ArXiv Collision Cognition Skill#

元数据#

字段值名称arxiv-collision-cognition版本1.0.0作者KingOfZhao发布日期2026-03-31置信度95%

核心能力#

将 ArXiv 论文与你当前项目的已知知识进行四向碰撞，提取跨域洞见：

1. 正面碰撞：论文核心方法与你的现有方案直接对比，找出可迁移技术
2. 反面碰撞：论文的局限性/失败案例是否正是你的场景优势？
3. 侧面碰撞：论文的非主要贡献（ablation / appendix）中是否有隐藏宝藏？
4. 整体碰撞：论文的研究方向趋势是否预示你项目的未来路径？

已知/未知分离 + 文件记忆 + 人机闭环全部在此场景中应用。

安装命令#

bash

clawhub install arxiv-collision-cognition
# 或手动安装
cp -r skills/arxiv-collision-cognition ~/.openclaw/skills/

核心论文库（按主题分组）#

自进化 / 自改进 Agent#

论文ArXiv ID对本项目的直接价值Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharing2602.04837群体进化+经验共享，与"人机闭环+四向碰撞"一致，可扩展为多Agent迭代SkillA Survey of Self-Evolving Agents2507.21046最全自进化综述（What/When/How），作为理论骨架SAGE: Multi-Agent Self-Evolution for LLM Reasoning2603.15255Challenger+Planner+Solver+Critic四Agent闭环，与四向碰撞推理几乎一致Self-evolving Embodied AI2602.04411记忆自更新+任务自切换+模型自进化，匹配具身智能演进路径

DiePre 视觉 / 照片→CAD / 工业线条检测#

论文ArXiv ID对本项目的直接价值Generating CAD Code with Vision-Language Models for 3D Designs2410.05340VLM生成CAD+迭代验证（CADCodeVerify），可升级照片→SVG/DXF管道From 2D CAD Drawings to 3D Parametric Models: A Vision-Language Approach2412.118922D图纸→参数化3D，解决透视矫正和参数化问题Tool-Augmented VLLMs as Generic CAD Task Solvers(ICCV 2025)VLLM+工具调用做通用CAD任务，适合封装OpenCV管道为可调用Skill

具身智能 / Vision-Language-Action (VLA)#

论文ArXiv ID对本项目的直接价值Efficient Vision-Language-Action Models for Embodied Manipulation2510.17111VLA高效优化综述（低延迟+内存优化），适合M1 Max本地部署Vlaser: Vision-Language-Action Model with Synergistic Embodied Reasoning2510.11027强化具身推理的VLA，可作为DiePre"照片→动作决策"理论基础

Agent 记忆系统#

论文ArXiv ID对本项目的直接价值Memory in the Age of AI Agents2512.13564最新Agent记忆综述（token/parametric/latent），可升级HEARTBEAT记忆模块Beyond RAG for Agent Memory2602.02007"Decoupling and Aggregation"超越传统RAG，解决长时序记忆问题

每日筛选工作流（HEARTBEAT 集成）#

text

每个心跳周期执行:
1. 从 ArXiv API 拉取匹配关键词的最新论文（每日上限5篇）
   关键词: self-evolving agent, vision-language model, embodied AI,
           CAD generation, agent memory, packaging quality control
2. 对每篇论文执行四向碰撞推理
3. 碰撞结果写入 collision_log/arxiv_{id}_{date}.json
4. 置信度 ≥ 85% 的洞见 → 加入 actionable_insights 队列
5. 人类标记已验证的洞见 → 更新项目知识库

调用方式#

python

from skills.arxiv_collision_cognition import ArxivCollisionCognition

acc = ArxivCollisionCognition(workspace=".")

# 用 ArXiv ID 触发碰撞
result = acc.collide(
    arxiv_id="2602.04837",
    project_context={
        "domain": "packaging quality control",
        "known": ["corrugated board defect types", "current detection accuracy 87%"],
        "unknown": ["thermal deformation modeling", "acoustic inspection feasibility"]
    }
)

print(result.confidence)         # 置信度
print(result.actionable_insights) # 可操作洞见列表
print(result.collision_log)      # 四向碰撞详情

# 批量碰撞（每日筛选）
results = acc.daily_screening(max_papers=5, keywords=["self-evolving", "CAD", "VLA"])