8月22日 早间 AI 资讯速览，共 10 条（来源：GitHub、arXiv）。

1. deepseek-ai/deepseek-harness（⭐181190）

DeepSeek Harness: Everything is a Plugin.

来源：GitHub · 阅读原文

2. anywhere-labs/deepseek-harness-desktop（⭐17549）

为 DeepSeek Harness (DSH) 插件生态打造的现代化桌面端解决方案。万物皆「插件」，桌面本身也是「插件」。

来源：GitHub · 阅读原文

3. guillaumemeyer/watermarks-remover（⭐16621）

Strip multi-vendor AI provenance marks: Unicode text hygiene, statistical rewrite hooks, and C2PA/metadata from PNG/JPEG/SVG/PDF/DOCX/HTML/MD

来源：GitHub · 阅读原文

4. awesome-dsh-plugin/awesome-dsh-plugin（⭐11151）

A curated list of plugins for DeepSeek Harness (dsh) · DeepSeek Harness 插件精选列表

来源：GitHub · 阅读原文

5. yjh051108/dsh-routing-suite（⭐6560）

dsh-routing-suite — injector + router-standard kit: install the runtime injector first, then the task-aware reasoning-mode router preset (measured P1-P23).

来源：GitHub · 阅读原文

6. zhu1090093659/dsh-web-ui（⭐5421）

Plugin and skin collection for DeepSeek Harness (DSH) Web UI - task board, git graph, right-side panel, remote mobile UI, pet, live token stats, and skin center.

来源：GitHub · 阅读原文

7. G-CARL: Grounded Checklist-Aligned Reward Learning for Patient-Oriented Medical Report Interpretation

Travel behavior research increasingly combines digital data collection with predictive modeling, yet these stages are often developed and evaluated separately. This study proposes a three-agent workfl

来源：arXiv · 阅读原文

8. Inducing Task Models from Computer-Use Traces

Recursive self-improvement (RSI) asks whether an AI system can improve the process that produces AI systems, so that the next system inherits the improvement. That process is the training algorithm: a

来源：arXiv · 阅读原文

9. Pandora's AI Model Routing Box: Efficient Allocation with Costly Value Estimation

Mid-training is increasingly recognized as a critical stage for shaping the capabilities of large language models. Recent work has shown that targeted mid-training can strengthen reasoning-intensive a

来源：arXiv · 阅读原文

10. Phantom Gains: Auditing Self-Improvement Against a Measured Null

Large language models often fail to answer questions about a bounded document collection when the source documents are not retrieved at inference time. We study this setting as document knowledge inte

来源：arXiv · 阅读原文

以上资讯由 AI 机器人「小赫」自动采集整理，内容版权归原作者所有。

本文链接:https://it72.com/12844.htm