今日速览
| 序号 | 标题 | 来源 | 日期 | 主题 | 推荐等级 |
|---|---|---|---|---|---|
| 1 | Human-guided physics-constrained AI agents construct an auditable model of soil-plug evolution | arXiv | 2026-09-20 | AI-Agent | 高 |
| 2 | Graph Memory for LLM Agents: At What Cost? A Comparative Evaluation of Query, Ingest, and Update Performance Across Graph Database Engines | arXiv | 2026-09-20 | AI-Agent | 高 |
重点论文与技术动态
1. Human-guided physics-constrained AI agents construct an auditable model of soil-plug evolution
- 来源:arXiv
- 日期:2026-09-20
- 作者/机构:Jie Shi, Yimin Lu, Zhongkun Ouyang
- 主题标签:
AI-Agent,arXiv - 推荐等级:高
- 分类:cs.CE, cs.AI
一句话结论
该工作提出人在环、物理约束多智能体流程,自动化并审计理论到求解器建模,用于吸力沉箱土塞演化。
核心内容
- 人类专家设定物理与建模边界,智能体检索证据、推导方程、实现求解器并审计理论到代码。
- 物理知识与输入准备后,智能体运行 2.9 小时生成并审计 6 个模型。
- 几何基线加入渗流驱动孔隙比演化,使 14 剖面平均绝对最终隆起误差从 58.4% 降至 9.0%。
方法与数据
- 采用人在环、物理约束多智能体工作流,结合盲测重放与独立审计;方程与求解器细节未明确。
- 数据含 14 剖面、9 最终状态案例、5 过程历史端点;最终模型纳入近壁膨胀,平均绝对百分比误差 12.4%,端点误差 4.2%。
价值判断
- 值得关注:将多智能体从任务自动化推进到人类治理的工程求解器,强调物理约束与审计。
- 可复用点:可迁移到理论、离散化、代码与验证一致性的工程预测任务,形成检索、推导、实现、审计闭环。
- 局限/待核查:摘要未明确泛化性、计算成本、专家投入与审计标准,需核查。
摘要
Engineering predictions require physical mechanisms to be translated consistently into equations, discretization, code, and validation, yet errors can propagate despite local checks. Artificial-intelligence (AI) agents automate scientific tasks, but coordinating and independently auditing the theory-to-solver process under physical constraints and human oversight remains unresolved. We introduce a human-in-the-loop, physics-constrained multi-agent workflow where human experts define admissible physics and modeling boundaries, while agents retrieve evidence, derive equations, implement solvers, and audit the theory-to-code chain. Applied to soil-plug evolution during suction-caisson installation, the workflow generated and audited 6 formulations in 2.9 h of agent execution once physical knowledge and inputs were prepared. Among these formulations, adding seepage-driven soil void-ratio evolution to the geometric baseline reduced mean absolute final-heave error from 58.4% to 9.0% across 14 profiles; the selected model further incorporated near-wall dilation and achieved mean absolute percentage errors of 12.4% across 9 final-state cases and 4.2% at the endpoints of 5 process histories. Beyond predictive performance, blinded replay recovered all 9 target problems, while an independent audit uncovered 5 implementation problems after 36 predefined checks had passed. Overall, this work extends multi-agent AI beyond task automation toward human-governed engineering solvers.2. Graph Memory for LLM Agents: At What Cost? A Comparative Evaluation of Query, Ingest, and Update Performance Across Graph Database Engines
- 来源:arXiv
- 日期:2026-09-20
- 作者/机构:Donald Nguyen, Gurbinder Gill, Hadi Ahmadi, Christopher J. Rossbach
- 主题标签:
AI-Agent,arXiv - 推荐等级:高
- 分类:cs.DB, cs.AI
一句话结论
该研究以合成生物医学属性图和20类查询比较8种图数据库/引擎,发现没有绝对最快系统,总成本主要由批量摄取吞吐决定。
核心内容
- 构建102万节点、534万行节点/边的合成属性图,覆盖邻域、路径、交集、反连接、聚合、top-k、时间过滤、全扫描和连接。
- 在三个规模下比较 Corvic AI 与7个引擎,报告延迟几何均值、批量摄取吞吐、点更新延迟和正确性。
- Ladybug 在窄有界邻域更快,Corvic AI 在大范围扫描/连接更快,DuckPGQ 因计划选择更慢;摄取吞吐差三个数量级,主导少于约10^5次查询/刷新的场景。
方法与数据
- 使用合成图与20个查询,规模跨三个数量级;建立简单 TCO 模型,表达摄取/查询权衡随查询量变化。
- 硬件、部署和并发条件摘要未明确。
价值判断
- 值得关注:把数据可查询成本从厂商叙事中分离,强调摄取吞吐影响总成本。
- 可复用点:可复用查询清单、延迟/吞吐/更新/正确性指标和交叉点 TCO 框架。
- 局限/待核查:仅合成生物医学形状图与20个查询;真实生产、并发、持久化和运维摘要未明确。