Topic: Awesome List: ai-agent-papers-2026

← Back to home

1. Less is More for RAG: Information Gain Pruning for Generator-Aligned Reranking and Evidence Selection

   Paper • Jan 24, 2026 • arxiv.org • Zhipeng Song, Yizhi Zhou, Xiangyu Kong, Jiulong Jiao, Xinrui Bao, Xu You, Xueqing Shi, Yuhang Zhou, Heng Qi

   Retrieval-augmented generation (RAG) grounds large language models with external evidence, but under a limited context budget, the key challenge is deciding which retrieved passages should be injec...

2. FastInsight: Fast and Insightful Retrieval via Fusion Operators for Graph RAG

   Paper • Jan 26, 2026 • arxiv.org • Seonho An, Chaejeong Hyun, Min-Soo Kim

   Existing Graph RAG methods aiming for insightful retrieval on corpus graphs typically rely on time-intensive processes that interleave Large Language Model (LLM) reasoning. To enable time-efficient...

3. FadeMem: Biologically-Inspired Forgetting for Efficient Agent Memory

   Paper • Feb 6, 2026 • arxiv.org • Lei Wei, Xiao Peng, Xu Dong, Niantao Xie, Bin Wang

   Large language models deployed as autonomous agents face critical memory limitations, lacking selective forgetting mechanisms that lead to either catastrophic forgetting at context boundaries or in...

4. Dep-Search: Learning Dependency-Aware Reasoning Traces with Persistent Memory

   Paper • Jan 26, 2026 • arxiv.org • Yanming Liu, Xinyue Peng, Zixuan Yan, Yanxin Shen, Wenjie Xu, Yuefeng Huang, Xinyi Wang, Jiannan Cao, Jianwei Yin, Xuhong Zhang

   Large Language Models (LLMs) have demonstrated remarkable capabilities in complex reasoning tasks, particularly when augmented with search mechanisms that enable systematic exploration of external ...

5. When Iterative RAG Beats Ideal Evidence: A Diagnostic Study in Scientific Multi-hop Question Answering

   Paper • Feb 5, 2026 • arxiv.org • Mahdi Astaraki, Mohammad Arshi Saloot, Ali Shiraee Kasmaee, Hamidreza Mahyar, Soheila Samiee

   Retrieval-Augmented Generation (RAG) extends large language models (LLMs) beyond parametric knowledge, yet it is unclear when iterative retrieval-reasoning loops meaningfully outperform static RAG,...

6. AMA: Adaptive Memory via Multi-Agent Collaboration

   Paper • Feb 2, 2026 • arxiv.org • Weiquan Huang, Zixuan Wang, Hehai Lin, Sudong Wang, Bo Xu, Qian Li, Beier Zhu, Linyi Yang, Chengwei Qin

   The rapid evolution of Large Language Model (LLM) agents has necessitated robust memory systems to support cohesive long-term interaction and complex reasoning. Benefiting from the strong capabilit...

7. MemCtrl: Using MLLMs as Active Memory Controllers on Embodied Agents

   Paper • Jan 28, 2026 • arxiv.org • Vishnu Sashank Dorbala, Dinesh Manocha

   Foundation models rely on in-context learning for personalized decision making. The limited size of this context window necessitates memory compression and retrieval systems like RAG. These systems...

8. A2RAG: Adaptive Agentic Graph Retrieval for Cost-Aware and Reliable Reasoning

   Paper • Jan 29, 2026 • arxiv.org • Jiate Liu, Zebin Chen, Shaobo Qiao, Mingchen Ju, Danting Zhang, Bocheng Han, Shuyue Yu, Xin Shu, Jingling Wu, Dong Wen, Xin Cao, Guanfeng Liu, Zhengyi Yang

   Graph Retrieval-Augmented Generation (Graph-RAG) enhances multihop question answering by organizing corpora into knowledge graphs and routing evidence through relational structure. However, practic...

9. When should I search more: Adaptive Complex Query Optimization with Reinforcement Learning

   Paper • Jan 29, 2026 • arxiv.org • Wei Wen, Sihang Deng, Tianjun Wei, Keyu Chen, Ruizhi Qiao, Xing Sun

   Query optimization is a crucial component for the efficacy of Retrieval-Augmented Generation (RAG) systems. While reinforcement learning (RL)-based agentic and reasoning methods have recently emerg...

10. ShardMemo: Masked MoE Routing for Sharded Agentic LLM Memory

    Paper • Jan 29, 2026 • arxiv.org • Yang Zhao, Chengxiao Dai, Yue Xiu, Mengying Kou, Yuliang Zheng, Dusit Niyato

    Agentic large language model (LLM) systems rely on external memory for long-horizon state and concurrent multi-agent execution, but centralized indexes and heuristic partitions become bottlenecks a...

11. E-mem: Multi-agent based Episodic Context Reconstruction for LLM Agent Memory

    Paper • Jan 29, 2026 • arxiv.org • Kaixiang Wang, Yidan Lin, Jiong Lou, Zhaojiacheng Zhou, Bunyod Suvonov, Jie Li

    The evolution of Large Language Model (LLM) agents towards System~2 reasoning, characterized by deliberative, high-precision problem-solving, requires maintaining rigorous logical integrity over ex...

12. ProRAG: Process-Supervised Reinforcement Learning for Retrieval-Augmented Generation

    Paper • Jan 29, 2026 • arxiv.org • Zhao Wang, Ziliang Zhao, Zhicheng Dou

    Reinforcement learning (RL) has become a promising paradigm for optimizing Retrieval-Augmented Generation (RAG) in complex reasoning tasks. However, traditional outcome-based RL approaches often su...

13. JADE: Bridging the Strategic-Operational Gap in Dynamic Agentic RAG

    Paper • Jan 29, 2026 • arxiv.org • Yiqun Chen, Erhan Zhang, Tianyi Hu, Shijie Wang, Zixuan Yang, Meizhi Zhong, Xiaochi Wei, Yan Gao, Yi Wu, Yao Hu, Jiaxin Mao

    The evolution of Retrieval-Augmented Generation (RAG) has shifted from static retrieval pipelines to dynamic, agentic workflows where a central planner orchestrates multi-turn reasoning. However, e...

14. DIVERGE: Diversity-Enhanced RAG for Open-Ended Information Seeking

    Paper • Jan 30, 2026 • arxiv.org • Tianyi Hu, Niket Tandon, Akhil Arora

    Existing retrieval-augmented generation (RAG) systems are primarily designed under the assumption that each query has a single correct answer. This overlooks common information-seeking scenarios wi...

15. Aggregation Queries over Unstructured Text: Benchmark and Agentic Method

    Paper • Feb 3, 2026 • arxiv.org • Haojia Zhu, Qinyuan Xu, Haoyu Li, Yuxi Liu, Hanchen Qiu, Jiaoyan Chen, Jiahui Jin

    Aggregation query over free text is a long-standing yet underexplored problem. Unlike ordinary question answering, aggregate queries require exhaustive evidence collection and systems are required ...

16. ProcMEM: Learning Reusable Procedural Memory from Experience via Non-Parametric PPO for LLM Agents

    Paper • Feb 2, 2026 • arxiv.org • Qirui Mi, Zhijian Ma, Mengyue Yang, Haoxuan Li, Yisen Wang, Haifeng Zhang, Jun Wang

    LLM-driven agents demonstrate strong performance in sequential decision-making but often rely on on-the-fly reasoning, re-deriving solutions even in recurring scenarios. This insufficient experienc...

17. SOPRAG: Multi-view Graph Experts Retrieval for Industrial Standard Operating Procedures

    Paper • Feb 2, 2026 • arxiv.org • Liangtao Lin, Zhaomeng Zhu, Tianwei Zhang, Yonggang Wen

    Standard Operating Procedures (SOPs) are essential for ensuring operational safety and consistency in industrial environments. However, retrieving and following these procedures presents unique cha...

18. AI Agent Systems for Supply Chains: Structured Decision Prompts and Memory Retrieval

    Paper • Feb 5, 2026 • arxiv.org • Konosuke Yoshizato, Kazuma Shimizu, Ryota Higa, Takanobu Otsuka

    This study investigates large language model (LLM) -based multi-agent systems (MASs) as a promising approach to inventory management, which is a key component of supply chain management. Although t...

19. Graph-based Agent Memory: Taxonomy, Techniques, and Applications

    Paper • Feb 5, 2026 • arxiv.org • Chang Yang, Chuang Zhou, Yilin Xiao, Su Dong, Luyao Zhuang, Yujing Zhang, Zhu Wang, Zijin Hong, Zheng Yuan, Zhishang Xiang, Shengyuan Chen, Huachi Zhou, Qinggang Zhang, Ninghao Liu, Jinsong Su, Xinrun Wang, Yi Chang, Xiao Huang

    Memory emerges as the core module in the Large Language Model (LLM)-based agents for long-horizon complex tasks (e.g., multi-turn dialogue, game playing, scientific discovery), where memory can ena...

20. Mitigating Hallucination in Financial Retrieval-Augmented Generation via Fine-Grained Knowledge Verification

    Paper • Feb 5, 2026 • arxiv.org • Taoye Yin, Haoyuan Hu, Yaxin Fan, Xinhao Chen, Xinya Wu, Kai Deng, Kezun Zhang, Feng Wang

    In financial Retrieval-Augmented Generation (RAG) systems, models frequently rely on retrieved documents to generate accurate responses due to the time-sensitive nature of the financial domain. Whi...