今日速览
| 序号 | 标题 | 来源 | 日期 | 主题 | 推荐等级 |
|---|---|---|---|---|---|
| 1 | AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents | arXiv | 2026-07-02 | AI-Agent | 高 |
| 2 | Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support | arXiv | 2026-07-02 | AI-Agent | 高 |
重点论文与技术动态
1. AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents
- 来源：arXiv
- 日期：2026-07-02
- 作者/机构：Xiangchen Cheng, Yunwei Jiang, Jianwen Sun, Zizhen Li, Chuanhao Li, Xiangcheng Cao
- 主题标签：
AI-Agent,arXiv - 推荐等级：高
- 分类：cs.AI, cs.CL
一句话结论
AgenticSTS构建了一个有界记忆测试平台，使长周期LLM代理的决策研究可控制且可隔离，避免上下文混杂问题。
核心内容
- 提出有界记忆合同，每个决策基于类型化检索组装的全新用户消息，无原始跨决策转录，确保提示上下文始终简洁。
- 在Slay the Spire 2游戏中实例化，该游戏需数百个战术决策，人类胜率16%而LLMs零胜，任务硬但未饱和。
- 实验显示添加技能层后胜率从3/10提升至6/10（p≈0.37），表明显式记忆层显著影响决策，但结果统计上不显著。
方法与数据
- 在Slay the Spire 2游戏中收集298个完成轨迹数据，包含条件标签、冻结记忆/技能快照及提示记录。
- 提供完整分析脚本和可复现测试平台，支持显式记忆层的孤立分析。
价值判断
- 值得关注：长周期LLM代理在复杂决策任务中表现不足（人类胜率16%，LLMs零胜），任务未饱和且具研究价值。
- 可复用点：测试平台和方法可重复用于研究显式记忆层对决策的影响，提供标准化评估框架。
- 局限/待核查：实验样本量小（10局），结果统计上不显著（Fisher精确检验p≈0.37），需更大规模验证。
摘要
Memory for a long-horizon LLM agent is a contract about what each future decision is allowed to see. The simplest contract appends past observations, tool calls, and reflections to every prompt, which makes prior context easy to access but also turns it into a jumbled mixture in which the effect of any single memory component is hard to isolate. We introduce and instrument an alternative bounded contract: every decision is made from a fresh user message assembled by typed retrieval, with no raw cross-decision transcript appended. The prompt thus stays bounded across runs of any length, and any single layer can be ablated in isolation. We instantiate the contract in Slay the Spire 2, a closed-rule stochastic deck-building game whose runs require hundreds of tactical and strategic decisions. A public online benchmark of frontier LLMs on the same game reports zero wins at the lowest difficulty across five configurations, and the developer-reported human win rate at the same difficulty is 16%; the task is hard but not saturated. Within our harness, a fixed-A0 ablation shows the largest observed difference when triggered strategic skills are enabled: the no-store baseline wins 3/10 games and adding the skill layer 6/10. At this sample size the comparison is directional rather than statistically decisive (Fisher exact p\approx0.37); a cross-backbone probe and public accumulating-context baselines are reported as operational comparisons rather than controlled tests of the contract variable itself. We release a reproducible testbed: 298 completed trajectories with condition tags, frozen memory/skill snapshots, prompt records, and analysis scripts -- an agent design and a validated, reusable methodology for studying how explicit memory layers shape long-horizon LLM-agent decisions.2. Copewell: A Multi-Agent Swarm Architecture for Equitable Mental Wellness Support
- 来源：arXiv
- 日期：2026-07-02
- 作者/机构：Seren Yenikent, Jack Vinijtrongjit, Katherine Ng
- 主题标签：
AI-Agent,arXiv - 推荐等级：高
- 分类：cs.AI, cs.CY, cs.HC
一句话结论
Copewell通过多代理群集架构在心理健康支持中实现了公平与安全原则的早期操作化。
一段话。该系统针对全球近10亿心理健康障碍患者，尤其解决75%低收入和中等收入国家无治疗率问题，通过整合多源数据、情绪映射和双模式干预，提供动态、个性化的支持，同时确保隐私与伦理安全。
核心内容
- 引入多源评估框架，整合自报告、生理和上下文数据以减轻算法偏见。
- 采用Russell环形模型进行情绪映射，动态路由用户至专业化AI代理。
- 实施双模式干预，结合对话支持与证据-based感官健康协议。
方法与数据
- 多源数据整合（自报告、生理、上下文）用于评估。
- 嵌入伦理监督代理确保隐私与安全。
价值判断
- 值得关注：解决低收入国家心理健康服务可及性问题，减少75%无治疗率。
- 可复用点：多源评估框架和情绪映射模型可迁移至其他健康AI应用。
- 局限/待核查：需未来实证评估验证干预效果。