paper-with-me

Papers

Learning to Retrieve: Dual-Level Long-Term Memory for Text-to-SQL Agents

2026-05-30 · Yibo Wang, Nikki Lijing Kuang, Philip S. Yu, Zhewei Yao, Yuxiong He arxiv

Interactive text-to-SQL agents solve database tasks through multi-turn interactions involving schema exploration, query execution, feedback interpretation, and decision revision. Long-term memory helps agents reuse past experiences, but existing retrieval methods remain limited. Static methods rely on fixed similarity heuristics that do not optimize downstream utility, while dynamic methods often learn from sparse final outcomes and retrieve memories at a single decision horizon. This is insufficient when memory usefulness changes across interaction stages, since memories useful for initial planning may differ from those needed for local, state-conditioned execution. We propose MERIT, a dynamic multi-horizon memory retrieval framework. MERIT maintains episode-level memory for global strategic guidance and turn-level memory for local decision support. Both levels use learned retrieval policies optimized with reinforcement learning. To train turn-level retrieval despite limited intermediate supervision, MERIT uses a lightweight Process Reward Model to provide dense proxy rewards for local memory selection. Experiments on BIRD-Interact show that MERIT outperforms no-memory, static-retrieval, and dynamic-retrieval baselines in success rate while reducing average interaction turns. Transfer results on Spider2-Snow further show positive cross-benchmark transfer without benchmark-specific tuning. These results suggest that multi-horizon retrieval improves experience reuse in interactive text-to-SQL agents.

📄 PDF Abstract BibTeX arXiv:2606.00547

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Mnemis: Dual-Route Retrieval on Hierarchical Graphs for Long-Term LLM Memory

2026-02-17 · Zihao Tang, Xin Yu, Ziyu Xiao, Zengxuan Wen 외 arxiv

AI Memory, specifically how models organizes and retrieves historical messages, becomes increasingly valuable to Large Language Models (LLMs), yet existing methods (RAG and Graph-RAG) primarily retrieve memory through si…

SGMem: Sentence Graph Memory for Long-Term Conversational Agents

2025-09-25 · Yaxiong Wu, Yongyue Zhang, Sheng Liang, Yong Liu arxiv

Long-term conversational agents require effective memory management to handle dialogue histories that exceed the context window of large language models (LLMs). Existing methods based on fact extraction or summarization …

Conversational Question AnsweringResponse Generation

ReMem-VLA: Empowering Vision-Language-Action Model with Memory via Dual-Level Recurrent Queries

2026-03-13 · Hang Li, Fengyi Shen, Dong Chen, Liudi Yang 외 arxiv

Vision-language-action (VLA) models for closed-loop robot control are typically cast under the Markov assumption, making them prone to errors on tasks requiring historical context. To incorporate memory, existing VLAs ei…

Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation

2026-07-08 · Hongyu Qu, Jianzhe Gao, Xiaobin Hu, Shaohuan Yang 외 arxiv

Mainstream Vision-Language-Action (VLA) models predict actions primarily from the current observation under a Markovian assumption, thus struggling with long-horizon, temporally dependent tasks. Existing memory-augmented…

Multimodal Reasoning

RRM: Experience-Driven Reflective Retrieval Memory for Long-Horizon Multimodal Reasoning

2026-07-30 · Jingxiang Fan, Junbao Zhuo, Bochao Zou arxiv

Existing multimodal long-term memory agents use external memory to overcome the limited context available for long videos. However, most methods emphasize what to store rather than how stored memory should be retrieved. …

Multimodal ReasoningAnswer Generation