5 papers
Zero-Mem: Zero-Token Memory Operations for LLM Agents
Yilin Xiao, Zhehan Zhu, Yujing Zhang +8
LLM agents need memory to act consistently over long interactions, yet many systems use additional LLM calls to operate that memory. Generating intermediate records and mediating t…
TOUR: A Trajectory-Level Unlearning Benchmark for Offline Reinforcement Learning
Chaofan Pan, Lingfei Ren, Xiangyu Jiang +6
Offline Reinforcement Learning (RL) agents are trained on fixed behavioral trajectories, which makes trajectory-level deletion important when selected data must be removed after tr…
Regime-Adaptive Continual Learning for Portfolio Management
Chaofan Pan, Lingfei Ren, Linbo Xiong +3
Financial markets are inherently non-stationary, exhibiting frequent regime shifts and structural changes that render traditional Portfolio Management (PM) approaches ineffective.…
LogicPoison: Logical Attacks on Graph Retrieval-Augmented Generation
Yilin Xiao, Jin Chen, Qinggang Zhang +6
Graph-based Retrieval-Augmented Generation (GraphRAG) enhances the reasoning capabilities of Large Language Models (LLMs) by grounding their responses in structured knowledge graph…
Multi-granularity Knowledge Transfer for Continual Reinforcement Learning
Chaofan Pan, Lingfei Ren, Yihui Feng +4
Continual reinforcement learning (CRL) empowers RL agents with the ability to learn a sequence of tasks, accumulating knowledge learned in the past and using the knowledge for prob…