1
Scaling Automatic Research Agents via World Models
用世界模型为自动研究代理注入规划能力，探索智能体规模化的前沿路径。
arXiv:2608.12564v1 Announce Type: new Abstract: Automating empirical research is a long-standing direction of AI. Recent automatic research (AutoResea…
2
The Algorithmic-Human Manager: AI, Apps, and Workers in the Indian Gig Economy
AI正在重塑印度零工经济的管理模式，这篇论文揭示了算法与人结合的独特实践。
arXiv:2606.19975v1 Announce Type: cross Abstract: This paper examines the impact of artificial intelligence and digital technologies on the blue-colla…
3
Closing the Auto-Research Loop: An AI Co-Scientist for Production Search Ranking
LLM智能体联手云算力，自动迭代实验闭环，直击旅游平台搜索排名优化痛点。
arXiv:2603.22376v2 Announce Type: replace-cross Abstract: We present an AI Co-Scientist framework that closes the research loop for the production sea…
4
Show HN: Lead Qualifier – research agent turning a LinkedIn URL into a dossier
输入LinkedIn链接，自动生成包含公司背景、融资、技术栈和社交画像的详细档案，并帮你起草外联文案，让销售准备事半功倍。
Article URL: https://lead.robowrite.ai/ Comments URL: https://news.ycombinator.com/item?id=48455685 Points: 2 # Comments: 0
5
Don't Gamble, GAMBLe: An Analytical Framework for AI-Driven Research Systems
提出GAMBLe分析框架，为AI驱动研究系统提供系统性评估方法，避免盲目探索。
arXiv:2606.02863v1 Announce Type: new Abstract: AI-Driven Research Systems (ADRS) -- systems coupling LLMs with automated evaluation to discover algor…
6
MLIPilot: LLM-Driven Auto-Research for Machine-Learned Interatomic Potentials
LLM驱动的自动研究框架，优化机器学习原子间势的构建，平衡精度与效率
arXiv:2605.30889v1 Announce Type: cross Abstract: Constructing production-quality machine-learned interatomic potentials (MLIPs) requires balancing ac…
7
Discovering Cooperative Pipelines: Autoresearch for Sequential Social Dilemmas
自动发现合作管道的新方法，攻克序贯社会困境中的多智能体协作难题。
arXiv:2605.30003v1 Announce Type: cross Abstract: We study two-level autoresearch for cooperation: an outer-loop AI agent autonomously redesigns the i…
8
DeepSeek陈德里开发自动研究Skill，写一篇论文人类只动脑2小时
DeepSeek研究员陈德里研发自动研究技能Agent，写一篇论文人类仅需脑力2小时，99%内容由AI完成，揭示迈向L5自主科研的关键瓶颈。
“1%是我写的，99%是Agent写的。”
9
How Far Are We From True Auto-Research?
系统评估AI自动生成论文的质量差距，ResearchArena框架揭示可行性不等于真智能。
arXiv:2605.19156v1 Announce Type: cross Abstract: Recent auto-research systems can produce complete papers, but feasibility is not the same as quality…