirCache: Activating Inter-modal Relevancy KV Cache Compression for Efficient Large Vision-Language Model Inference
Kai Huang, Hao Zou, Bochen Wang +3
Recent advancements in Large Visual Language Models (LVLMs) have gained significant attention due to their remarkable reasoning capabilities and proficiency in generalization. Howe…