2026年7月23日 技术热点总结

项目简介：Dotted thought-orb loading indicators for AI & agent UIs — six tuned states, two sizes, auto dark/light

技术栈：TypeScript

核心介绍：Dotted thought-orb loading indicators for AI & agent UIs. Six hand-tuned animated states, each shipped at two purpose-tuned sizes, rendered on a plain 2D canvas — no WebGL, no filters, works identically in Chrome, Safari and Firefox. Live demo · Repository · Report an issue

项目数据：⭐ 603 Stars，🍴 39 Forks

项目简介：一个先接住情绪、再分析关系并给出可执行策略的 Codex 恋爱军师，内置心理、法律、社会、人文、哲学、婚姻家庭与性学知识库，支持多元关系。

技术栈：Python、ai-agent、chinese、codex、codex-skill、lgbtq、psychology、relationship-advice

核心介绍：> 把情绪支持、关系科学与可执行策略，装进一个真正站在用户一边的 AI 恋爱军师。 如果它让你在感情里少一次内耗、多一次清醒，欢迎点亮一个🌟 Star。你的支持会让更多正在关系里犹豫的人找到它，也会推动知识库继续更新。 很多恋爱建议只有两个答案：“勇敢去追”或“赶紧分手”。狗头军师试图做得更认真一点：先接住用户的情绪，再把事实、猜测和未知拆开；既分析心动，也计算现实、互惠、风险、机会成本和长期选择权；最后给出可以真正执行的下一步。 它不只是一个话术库。它能读取聊天截图、导出文本和用户转述，区分素材来源与事实边界；也能把分析落到一句可发送的话、一次具体邀约、第一次见面安排或一段可复盘的对话演练。它是一套覆盖恋爱全周期、面向多元关系、能够解释建议理由的 Codex Skill。 狗头军师支持异性恋、男同性恋、女同性恋、双性恋、泛性恋、无性恋等不同性取向，也尊重跨性别、非二元性别及其他性别身份。它同样可以讨论单身约会、长期伴侣…

项目简介：AI video skill for Claude Code & Codex — cinematic product videos with Remotion: 106 shot recipe cards, 161 motion previews, a production-ready template

技术栈：TypeScript、agent-skills、ai-agents、ai-video、claude-code、claude-code-skills、claude-skills、codex、motion-design

核心介绍：English | 中文 motion-design studio: point it at your product and it storyboards, animates, and sound-designs a cinematic promo, marketing, launch, or demo video with Remotion — real page captures, 2.5D camera moves, beat-synced cuts, and film-grade SFX included. The 38-second Gallery intro below was itself p…

项目简介：Get notified when big trades happen on Hyperliquid.

技术栈：TypeScript

核心介绍：This bot monitors Hyperliquid trades and sends notifications to a Telegram channel when significant trades occur. Before you begin, make sure you have the following installed:

关键特性：Real-time trade monitoring on Hyperliquid。

项目数据：⭐ 503 Stars，🍴 0 Forks

We present AutoIndex, a framework for learning representation programs: executable transformations that map raw documents into the representations exposed to a retrieval system. Rather than tuning retrievers, rerankers, or a small set of preprocessing hyperparameters, AutoIndex searches over programs that slice, enrich, normalize, reweight, or reorganize documents before indexing. At each iteration, AutoIndex performs validation-guided program search, in which agents diagnose failures of the current program and synthesize candidate updates, retaining only updates that improve retrieval qual…

Diffusion Language Models (DLMs) have recently emerged as a promising alternative to autoregressive models. Unlike standard diffusion-based approaches, DLMs are not explicitly conditioned on a timestep, raising a natural question: do these models internally represent denoising progress, and how is such information used downstream? In this work, we show that DLMs do in fact encode a latent representation related to the diffusion timestep within their residual streams. We find that this signal can be reliably extracted using probes across layers, indicating that denoising progress is decodabl…

Modern ASR models trained on heterogeneously annotated data treat transcription style (verbatim vs. intended) as an uncontrolled latent variable, causing measurable decoding instability, evaluation confounding (up to 60% of reported WER attributable to style mismatch), and unreliable word-level timing. We show that models already encode both styles; the challenge is controlled activation. Using coverage-aware decoder task tokens trained on parallel verbatim/intended transcript pairs, we raise German disfluency F1 from 10% to 79% zero-shot, despite English-only training. Full English-only fi…

Multimodal humor in memes, cartoons, and comics remains difficult for AI systems because intended meaning depends on non-literal mechanisms, shared cultural knowledge, and communicative intent rather than literal scene description. This survey focuses on visual humor understanding in single-image and multi-panel artifacts, while treating humor generation as an emerging downstream frontier. We position the literature against prior humor, sarcasm, and general MLLM surveys and organize it using a capability-centric hierarchy spanning recognition, interpretation and reasoning, and generation. U…

Accurate agricultural field boundary delineation at large scale is a foundational task for food security, supply chain transparency, and carbon accounting. While vision foundation models like SAM show remarkable zero-shot capabilities, they frequently fail in geospatial domains due to topological complexity, cropland texturing patterns, and a lack of physical scale awareness. In this work, we introduce Delineate Anything v2, a globally scalable foundation model designed specifically for wide-area field boundary mapping. We construct FBIS-73M, a 73-million-instance multi-resolution dataset s…

Optimizer state is the largest single line item in the memory budget of mixture-of-experts (MoE) training: on a 6.78B-parameter MoE language model, AdamW keeps 50.6 GB of first and second moments to update 12.6 GB of bfloat16 weights. We study SkewAdam, an optimizer built on the observation that the three parameter populations of an MoE – the dense backbone, the experts, and the router – differ enough in size and gradient statistics that they should not receive the same state. SkewAdam keeps float32 momentum plus a factored second moment for the backbone (5% of parameters), a factored secon…