2026年9月19日 技术热点总结

项目简介：Claude Code plugin that replaces the compaction summary with Jev decisions: every tool call and result is scored in one fast request, stale ones are dropped or truncated, everything kept stays verbatim.

技术栈：TypeScript

核心介绍：Claude Code plugin that replaces the compaction summary with Jev decisions: every tool call and result is scored in one fast request, stale ones are dropped or truncated, everything kept stays verbatim. Also usable as an npm Most context compaction asks an LLM to summarize old turns. A summary is lossy: a file path, exact error, constraint, or command c…

项目简介：Semantic ifs from open models, on a 3090 at home. Independent; not affiliated with Jev or TypeSafe.

技术栈：Python

核心介绍：> Independent research project. SemIf was formerly called OpenJev. It is not affiliated with or endorsed by TypeSafe. Jev, TypeSafe, and other names and marks are the property of their respective owners. No infringement is intended. Most agent decisions are small: *route this*, *retry that*, *does the evidence support X?* A chat model can answer them, but it spends time generating text that software immediately parses back into an if statement.

**项目数据：*…

项目简介：Private source preview: native Jianying drafts, isolated editing/export, and standalone Agent Skill.

技术栈：Python

核心介绍：剪映专业版 macOS 的本地自动化工具：从剪辑计划生成可继续编辑的原生草稿， 在独立副本里修改已有多轨草稿，并在明确要求时调用本机原生引擎导出 MP4。 这是私有源码预览，不是剪映官方 SDK，也不是无需安装剪映的独立剪辑软件。 官方程序库、账号资料、真实素材与效果资源均不随仓库分发。 Skill 是调用入口，不包含剪映引擎。单独安装 Skill 后仍需取得本项目的访问权限、 检出核心代码，并安装匹配版本的剪映。 成功生成文件不等于原生播放、冷重开、视觉一致性、素材商用许可都已通过。 详见 验证状态 和 分发范围。 版本号、build、bundle ID、官方库哈希和写入前完整应用签名都会核对。 不支持的版本会停止，不自动改常量、降级效果或下载资源。 这份预览尚未完成另一台干净机器的安装验收。 仓库是私有的，以下检出操作需要已获授权的 GitHub 账号： gh repo clone mcncarl/jianying-headless python3 tools/build_native_codec…

项目简介：One AI trade decision every Monad block. Jev on Kuru MON-USDC.

技术栈：TypeScript

核心介绍：One decision every Monad block. A TypeSafe Jev model watches the Kuru MON-USDC order book and answers buy or sell every ~300 ms. Every block posts a real post-only limit order on that side, one tick inside the touch, replacing the last one. Fills happen when a taker hits it, so the bot earns the spread instead of paying it. A small server streams every block to the dashboard.

项目数据：⭐ 864 Stars，🍴 162 Forks

项目简介：text-humanizer is an open-source project designed to convert text generated by AI to its “humanized” version. Bypasses the most of AI detectors such as Turnitin or GPTZero

技术栈：Python、ai-humanization、gptzero、gptzero-bypass、text-humanizer、turnitin、turnitin-bypass

核心介绍：English | 中文 text-Humanizer is a completely free open-sourced project designed to humanize AI-generated text through a multilingual LLM-powered rewriting pipeline. By leveraging large language models for semantic-preserving text transformation, the system rewrites content with more natural sentence structures,…

LLM assistants are widely used for daily social advice, yet evaluating their social reasoning in such consultation settings remains challenging since (i) it requires setups where the assistant learns about social situations from subjective user narratives, and (ii) social properties, such as others' intentions, typically lack verifiable ground truth. To address these challenges, we introduce Fuse, a multi-agent simulation framework for studying user-mediated social reasoning. In Fuse, a target agent with a hidden motive interacts with other agents including one representing the user, who th…

Information retrieval is increasingly important as LLM agents tackle complex tasks involving diverse information needs. Because retrieval relies on an index that represents each document through index keys, retrieval quality depends heavily on how effectively these keys expose the knowledge contained in each document. However, effective index representations vary across retrieval environments, making it difficult for any fixed optimization strategy to perform consistently. Yet evolving an index to its retrieval environment remains largely human-driven, requiring humans to diagnose retrieval…

Test-time scaling can improve large language model reasoning by generating and combining multiple candidate responses. In sampling-based methods, the inference budget is often described by the number of generated candidates, N. However, N tells us how many candidates are generated, not how they are executed. The same candidate budget can be produced in one batched generation call or split across several sequential calls with smaller batch sizes. We first study the effect of increasing N on reasoning accuracy using Phi-3-mini and Qwen2.5-1.5B on 500 GSM8K prompts. As expected, increasing N f…

Question answering has advanced rapidly with large language models, but predominantly for high-resource languages, in both text and spoken settings. Spoken question answering (SQA) benchmark for Telugu remains unexplored, and the reliability of automatic evaluation in this setting remains unquantified. We introduce VākQA, a Telugu SQA benchmark of 2,001 factoid question-answer pairs across six domains, with 2.53 hours of speech audio, bilingual transcriptions, and human-verified reference answers. We first validate evaluation methods against human judgements: Gemini-as-a-judge best approxim…

Agent trajectories record what an agent does and what happens next. Yet standard supervised fine-tuning (SFT) applies loss only to agent-authored action tokens, using environment observations as context but not as prediction targets. We ask whether this convention provides the best initialization for subsequent reinforcement learning. We introduce ActObs, which also supervises the observation tokens already present in each trajectory. Although deployed agents never generate observations, learning to predict them encourages the policy to model action consequences without adding data, paramet…

We present Srijika, a system for producing installable OpenType fonts for nine Brahmic scripts: Devanagari, Tamil, Bengali, Telugu, Kannada, Malayalam, Gujarati, Gurmukhi, and Odia. Rather than generating fonts from scratch, Srijika restyles glyph outlines from shaping-complete template fonts. It preserves the template's cmap and GSUB closure and its GPOS data under a documented metric policy, making every output a complete font by construction. This addresses a central challenge of Indic font generation: hundreds to thousands of conjuncts, half forms, and matra variants must remain mutuall…