Tag
A free software developer shares their move away from Debian to Chimera Linux, drawn by its firm anti-LLM contribution policy, unconventional tech stack, and welcoming community.
The paper argues that standard AI-text detectors measure the wrong thing for academic venues and proposes policy-conditioned AI-use detection, an evidentiary framework that takes venue policies as explicit inputs, reports calibrated hypotheses and uncertainty instead of binary verdicts, and builds benchmarks from reproducible compliant/non-compliant workflow pipelines.
StepCAD 提出了一种将状态条件化 CAD 策略与 IoU 引导树搜索相结合的生成式优化方法，从 3D 网格重建可执行 CAD 程序，并发布了包含 150 万个程序的 ARCADE-1.5M 大规模数据集，在多个基准上实现了最先进的几何重建精度。
The article proposes an LLM policy for the GNOME project that bans LLM-generated contributions to protect the human-centric values of the community, contrasting with KDE's approach and emphasizing social norms over workflow management.
Flathub announces it will no longer accept AI-generated submissions, citing volunteer reviewer burnout and spam from automated tooling like OpenClaw. The post explores the history of Flathub's review process and the community's reaction.
NLNet Labs announces a policy restricting the use of LLMs in code and documentation contributions, requiring disclosure of LLM use and prohibiting AI-generated code.
This paper presents a two-level autoresearch framework where an outer-loop AI agent autonomously optimizes inner-loop LLM policy-synthesis pipelines for multi-agent sequential social dilemmas, achieving superior performance and discovering objective-specific mechanisms like fairness under a maximin welfare objective.