Weekly Digest — 2025-W42

Anthropic's Interactive Prompt Engineering Tutorial

Beta release of Archon OS - the knowledge and task management backbone for AI coding assistants.

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

Collection of extracted System Prompts from popular chatbots like ChatGPT, Claude & Gemini

Klavis AI (YC X25): MCP integration platforms that let AI agents use tools reliably at any scale

A collective list of free APIs

Node Version Manager - POSIX-compliant bash script to manage multiple active node.js versions

100+ open-source clones of popular sites like Airbnb, Amazon, Instagram, Netflix, Tiktok, Spotify, Whatsapp, Youtube etc. See source code, demo links, tech stack, github stars.

Agentic AI Framework for Java Developers

📚 从零开始的大语言模型原理与实践教程

🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!

Next Generation Server Toolkit. Create web servers with everything you need and deploy them wherever you prefer.

Large language models leverage internet-scale text data, yet embodied AI remains constrained by the prohibitive costs of physical trajectory collection. Desktop environments -- particularly gaming -- offer a compelling alternative: they provide rich sensorimotor interactions at scale while maintaining the structured observation-action coupling essential for embodied learning. We present D2E (Desktop to Embodied AI), a framework that demonstrates desktop interactions can serve as an effective pretraining substrate for robotics embodied AI tasks. Unlike prior work that remained domain-specific (e.g., VPT for Minecraft) or kept data proprietary (e.g., SIMA), D2E establishes a complete pipeline from scalable desktop data collection to verified transfer in embodied domains. Our framework comprises three components: (1) the OWA Toolkit that unifies diverse desktop interactions into a standardized format with 152x compression, (2) the Generalist-IDM that achieves strong zero-shot generalization across unseen games through timestamp-based event prediction, enabling internet-scale pseudo-labeling, and (3) VAPT that transfers desktop-pretrained representations to physical manipulation and navigation. Using 1.3K+ hours of data (259 hours of human demonstrations, and 1K+ hours of pseudo-labeled gameplay), we achieve a total of 96.6% success rate on LIBERO manipulation and 83.3% on CANVAS navigation benchmarks. This validates that sensorimotor primitives in digital interactions exhibit sufficient invariance to transfer meaningfully to physical embodied tasks, establishing desktop pretraining as a practical paradigm for robotics. We will make all our work public, including the OWA toolkit, datasets of human-collected and pseudo-labeled, and VAPT-trained models available at https://worv-ai.github.io/d2e/

Camera-centric understanding and generation are two cornerstones of spatial intelligence, yet they are typically studied in isolation. We present Puffin, a unified camera-centric multimodal model that extends spatial awareness along the camera dimension. Puffin integrates language regression and diffusion-based generation to interpret and create scenes from arbitrary viewpoints. To bridge the modality gap between cameras and vision-language, we introduce a novel paradigm that treats camera as language, enabling thinking with camera. This guides the model to align spatially grounded visual cues with photographic terminology while reasoning across geometric context. Puffin is trained on Puffin-4M, a large-scale dataset of 4 million vision-language-camera triplets. We incorporate both global camera parameters and pixel-wise camera maps, yielding flexible and reliable spatial generation. Experiments demonstrate Puffin superior performance over specialized models for camera-centric generation and understanding. With instruction tuning, Puffin generalizes to diverse cross-view tasks such as spatial imagination, world exploration, and photography guidance. We will release the code, models, dataset pipeline, and benchmark to advance multimodal spatial intelligence research.

As the volume of peer-reviewed research surges, scholars increasingly rely on social platforms for discovery, while authors invest considerable effort in promoting their work to ensure visibility and citations. To streamline this process and reduce the reliance on human effort, we introduce Automatic Promotion (AutoPR), a novel task that transforms research papers into accurate, engaging, and timely public content. To enable rigorous evaluation, we release PRBench, a multimodal benchmark that links 512 peer-reviewed articles to high-quality promotional posts, assessing systems along three axes: Fidelity (accuracy and tone), Engagement (audience targeting and appeal), and Alignment (timing and channel optimization). We also introduce PRAgent, a multi-agent framework that automates AutoPR in three stages: content extraction with multimodal preparation, collaborative synthesis for polished outputs, and platform-specific adaptation to optimize norms, tone, and tagging for maximum reach. When compared to direct LLM pipelines on PRBench, PRAgent demonstrates substantial improvements, including a 604% increase in total watch time, a 438% rise in likes, and at least a 2.9x boost in overall engagement. Ablation studies show that platform modeling and targeted promotion contribute the most to these gains. Our results position AutoPR as a tractable, measurable research problem and provide a roadmap for scalable, impactful automated scholarly communication.

Recent diffusion models achieve the state-of-the-art performance in image generation, but often suffer from semantic inconsistencies or hallucinations. While various inference-time guidance methods can enhance generation, they often operate indirectly by relying on external signals or architectural modifications, which introduces additional computational overhead. In this paper, we propose Tangential Amplifying Guidance (TAG), a more efficient and direct guidance method that operates solely on trajectory signals without modifying the underlying diffusion model. TAG leverages an intermediate sample as a projection basis and amplifies the tangential components of the estimated scores with respect to this basis to correct the sampling trajectory. We formalize this guidance process by leveraging a first-order Taylor expansion, which demonstrates that amplifying the tangential component steers the state toward higher-probability regions, thereby reducing inconsistencies and enhancing sample quality. TAG is a plug-and-play, architecture-agnostic module that improves diffusion sampling fidelity with minimal computational addition, offering a new perspective on diffusion guidance.

Large Language Models (LLMs) have shown remarkable success, and their multimodal expansions (MLLMs) further unlock capabilities spanning images, videos, and other modalities beyond text. However, despite this shift, prompt optimization approaches, designed to reduce the burden of manual prompt crafting while maximizing performance, remain confined to text, ultimately limiting the full potential of MLLMs. Motivated by this gap, we introduce the new problem of multimodal prompt optimization, which expands the prior definition of prompt optimization to the multimodal space defined by the pairs of textual and non-textual prompts. To tackle this problem, we then propose the Multimodal Prompt Optimizer (MPO), a unified framework that not only performs the joint optimization of multimodal prompts through alignment-preserving updates but also guides the selection process of candidate prompts by leveraging earlier evaluations as priors in a Bayesian-based selection strategy. Through extensive experiments across diverse modalities that go beyond text, such as images, videos, and even molecules, we demonstrate that MPO outperforms leading text-only optimization methods, establishing multimodal prompt optimization as a crucial step to realizing the potential of MLLMs.

Large Language Models (LLMs) have achieved remarkable success through imitation learning on vast text corpora, but this paradigm creates a training-generation gap and limits robust reasoning. Reinforcement learning (RL) offers a more data-efficient solution capable of bridging this gap, yet its application has been constrained by a critical data bottleneck: existing RL datasets are orders of magnitude smaller and less diverse than web-scale pre-training corpora. To address this, we introduce the Webscale-RL pipeline, a scalable data engine that systematically converts large-scale pre-training documents into millions of diverse, verifiable question-answer pairs for RL. Using this pipeline, we construct the Webscale-RL dataset, containing 1.2 million examples across more than 9 domains. Our experiments show that the model trained on this dataset significantly outperforms continual pretraining and strong data refinement baselines across a suite of benchmarks. Notably, RL training with our dataset proves substantially more efficient, achieving the performance of continual pre-training with up to 100times fewer tokens. Our work presents a viable path toward scaling RL to pre-training levels, enabling more capable and efficient language models.

We propose QeRL, a Quantization-enhanced Reinforcement Learning framework for large language models (LLMs). While RL is essential for LLMs' reasoning capabilities, it is resource-intensive, requiring substantial GPU memory and long rollout durations. QeRL addresses these issues by combining NVFP4 quantization with Low-Rank Adaptation (LoRA), accelerating rollout phase of RL while reducing memory overhead. Beyond efficiency, our findings show that quantization noise increases policy entropy, enhancing exploration, and enabling the discovery of better strategies during RL. To further optimize exploration, QeRL introduces an Adaptive Quantization Noise (AQN) mechanism, which dynamically adjusts noise during training. Experiments demonstrate that QeRL delivers over 1.5 times speedup in the rollout phase. Moreover, this is the first framework to enable RL training of a 32B LLM on a single H100 80GB GPU, while delivering overall speedups for RL training. It also achieves faster reward growth and higher final accuracy than 16-bit LoRA and QLoRA, while matching the performance of full-parameter fine-tuning on mathematical benchmarks such as GSM8K (90.8%) and MATH 500 (77.4%) in the 7B model. These results establish QeRL as an efficient and effective framework for RL training in LLMs.

Latent generative modeling, where a pretrained autoencoder maps pixels into a latent space for the diffusion process, has become the standard strategy for Diffusion Transformers (DiT); however, the autoencoder component has barely evolved. Most DiTs continue to rely on the original VAE encoder, which introduces several limitations: outdated backbones that compromise architectural simplicity, low-dimensional latent spaces that restrict information capacity, and weak representations that result from purely reconstruction-based training and ultimately limit generative quality. In this work, we explore replacing the VAE with pretrained representation encoders (e.g., DINO, SigLIP, MAE) paired with trained decoders, forming what we term Representation Autoencoders (RAEs). These models provide both high-quality reconstructions and semantically rich latent spaces, while allowing for a scalable transformer-based architecture. Since these latent spaces are typically high-dimensional, a key challenge is enabling diffusion transformers to operate effectively within them. We analyze the sources of this difficulty, propose theoretically motivated solutions, and validate them empirically. Our approach achieves faster convergence without auxiliary representation alignment losses. Using a DiT variant equipped with a lightweight, wide DDT head, we achieve strong image generation results on ImageNet: 1.51 FID at 256x256 (no guidance) and 1.13 at both 256x256 and 512x512 (with guidance). RAE offers clear advantages and should be the new default for diffusion transformer training.

Recent advances in multimodal large language models (MLLMs) have demonstrated substantial potential in video understanding. However, existing benchmarks fail to comprehensively evaluate synergistic reasoning capabilities across audio and visual modalities, often neglecting either one of the modalities or integrating them in a logically inconsistent manner. To bridge this gap, we introduce OmniVideoBench, a large-scale and rigorously designed benchmark dedicated to assessing synergistic audio-visual understanding, with a strong emphasis on modality complementarity and logical consistency. Specifically, OmniVideoBench comprises 1000 high-quality question-answer(QA) pairs, each annotated with step-by-step reasoning traces, derived from 628 diverse videos ranging from several seconds to 30 minutes, and manually verified to guarantee complete correctness and uniqueness. Moreover, OmniVideoBench encompasses 13 carefully designed question types, covering temporal reasoning, spatial localization, counting, causal inference, summarization, and beyond, thereby capturing the essential challenges of video understanding. Evaluation of multiple MLLMs on OmniVideoBench reveals a pronounced gap between model performance and human reasoning, with open-source models lagging significantly behind their closed-source counterparts, underscoring the inherent difficulty of genuine audio-visual reasoning. We will release OmniVideoBench to foster the development of MLLMs with stronger and more generalizable reasoning capabilities.

Autoregressive (AR) models remain the standard for natural language generation but still suffer from high latency due to strictly sequential decoding. Recent diffusion-inspired approaches, such as LlaDA and Dream, mitigate this by generating in parallel, yet they suffer from two core limitations: information loss, as predictive distributions for non-finalized tokens are discarded at each step, and premature commitment, where local decisions are made without sufficient global coordination. We introduce Latent Refinement Decoding (LRD), a two-stage framework with Latent Refinement and a Predictive Feedback Loop. The first stage maintains masked positions as distributional mixtures of predicted tokens and the mask embedding, allowing the model to establish more globally consistent beliefs. The second stage progressively finalizes confident tokens while retaining uncertain ones for iterative feedback. KL-divergence dynamics provide a principled and reliable criterion for convergence and early stopping. Experiments across coding (HumanEval +6.3, MBPP +2.6) and reasoning (GSM8K +2.9, MATH500 +3.8) show that LRD improves accuracy while delivering speedups of up to 10.6x, making it a strong and versatile alternative for parallel sequence generation.

Reinforcement Learning with Verifiable Rewards (RLVR) has recently emerged as a promising framework for improving reasoning abilities in Large Language Models (LLMs). However, policy optimized with binary verification prone to overlook potential valuable exploration in reasoning trajectory. In view of heavy annotation cost of golden Process Reward Models (PRMs), recent works attempt using auxiliary signals for reward shaping of process tokens, involving entropy and likelihood collected from logit space. In this work, we offer a novel perspective on shaping RLVR with flow rewards derived from latent space, and propose RLFR, where the flow fields of model latents are constructed from either off-policy high-quality data and on-policy rejection sampling data, and the velocity deviations of policy latents within it are quantified to serve as a reward signal. RLFR first demonstrates that a well-established flow field can be a sound environment for reward signal collection, highlighting the expressive latent space is much underexplored. Moreover, RLFR is able to compress any off-policy expert data as reference for constituting reward signals, and we show that the efficient context dependence compressed within the hidden states are utilized, rather than individual token-level denotation for context comprehending. Experiments on both language and multimodal reasoning benchmarks demonstrate the reliability of flow rewards, and suggesting a promising paradigm for reward shaping with auxiliary signals.

While Reinforcement Learning with Verifiable Rewards (RLVR) has advanced the reasoning capabilities of Large Vision-Language Models (LVLMs), most existing methods in multimodal reasoning neglect the critical role of visual perception within the RLVR optimization process. In this paper, we undertake a pioneering exploration of multimodal RLVR through the novel perspective of token perception, which measures the visual dependency of each generated token. With a granular analysis of Chain-of-Thought (CoT) processes, we uncover two key insights: first, token perception in a rollout trajectory is sparsely distributed, where only a small fraction of tokens have high visual dependency for visually-grounded reasoning; second, different trajectories exhibit significant divergence in their overall visual dependency. Based on these observations, we propose Visually-Perceptive Policy Optimization (VPPO), a novel policy gradient algorithm that explicitly leverages token perception to refine the learning signal. Specifically, VPPO achieves this through a dual mechanism: it reweights a trajectory's advantage by its overall visual dependency, and focuses policy updates exclusively on perceptually pivotal tokens. On a comprehensive suite of eight perception and reasoning benchmarks, VPPO demonstrates substantial gains over leading open-source RL-tuned models, with its effectiveness consistently validated across 7B and 32B model scales. Our findings not only establish a new token-level perceptual perspective for analyzing multimodal RLVR but also present a novel and effective optimization strategy to significantly enhance the multimodal reasoning capabilities of LVLMs.

发表于《自然》的新研究显示，高龄父亲将致病突变遗传给孩子的风险比我们想象的要高。基因组测序显示，在 30 岁出头的男性中，大约每 50 个精子中就有 1 个携带致病突变；而到 70 岁时，这一比例上升到近 1/20。 研究人员建议，如果年轻男性认为自己要年纪大一些时再有孩子，他们可以考虑冷冻精子；而计划组建家庭的年长男性则可以考虑现有的各种筛查技术。最近的研究表明，我们每个人体内的大多数细胞中都有约 70 个父母都没有的新突变，其中 80% 的突变源于父亲的睾丸，这还不包括母亲卵子中更常见的大规模染色体异常。

法拉利宣布其首款电动跑车 Elettrica 将于明年夏天推出。Elettrica 的最高时速 310 公里/小时，百公里加速仅需 2.5 秒，续航里程 530 公里，最高 350 kW 的超快直流充电，电池容量 122 kWh，能量密度 195 Wh/kg——法拉利称这是量产电动汽车中最高的。电动汽车通常因为发动机过于安静而会去模拟机械发动机的轰鸣声，Elettrica 采用了不同的方法：安装在逆变器上的传感器会探测动力系统的真实机械振动，然后将其放大，创造出一种反映驾驶方式的不断变化的自然音调。法拉利称声音为司机提供了一种反馈功能，司机如果喜欢安静驾驶可以选择将其关闭。

Firefox 多年来一直支持创建多个配置文件去存储个人信息，以便将工作与个人浏览分开、测试不同设置，或与他人共享计算机。但 Firefox 没有让配置文件更容易被发现或管理。现在情况即将发生改变，Mozilla 宣布将推出配置文件管理功能，用户能更轻松地创建和切换配置文件。该功能将于 10 月 14 日起逐步推广给用户。

根据发表在《Lancet Regional Health – Western Pacific》上的一项研究，研究人员分析了 2019-2020 年间斯里兰卡、印度尼西亚、马来西亚、越南和菲律宾十所医院收集的近 1.5 万份患病婴儿的血液样本，发现人类与耐药细菌的战争并不顺利，新生儿血液中的耐药菌（或称之为超级细菌）十分普遍。近八成新生儿感染的是革兰氏阴性菌如大肠杆菌（E. coli,）、克雷伯菌（Klebsiella）和不动杆菌（Acinetobacter）。革兰氏阴性菌因其细胞膜结构，比革兰氏阳性菌更容易产生抗生素耐药性。研究人员称，新生儿出生几天后就会感染耐药菌。研究还发现，真菌感染导致了近十分之一的婴儿严重感染。

一项研究发现，一颗自由漂浮的行星吞噬了数量惊人的物质——每秒可以吃掉 60 亿吨气体和尘埃。这一发现模糊了行星与恒星之间的界限，暗示着恒星和行星的形成过程比想象中更相似。流浪行星是一种不围绕任何母恒星上的自由漂浮的气体星球，它们极其常见，甚至可能超过银河系中的恒星数量。但流浪行星的形成方式令天文学家困惑不已：它们会像其他行星一样先是围绕恒星运行，然后被放逐后独自在银河系中漫游吗？亦或者它们可以像恒星一样自行形成？天文学家最近发现了一颗名为 Cha 1107-7626 的流浪天体正以惊人的井喷式速度增长。早在 2008 年，该天体因其周围形成了看起来像原始行星盘的物质，曾首次引起天文学家的注意。6 月 Cha 1107-7626 突然开始以之前近 10 倍的速度消耗物质，并持续了两个月。这达到了以往只有在恒星中才能看到的增长速度。研究团队认为，一定有一种类似于恒星中发现的机制在起作用，即强磁场将物质从远处的气体和尘埃体积中通过狭窄的通道输送。但目前尚不清楚这颗行星是如何或为什么突然开始消耗如此多的质量。

金星曾经被认为是一个十分干燥、富含硫酸大气的行星，美国科学家重新分析了先驱者金星计划留下的资料，发现金星大气不只是硫酸量比先前认为的少，还有比预期更多的水和氧化铁。先驱者金星2号任务搭载了一大三小共计 4 架金星大气层的探测器，让探测器在落下的过程持续收集金星大气的成分等等数据。在降落过程中探测器也同时收集到了金星大气中的气胶，而气胶在进入探测器后分解，成分也因此被探测器记录下来。然而这些资料一直被尘封在 NASA 档案馆里面，直到最近研究团队才在一组微缩胶卷上找到。重新分析发现了金星气胶粒子中含有水、二氧化硫、氧分子和氧化铁的证据。水的含量比先前预期地高，大约是过去估计的三倍──水占气胶质量的约 60%。

AMD 和英特尔去年 10 月宣布组建一个 x86 生态系统顾问团队，致力于在 x86 架构实现上有更高的一致性。x86 生态系统某种程度上由 AMD 和英特尔共同开发，但两家公司保持着距离，导致了部分指令集架构存在低效和偏差问题。高级矢量扩展指令集（Advanced Vector Extensions 或 AVX）就是一个典型例子，AVX-512 在多年里只能通过英特尔平台使用，AMD 是从 2022 年的 Zen 4 起加入了对 AVX-512 的初步支持，2024 年发布的 Zen 5 才完整支持 512 位数据路径。过去一年 x86 生态系统顾问团的成果包括：内存标记指令集架构 ChkTag，Flexible Return Event Delivery (FRED)，AVX 指令集的下一代 AVX10，用于矩阵乘法的 Advanced Matrix Extensions (AMX) ACE，等等。

美国的 AI 产业在蓬勃发展，但制造业则陷入了更深层的衰退。美国制造业在 1979 年的巅峰时期雇佣了约 1950 万名工人。这一数字此后已缩减至不到 1300 万，而截至今年八月的一年内，制造业又减少了约 7.8 万个工作岗位。普查数据也显示，新成立的制造商数量在减少。美国经济分析局的数据显示，截至 7 月的一年内工厂投资下降了约 6%，这是自 2021 年初以来的首次下降。特朗普的关税政策也导致制造商们的利润下降。制造业的低迷与 AI 投资的巨额增长形成了鲜明对比。AI 行业使用的硬件大部分都免增关税。2025 年上半年，美国数据中心投资同比增长近 37%，而同期工厂建设则下降了约 3%。美国对计算设备的投资同比增长逾 45%，而传统工业设备的支出几乎没有变化。

日本三重大学的研究团队发现，1982-2023 年的 42 年间，日本的“夏季时长”增加了约 3 周。“冬季时长”基本未变，春季与秋季则不断缩短。团队警示称：“全球变暖导致的海面水温上升是主要原因。若变暖趋势持续，‘长夏+长冬’的两季化现象将进一步加剧”。42 年间夏季起始日提前了约 12.6 天，结束日推迟了约 8.8 天，总时长增加了约 21.4 天。以 2023 年为例，日本夏季为 6 月11 日至 10 月 9日，共计 121 天。

教宗良十四世上周接见了出席第 39 届国际新闻机构协会大会的参与者，督促在充斥“垃圾”资讯和数字媒体的时代中培育“良知”和“批判性思维”。他说，“我们并非注定要生活在一个真实与虚构再也难以区分的世界里”，他引用汉娜‧鄂兰（Hannah Arendt）的一句名言：“极权统治的理想臣民不是坚定的纳粹份子或共产主义者，而是那些再也无法分辨事实与虚构孰真孰假的人。”教宗说：“算法以前所未有的规模和速度生成内容与数据，但谁在掌控它们？人工智能正在改变我们获取资讯与交流的方式，但又是谁在引导它们，以及出于什么样的目的？”教宗提醒：“我们必须保持警惕，确保科技不会取代人类，以及对资讯和算法的管理今天不会被少数人掌控。”

尽管美国公司如 OpenAI、Anthropic 和 Google 的大模型在基准测试上处于世界领先水平，但这些模型基本上都是私有不公开权重的，根据 Hugging Face 和 LMArena 的统计，中国公司 DeepSeek 和阿里巴巴发布了下载量最多的公开权重的开放模型。Meta 一度倡导开放模型，扎克伯格（Mark Zuckerberg）去年还表示如果 AI 公司能分享模型，世界将会从中受益。但此后 Meta 放慢了公开其最新模型的步伐，扎克伯格如今表示要将最好的模型留给自己。

10 月 14 日，微软终止了对 Windows 10 的主流支持，进入到了扩展支持阶段。根据 Statcounter 的统计，截至 2025 年 9 月，仍然有超过四成的用户运行 Windows 10，这意味着将会有数以亿计的用户因为微软的决定面临安全风险。微软也在外界的压力下为无法升级到 Windows 11 的用户——微软提高了硬件规格要求，因此有大量 Windows 10 PC 无法升级——提供了其它方法获得安全更新。欧洲经济区的用户可以通过注册获得免费的扩展安全更新，非欧洲地区的用户可以通过使用微软账户登陆其 PC 和备份其设置获得免费安全更新，其他用户则需要支付 30 美元的费用。