Efficient Agents

A new survey paper examines efficiency in large language model agentic systems across three core components — memory, tool learning, and planning — measuring costs including latency, tokens, and steps. The paper reports that recent approaches frequently converge on shared high-level principles such as bounded context via compression and retrieval, reduced action cost via budgeted tool use and caching, and controlled search via hierarchical planning and pruning. It characterizes efficiency two ways: comparing effectiveness under a fixed cost budget and comparing cost at a comparable level of effectiveness, framed as a Pareto frontier, and consolidates efficiency metrics and benchmarks for the components.

Toward Efficient Agents: A Survey of Memory, Tool learning, and Planning Abstract Recent years have witnessed increasing interest in extending large language models into agentic systems. While the effectiveness of agents has continued to improve, efficiency , which is crucial for real-world deployment, has often been overlooked. This paper therefore investigates efficiency from three core components of agents: memory, tool learning, and planning , considering costs such as latency, tokens, steps, etc. Aimed to conducting comprehensive research addressing the efficiency of the agentic system itself, we review a broad range of recent approaches that differ in implementation yet frequently converge on shared high-level principles including but not limited to bounded context via compression and retrieval, reduced action cost via budgeted tool use and caching, and controlled search via hierarchical planning and pruning for improving efficiency, which we discuss in detail. Accordingly, we characterize efficiency in two complementary ways: comparing effectiveness under a fixed cost budget, and comparing cost at a comparable level of effectiveness. This trade-off can also be viewed through the Pareto frontier between effectiveness and cost. From this perspective, we also examine efficiency oriented benchmarks by summarizing evaluation protocols for these components and consolidating commonly reported efficiency metrics from both benchmark and methodological studies. Moreover, we discuss the key challenges and future directions, with the goal of providing promising insights. Paper List Navigation 📑 Table of Memory Contents In the paper, we organize memory into construction, management, and access. Since many papers overlap across these stages, this part is primarily organized around memory construction to avoid redundancy. 🔄 Working Memory 📝 Textual Memory - 2025-10 AgentFold: Long-Horizon Web Agents with Proactive Context Management https://arxiv.org/abs/2510.24699 - 2025-07 MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent https://arxiv.org/abs/2507.02259 - 2025-06 MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents https://arxiv.org/abs/2506.15841 NeurIPS WS 2025 COLM WS 2025- 2025-04 Dynamic Cheatsheet: Test-Time Learning with Adaptive Memory https://arxiv.org/abs/2504.07952 - 2024-02 Compress to Impress: Unleashing the Potential of Compressive Memory in Real-World Long-Term Conversations https://arxiv.org/abs/2402.11975 COLING 2025 🧩 Latent Memory - 2026-01 FlashMem: Distilling Intrinsic Latent Memory via Computation Reuse https://arxiv.org/abs/2601.05505 - 2025-09 MemGen: Weaving Generative Latent Memory for Self-Evolving Agents https://arxiv.org/abs/2509.24704 - 2025-02 M+: Extending MemoryLLM with Scalable Long-Term Memory https://arxiv.org/abs/2502.00592 ICML 2025- 2025-01 Titans: Learning to Memorize at Test Time https://arxiv.org/abs/2501.00663 - 2024-09 MemoRAG: Boosting Long Context Processing with Global Memory-Enhanced Retrieval Augmentation https://arxiv.org/abs/2409.05591 TheWebConf 2025- 2024-07 Memory³: Language Modeling with Explicit Memory https://arxiv.org/abs/2407.01178 - 2024-02 MEMORYLLM: Towards Self-Updatable Large Language Models https://arxiv.org/abs/2402.04624 ICML 2024- 2024-01 Long Context Compression with Activation Beacon https://arxiv.org/abs/2401.03462 ICLR 2025 💾 External Memory 📦 Item-based Memory - 2026-03 Evoking User Memory: Personalizing LLM via Recollection-Familiarity Adaptive Retrieval https://arxiv.org/abs/2603.09250 ICLR 2026- 2026-01 MemRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic Memory https://arxiv.org/abs/2601.03192 - 2025-10 Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models https://arxiv.org/abs/2510.04618 - 2025-09 ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory https://arxiv.org/abs/2509.25140 - 2025-08 Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning https://arxiv.org/abs/2508.19828 - 2025-08 Memento: Fine-tuning LLM Agents without Fine-tuning LLMs https://arxiv.org/abs/2508.16153 - 2025-07 Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solving https://arxiv.org/abs/2507.06229 ICML 2025 Workshop- 2025-06 Cost-Efficient Serving of LLM Agents via Test-Time Plan Caching Agentic Plan Caching: Test-Time Memory for Fast and Cost-Efficient LLM Agents https://arxiv.org/abs/2506.14852 NeurIPS 2025 - 2025-05 From Single to Multi-Granularity: Toward Long-Term Memory Association and Selection of Conversational Agents https://arxiv.org/abs/2505.19549 ICLR 2026- 2025-04 Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory https://arxiv.org/abs/2504.19413 - 2025-03 MemInsight: Autonomous Memory Augmentation for LLM Agents https://arxiv.org/abs/2503.21760 EMNLP 2025- 2025-03 In Prospect and Retrospect: Reflective Memory Management for Long-term Personalized Dialogue Agents https://arxiv.org/abs/2503.08026 ACL 2025 - 2025-02 A-MEM: Agentic Memory for LLM Agents https://arxiv.org/abs/2502.12110 NeurIPS 2025- 2025-02 On Memory Construction and Retrieval for Personalized Conversational Agents https://arxiv.org/abs/2502.05589 ICLR 2025- 2024-06 Hello Again LLM-powered Personalized Agent for Long-term Dialogue https://arxiv.org/abs/2406.05925 NAACL 2025- 2024-04 "My agent understands me better": Integrating Dynamic Human-like Memory Recall and Consolidation in LLM-Based Agents https://arxiv.org/abs/2404.00573 CHI EA 2024 - 2023-10 RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentation https://arxiv.org/abs/2310.04408 ICLR 2024- 2023-08 ExpeL: LLM Agents Are Experiential Learners https://arxiv.org/abs/2308.10144 AAAI 2024- 2023-08 MemoChat: Tuning LLMs to Use Memos for Consistent Long-Range Open-Domain Conversation https://arxiv.org/abs/2308.08239 - 2023-05 MemoryBank: Enhancing Large Language Models with Long-Term Memory https://arxiv.org/abs/2305.10250 AAAI 2024 🕸️ Graph-based Memory - 2026-01 MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents https://arxiv.org/abs/2601.03236 - 2025-10 D-SMART: Enhancing LLM Dialogue Consistency via Dynamic Structured Memory And Reasoning Tree https://arxiv.org/abs/2510.13363 - 2025-04 Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory https://arxiv.org/abs/2504.19413 - 2025-01 Zep: A Temporal Knowledge Graph Architecture for Agent Memory https://arxiv.org/abs/2501.13956 - 2024-07 AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents https://arxiv.org/abs/2407.04363 IJCAI 2025- 2024-06 GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models https://arxiv.org/abs/2406.14550 EMNLP 2024 Findings - 2024-02 KG-Agent: An Efficient Autonomous Agent Framework for Complex Reasoning over Knowledge Graph https://arxiv.org/abs/2402.11163 ACL 2025 🪜 Hierarchical Memory - 2025-10 Beyond a Million Tokens: Benchmarking and Enhancing Long-Term Memory in LLMs https://arxiv.org/abs/2510.27246 ICLR 2026- 2025-10 LightMem: Lightweight and Efficient Memory-Augmented Generation https://www.arxiv.org/abs/2510.18866 - 2025-07 Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agents https://arxiv.org/abs/2507.22925 - 2025-07 MemOS: A Memory OS for AI System https://arxiv.org/abs/2507.03724 - 2025-06 Memory OS of AI Agent https://arxiv.org/abs/2506.06326 EMNLP 2025- 2024-08 HiAgent: Hierarchical Working Memory Management for Solving Long-Horizon Agent Tasks with Large Language Model https://arxiv.org/abs/2408.09559 ACL 2025- 2024-02 A Human-Inspired Reading Agent with Gist Memory of Very Long Contexts https://arxiv.org/abs/2402.09727 ICML 2024- 2023-10 MemGPT: Towards LLMs as Operating Systems https://arxiv.org/abs/2310.08560 👥 Multi-Agent Memory 🏠 Local Memory - 2025-08 Intrinsic Memory Agents: Heterogeneous Multi-Agent LLM Systems through Structured Contextual Memory https://arxiv.org/abs/2508.08997 - 2025-04 AgentNet: Decentralized Evolutionary Coordination for LLM-based Multi-Agent Systems https://arxiv.org/abs/2504.00587 NeurIPS 2025- 2025-02 LLM-Powered Decentralized Generative Agents with Adaptive Hierarchical Knowledge Graph for Cooperative Planning https://arxiv.org/abs/2502.05453 🔀 Mixed Memory - 2025-10 LEGOMem: Modular Procedural Memory for Multi-agent LLM Systems for Workflow Automation https://arxiv.org/abs/2510.04851 AAMAS 2026 Extended Abstract - 2025-05 Collaborative Memory: Multi-User Memory Sharing in LLM Agents with Dynamic Access Control https://arxiv.org/abs/2505.18279 - 2025-01 SRMT: Shared Memory for Multi-agent Lifelong Pathfinding https://arxiv.org/abs/2501.13200 📑 Table of Tool Learning Contents Tool Learning framework encompasses tool selection, tool calling, and tool-integrated reasoning for enhanced agent capabilities. 🔧 Tool Selection 🔍 External Retriever - 2025-10 ToolScope: Enhancing LLM Agent Tool Use through Tool Merging and Context-Aware Filtering https://arxiv.org/abs/2510.20036 - 2024-10 Toolshed: Scale Tool-Equipped Agents with Advanced RAG-Tool Fusion and Tool Knowledge Bases https://arxiv.org/abs/2410.14594 ICAART 2025- 2024-10 From Exploration to Mastery: Enabling LLMs to Master Tools via Self-Driven Interactions https://arxiv.org/abs/2410.08197 ICLR 2025 oral- 2024-02 AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls https://arxiv.org/abs/2402.04253 ICML 2024- 2023-12 ProTIP: Progressive Tool Retrieval Improves Planning https://arxiv.org/abs/2312.10332 EACL 2024 Workshop 🏷️ Multi-Label Classification MLC - 2024-09 Efficient and Scalable Estimation of Tool Representations in Vector Space https://arxiv.org/abs/2409.02141 - 2024-09 TinyAgent: Function Calling at the Edge https://arxiv.org/abs/2409.00608 EMNLP 2024 Demo 📚 Vocabulary-based Retrieval - 2025-03 Chain-of-Tools: Utilizing Massive Unseen Tools in the CoT Reasoning of Frozen Language Models https://arxiv.org/abs/2503.16779 - 2024-10 Toolken+: Improving LLM Tool Usage with Reranking and a Reject Option https://arxiv.org/abs/2410.12004 EMNLP 2024 Findings - 2024-10 ToolGen: Unified Tool Retrieval and Calling via Generation https://arxiv.org/abs/2410.03439 ICLR 2025- 2024-07 Concise and Precise Context Compression for Tool-Using Language Models https://arxiv.org/abs/2407.02043 ACL 2024 Findings - 2023-05 ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings https://arxiv.org/abs/2305.11554 NeurIPS 2023 oral ▶️ Tool Calling 📝 In-Place Parameter Filling - 2024-01 Efficient Tool Use with Chain-of-Abstraction Reasoning https://arxiv.org/abs/2401.17464 COLING 2025 - 2023-02 Toolformer: Language Models Can Teach Themselves to Use Tools https://arxiv.org/abs/2302.04761 NeurIPS 2023 oral ⚡ Parallel Tool Calling - 2026-02 W&D: Scaling Parallel Tool Calling for Efficient Deep Research Agents https://arxiv.org/abs/2602.07359 - 2024-11 CATP-LLM: Empowering Large Language Models for Cost-Aware Tool Planning https://arxiv.org/abs/2411.16313 ICCV 2025- 2024-05 An LLM-Tool Compiler for Fused Parallel Function Calling https://arxiv.org/abs/2405.17438 - 2023-12 An LLM Compiler for Parallel Function Calling https://arxiv.org/abs/2312.04511 ICML 2024 💰 Cost-Aware Tool Calling - 2025-07 A Joint Optimization Framework for Enhancing Efficiency of Tool Utilization in LLM Agents https://aclanthology.org/2025.findings-acl.1149/ ACL 2025 Findings- 2025-05 Distilling LLM Agent into Small Models with Retrieval and Code Tools https://arxiv.org/abs/2505.17612 - 2025-03 Alignment for Efficient Tool Calling of Large Language Models https://arxiv.org/abs/2503.06708 EMNLP 2025 - 2025-02 ToolCoder: A Systematic Code-Empowered Tool Learning Framework for Large Language Models https://arxiv.org/abs/2502.11404 ACL 2025- 2024-02 Budget-Constrained Tool Learning with Planning https://arxiv.org/abs/2402.15960 ACL 2024 Findings- 2024-01 TroVE: Inducing Verifiable and Efficient Toolboxes for Solving Programmatic Tasks https://arxiv.org/abs/2401.12869 ICML 2024 ⚙️ Efficient Test-Time Scaling 🎯 Efficient Tool Calling with Post-training 🛠️ Tool-Integrated Reasoning TIR ✅ Selective Invocation - 2025-09 TableMind: An Autonomous Programmatic Agent for Tool-Augmented Table Reasoning https://arxiv.org/abs/2509.06278 WSDM 2026- 2025-05 Tool-Star: Empowering LLM-Brained Multi-Tool Reasoner via Reinforcement Learning https://arxiv.org/abs/2505.16410 - 2025-02 SMART: Self-Aware Agent for Tool Overuse Mitigation https://arxiv.org/abs/2502.11435 ACL 2025 Findings- 2024-03 Agent-FLAN: Designing Data and Methods of Effective Agent Tuning for Large Language Models https://arxiv.org/abs/2403.12881 ACL 2024 Findings 🎓 Cost-Aware Policy Optimization - 2026-01 ET-Agent: Incentivizing Effective Tool-Integrated Reasoning Agent via Behavior Calibration https://arxiv.org/abs/2601.06860 - 2025-10 PORTool: Tool-Use LLM Training with Rewarded Tree https://arxiv.org/abs/2510.26020 - 2025-10 A$^2$FM: An Adaptive Agent Foundation Model for Tool-Aware Hybrid Reasoning https://arxiv.org/abs/2510.12838 - 2025-09 Toward Effective Tool-Integrated Reasoning via Self-Evolved Preference Learning https://arxiv.org/abs/2509.23285 - 2025-09 TableMind: An Autonomous Programmatic Agent for Tool-Augmented Table Reasoning https://arxiv.org/abs/2509.06278 WSDM 2026- 2025-07 Agentic Reinforced Policy Optimization https://arxiv.org/abs/2507.19849 - 2025-07 AutoTIR: Autonomous Tools Integrated Reasoning via Reinforcement Learning https://arxiv.org/abs/2507.21836 - 2025-05 Reinforced Internal-External Knowledge Synergistic Reasoning for Efficient Adaptive Search Agent https://arxiv.org/abs/2505.07596 - 2025-05 Agentic Reasoning and Tool Integration for LLMs via Reinforcement Learning https://arxiv.org/abs/2505.01441 - 2025-04 ToolRL: Reward is All Tool Learning Needs https://arxiv.org/abs/2504.13958 NeurIPS 2025- 2025-04 ReTool: Reinforcement Learning for Strategic Tool Use in LLMs https://arxiv.org/abs/2504.11536 - 2025-04 Synthetic Data Generation & Multi-Step RL for Reasoning & Tool Use https://arxiv.org/abs/2504.04736 COLM 2025 - 2025-04 Acting Less is Reasoning More Teaching Model to Act Efficiently https://arxiv.org/abs/2504.14870 📑 Table of Planning Contents Planning framework encompasses single-agent planning efficiency and multi-agent collaborative strategies for enhanced decision-making. 🤖 Single-Agent Planning Efficiency 💰 Adaptive Budgeting and Control - 2026-03 SpecEyes: Accelerating Agentic Multimodal LLMs via Speculative Perception and Planning https://arxiv.org/abs/2603.23483 - 2025-11 Budget-Aware Tool-Use Enables Effective Agent Scaling https://arxiv.org/abs/2511.17006 - 2025-09 Learning When to Plan: Efficiently Allocating Test-Time Compute for LLM Agents https://arxiv.org/abs/2509.03581 - 2025-06 Query-Level Uncertainty in Large Language Models https://arxiv.org/abs/2506.09669 - 2023-12 ReST meets ReAct: Self-Improvement for Multi-Step Reasoning LLM Agent https://arxiv.org/abs/2312.10003 ICLR 2024 Workshop - 2023-05 SwiftSage: A Generative Agent with Fast and Slow Thinking for Complex Interactive Tasks https://arxiv.org/abs/2305.17390 NeurIPS 2023 spotlight- 2023-03 Reflexion: Language Agents with Verbal Reinforcement Learning https://arxiv.org/abs/2303.11366 NeurIPS 2023 🔍 Structured Search - 2025-05 Cost-Augmented Monte Carlo Tree Search for LLM-Assisted Planning https://arxiv.org/abs/2505.14656 - 2023-12 ProTIP: Progressive Tool Retrieval Improves Planning https://arxiv.org/abs/2312.10332 - 2023-10 ToolChain : Efficient Action Space Navigation in Large Language Models with A Search https://arxiv.org/abs/2310.13227 ICLR 2024 poster - 2023-10 Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models https://arxiv.org/abs/2310.04406 ICML 2024 📋 Task Decomposition - 2025-12 Video-Browser: Towards Agentic Open-web Video Browsing https://arxiv.org/abs/2512.23044 - 2025-05 Alita: Generalist Agent Enabling Scalable Agentic Reasoning with Minimal Predefinition and Maximal Self-Evolution https://arxiv.org/abs/2505.20286 - 2025-03 ReSo: A Reward-driven Self-organizing LLM-based Multi-Agent System for Reasoning Tasks https://arxiv.org/abs/2503.02390 EMNLP 2025- 2024-11 BudgetMLAgent: A Cost-Effective LLM Multi-Agent system for Automating Machine Learning Tasks https://arxiv.org/abs/2411.07464 AIMLSystems 2024 - 2024-02 AutoGPT+P: Affordance-based Task Planning with Large Language Models https://arxiv.org/abs/2402.10778 - 2023-05 ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Models https://arxiv.org/abs/2305.18323 - 2023-03 HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face https://arxiv.org/abs/2303.17580 NeurIPS 2023 🎯 Policy Optimization - 2025-09 Planner-R1: Reward Shaping Enables Efficient Agentic RL with Smaller LLMs https://arxiv.org/abs/2509.25779 - 2025-08 Encouraging Good Processes Without the Need for Good Answers: Reinforcement Learning for LLM Agent Planning https://arxiv.org/abs/2508.19598 EMNLP 2025 Industry - 2025-05 Planning without Search: Refining Frontier LLMs with Offline Goal-Conditioned RL https://arxiv.org/abs/2505.18098 NeurIPS 2025- 2025-02 QLASS: Boosting Language Agent Inference via Q-Guided Stepwise Search https://arxiv.org/abs/2502.02584 ICML 2025- 2024-03 Trial and Error: Exploration-Based Trajectory Optimization for LLM Agents https://arxiv.org/abs/2403.02502 ACL 2024 🧠 Memory and Skill Acquisition - 2025-10 GAP: Graph-Based Agent Planning with Parallel Tool Use and Reinforcement Learning https://arxiv.org/abs/2510.25320 - 2024-07 Sibyl: Simple yet Effective Agent Framework for Complex Real-world Reasoning https://arxiv.org/abs/2407.10718 - 2024-06 GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models https://arxiv.org/abs/2406.14550 EMNLP 2024 Findings - 2024-02 Graph-enhanced Large Language Models in Asynchronous Plan Reasoning https://arxiv.org/abs/2402.02805 ICML 2024- 2023-05 Voyager: An Open-Ended Embodied Agent with Large Language Models https://arxiv.org/abs/2305.16291 TMLR 2024 👥 Multi-Agent Collaborative Efficiency 🕸️ Topological Efficiency and Sparsification - 2025-09 MARS: toward more efficient multi-agent collaboration for LLM reasoning https://arxiv.org/abs/2509.20502 - 2025-08 SafeSieve: From Heuristics to Experience in Progressive Pruning for LLM-based Multi-Agent Communication https://arxiv.org/abs/2508.11733 AAAI 2026- 2025-03 AgentDropout: Dynamic Agent Elimination for Token-Efficient and High-Performance LLM-Based Multi-Agent Collaboration https://arxiv.org/abs/2503.18891 ACL 2025- 2025-02 S$^2$-MAD: Breaking the Token Barrier to Enhance Multi-Agent Debate Efficiency https://arxiv.org/abs/2502.04790 NAACL 2025 - 2024-10 Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems https://arxiv.org/abs/2410.02506 ICLR 2025- 2024-09 GroupDebate: Enhancing the Efficiency of Multi-Agent Debate Using Group Discussion https://arxiv.org/abs/2409.14051 - 2024-06 Scaling Large Language Model-based Multi-Agent Collaboration https://arxiv.org/abs/2406.07155 ICLR 2025- 2024-06 Chain of Agents: Large Language Models Collaborating on Long-Context Tasks https://arxiv.org/abs/2406.02818 NeurIPS 2024 ⚙️ Protocol and Context Optimization - 2025-10 Stop Wasting Your Tokens: Towards Efficient Runtime Multi-Agent Systems https://arxiv.org/abs/2510.26585 - 2025-09 Free-MAD: Consensus-Free Multi-Agent Debate https://arxiv.org/abs/2509.11035 - 2025-07 CONSENSAGENT: Towards Efficient and Effective Consensus in Multi-Agent LLM Interactions Through Sycophancy Mitigation https://aclanthology.org/2025.findings-acl.1141/ ACL 2025 Findings - 2025-07 CodeAgents: A Token-Efficient Framework for Codified Multi-Agent Reasoning in LLMs https://arxiv.org/abs/2507.03254 - 2024-05 Smurfs: Multi-Agent System using Context-Efficient DFSDT for Tool Planning https://arxiv.org/abs/2405.05955 NAACL 2025 📊 Distilling Coordination into Planning - 2025-11 SMAGDi: Socratic Multi Agent Interaction Graph Distillation for Efficient High Accuracy Reasoning https://arxiv.org/abs/2511.05528 NeurIPS 2025 Workshop - 2025-06 Debate, Reflect, and Distill: Multi-Agent Feedback with Tree-Structured Preference Optimization for Efficient Language Model Enhancement https://arxiv.org/abs/2506.03541 ACL 2025 Findings- 2024-02 MAGDi: Structured Distillation of Multi-Agent Interaction Graphs Improves Reasoning in Smaller Language Models https://arxiv.org/abs/2402.01620 ICML 2024 BibTeX @misc{yang2026efficientagentsmemorytool, title={Toward Efficient Agents: Memory, Tool learning, and Planning}, author={Xiaofang Yang and Lijun Li and Heng Zhou and Tong Zhu and Xiaoye Qu and Yuchen Fan and Qianshan Wei and Rui Ye and Li Kang and Yiran Qin and Zhiqiang Kou and Daizong Liu and Qi Li and Ning Ding and Siheng Chen and Jing Shao}, year={2026}, eprint={2601.14192}, archivePrefix={arXiv}, primaryClass={cs.AI}, url={https://arxiv.org/abs/2601.14192}, }