arXiv:2604.26235cs.CRcs.AI2026-04

LATTICE评测加密货币智能体在真实场景中的决策辅助能力。

LATTICE: Evaluating Decision Support Utility of Crypto Agents

论文配图:LATTICE: Evaluating Decision Support Utility of Crypto Agents
图 1 · 摘自论文原文
  • 设计六维评估体系与16类任务,覆盖完整智能体工作流。
  • 用LLM裁判自动评分,无需专家标注或外部数据。
  • 评估六款真实产品,发现性能差异在维度级更显著。

我们提出LATTICE,一个用于评估加密货币智能体在真实用户场景中决策支持效用的基准。现有基准多聚焦于推理或结果评估,未衡量智能体辅助用户决策的能力。LATTICE通过三方面弥补这一缺口:(1) 定义六项评估维度,涵盖关键决策支持属性;(2) 提出16种任务类型,覆盖端到端加密助手工作流;(3) 使用大模型裁判基于这些维度和任务自动评分。关键在于,维度与任务可大规模用大模型裁判评估,无需依赖专家标注或外部数据。相较之下,其裁判标准可随新维度、任务、标准和人类反馈持续审计与更新,确保评估的可靠与可扩展。不同于其他比较通用框架的基座模型,我们使用LATTICE评估实际部署的加密助手产品,凸显编排与用户体验设计对智能体质量的重要性。本文在1,200个多样化查询上评估六款真实世界加密助手,并报告各维度、任务与查询类别下的表现细分。实验表明,多数助手总体得分相近,但在维度与任务层级差异显著,暗示决策支持质量存在实质性权衡:不同优先级的用户可能更适合不同的助手,而非仅由总分决定。为支持可复现研究,我们开源了本文所用全部LATTICE代码与数据。

原文摘要 · Abstract (English)

We introduce LATTICE, a benchmark for evaluating the decision support utility of crypto agents in realistic user-facing scenarios. Prior crypto agent benchmarks mainly focus on reasoning-based or outcome-based evaluation, but do not assess agents' ability to assist user decision-making. LATTICE addresses this gap by: (1) defining six evaluation dimensions that capture key decision support properties; (2) proposing 16 task types that span the end-to-end crypto copilot workflow; and (3) using LLM judges to automatically score agent outputs based on these dimensions and tasks. Crucially, the dimensions and tasks are designed to be evaluable at scale using LLM judges, without relying on ground truth from expert annotators or external data sources. In lieu of these dependencies, LATTICE's LLM judge rubrics can be continually audited and updated given new dimensions, tasks, criteria, and human feedback, thus promoting reliable and extensible evaluation. While other benchmarks often compare foundation models sharing a generic agent framework, we use LATTICE to assess production-level agents used in actual crypto copilot products, reflecting the importance of orchestration and UI/UX design in determining agent quality. In this paper, we evaluate six real-world crypto copilots on 1,200 diverse queries and report breakdowns across dimensions, tasks, and query categories. Our experiments show that most of the tested copilots achieve comparable aggregate scores, but differ more significantly on dimension-level and task-level performance. This pattern suggests meaningful trade-offs in decision support quality: users with different priorities may be better served by different copilots than the aggregate rankings alone would indicate. To support reproducible research, we open-source all LATTICE code and data used in this paper.

智能体评估加密货币决策支持

Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。