activity
20242026
collaborators

7 papers

cs.CL2026

Metis: Memory Foundation Model

Zeyu Zhang, Ziliang Guo, Yihang Sun +14

Recent advances in AI agents have increasingly internalized native capabilities into their underlying foundation models, giving rise to multimodal foundation models and large reaso…

cs.AI2026

DELTAMEM: Incremental Experience Memory for LLM Agents via Residual Trees

Haoran Tan, Zeyu Zhang, Zhicheng Cao +2

Large Language Model (LLM)-based agents increasingly rely on memory to learn from experiences over continual interactions. However, storing experiences as independent, flat units l…

cs.AI2026

From Coarse to Fine: Self-Adaptive Hierarchical Planning for LLM Agents

Haoran Tan, Zeyu Zhang, Chen Ma +3

Large language model-based agents have recently emerged as powerful approaches for solving dynamic and multi-step tasks. Most existing agents employ planning mechanisms to guide lo…

cs.CV2026

Explore Like Humans: Autonomous Exploration with Online SG-Memo Construction for Embodied Agents

Xu Chen, Shichao Xie, Zhining Gu +7

Constructing structured spatial memory is essential for enabling long-horizon reasoning in complex embodied navigation tasks. Current memory construction predominantly relies on a…

cs.AI2026

NextMem: Towards Latent Factual Memory for LLM-based Agents

Zeyu Zhang, Rui Li, Xiaoyan Zhao +4

Memory is critical for LLM-based agents to preserve past observations for future decision-making, where factual memory serves as its foundational part. However, existing approaches…

cs.AI2025

Explicit v.s. Implicit Memory: Exploring Multi-hop Complex Reasoning Over Personalized Information

Zeyu Zhang, Yang Zhang, Haoran Tan +2

In large language model-based agents, memory serves as a critical capability for achieving personalization by storing and utilizing users' information. Although some previous studi…

cs.LG2024

Generalizing Knowledge Graph Embedding with Universal Orthogonal Parameterization

Rui Li, Chaozhuo Li, Yanming Shen +2

Recent advances in knowledge graph embedding (KGE) rely on Euclidean/hyperbolic orthogonal relation transformations to model intrinsic logical patterns and topological structures.…