每日研究简报 2026-09-09

技术人视角 · 今日四栏精选：arXiv 论文 / GitHub 开源 / HuggingFace 热门 / 行业资讯。

一 · arXiv 最新论文

摘要：We study the problem of navigating cluttered indoor environments with a humanoid robot. Unlike conventional methods that model navigation as a 2D path planning problem, humanoid traversal in cluttered environments requires continuous geometry-aware whole-body adaptation, including coordinated arm placement, torso adjustment, and gait modulation for collision-free movement through complex 3D spaces
领域：AI / 大模型
推荐理由：近期提交，偏开发者/研究视角，值得速览。
链接：http://arxiv.org/abs/2609.09158v1

Learning Length-Extrapolatable Recurrent Models

摘要：Recurrent models provide a natural path to long-context modeling, yet models trained with backpropagation through time (BPTT) often fail beyond their training horizon. Classical analyses emphasize gradients that vanish or explode along temporal paths. However, dense per-token losses can still train a shared recurrent rule despite severe decay, showing that decay alone does not determine whether le
领域：AI / 大模型
推荐理由：近期提交，偏开发者/研究视角，值得速览。
链接：http://arxiv.org/abs/2609.09157v1

ReCite: Agentic Reasoning for Faithful Citation

摘要：Accurate citations are the foundation of academic writing, tracing intellectual origins and substantiating core claims. However, manually navigating the growing volume of scientific literature is increasingly difficult, prompting reliance on automatic citation recommendation. While modern retrieval-augmented architectures have largely mitigated the fabrication of non-existent papers, current syste
领域：AI / 大模型
推荐理由：近期提交，偏开发者/研究视角，值得速览。
链接：http://arxiv.org/abs/2609.09156v1

SyncWorld: Visual Calibration Enables World Models as Zero-Shot Simulators

摘要：World models are increasingly used as policy-in-the-loop imagination environments, where reliable rollouts require fine-grained controllability with respect to low-level robot actions. A key obstacle to scaling such models in robotics is that actions are not a universal language in pixel space: changes in visual environment, camera view, robot placement, or embodiment alter how the same numerical
领域：AI / 大模型
推荐理由：近期提交，偏开发者/研究视角，值得速览。
链接：http://arxiv.org/abs/2609.09155v1

Procedural Graphs: Self-Evolving Execution Structures for LLM Agents

摘要：Large language models are increasingly deployed as agents that plan over long horizons and act through external tools. Most agents select actions through unconstrained generation over an accumulating history, leaving implicit the procedural knowledge of what to do, in what order, and under which conditions. As trajectories lengthen, agents can lose track of their objectives, invoke tools out of or
领域：AI / 大模型
推荐理由：近期提交，偏开发者/研究视角，值得速览。
链接：http://arxiv.org/abs/2609.09153v1

Silver Rate Is (Almost) Optimal for Gradient Descent Acceleration

摘要：We study how far gradient descent (GD) can be accelerated by predetermined nonnegative stepsizes in smooth convex optimization. Writing $p_{\mathrm{sil}}=\log_2(1+\sqrt{2})$, we prove an $Ω\left(n^{-p_{\mathrm{sil}}-O(\sqrt{\log\log n/\log n})}\right)$ non-anytime lower bound. In the anytime setting, every infinite nonnegative schedule has infinitely many horizons with error $Ω\left(n^{-\frac{2p_{
领域：AI / 大模型
推荐理由：近期提交，偏开发者/研究视角，值得速览。
链接：http://arxiv.org/abs/2609.09152v1

Copying explains the collective behavior of AI agents in the wild

摘要：In June 2026, thousands of AI agents found that a small public wiki would accept edits from inside their sandboxes, and started using it to help one another pass a timed test. Each agent lived for about an hour and remembered nothing afterwards. Nobody asked them to cooperate, and the wiki had not been built for them. The complete record of what they wrote is public, and it is unusually informativ
领域：AI / 大模型
推荐理由：近期提交，偏开发者/研究视角，值得速览。
链接：http://arxiv.org/abs/2609.09150v1

Point4D: Long-range 4D Motion Reconstruction

摘要：We introduce Point4D, a feed-forward model for 4D reconstruction of long-range video sequences. Point4D is able to reliably infer dense per-point 3D trajectories across multi-hundred-frame videos, unlike existing 4D methods that are limited to short input windows of at most a few dozen frames. A key innovation that enables this is our flexible 3D query-based motion decoder that decouples trajector
领域：AI / 大模型
推荐理由：近期提交，偏开发者/研究视角，值得速览。
链接：http://arxiv.org/abs/2609.09145v1

二 · GitHub 热门开源

openclaw/openclaw

简介：The AI that really does things. Any OS. Any Platform. The lobster way. 🦞
热度：389273⭐
推荐理由：近期活跃且星标领先，值得关注。
链接：https://github.com/openclaw/openclaw

obra/superpowers

简介：An agentic skills framework & software development methodology that works.
热度：283702⭐
推荐理由：近期活跃且星标领先，值得关注。
链接：https://github.com/obra/superpowers

NousResearch/hermes-agent

简介：The agent that grows with you
热度：243656⭐
推荐理由：近期活跃且星标领先，值得关注。
链接：https://github.com/NousResearch/hermes-agent

n8n-io/n8n

简介：Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
热度：203816⭐
推荐理由：近期活跃且星标领先，值得关注。
链接：https://github.com/n8n-io/n8n

Significant-Gravitas/AutoGPT

简介：AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
热度：187220⭐
推荐理由：近期活跃且星标领先，值得关注。
链接：https://github.com/Significant-Gravitas/AutoGPT

firecrawl/firecrawl

简介：The context API to search, scrape, and interact with the web at scale. 🔥
热度：178186⭐
推荐理由：近期活跃且星标领先，值得关注。
链接：https://github.com/firecrawl/firecrawl

f/prompts.chat

简介：f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
热度：169761⭐
推荐理由：近期活跃且星标领先，值得关注。
链接：https://github.com/f/prompts.chat

Snailclimb/JavaGuide

简介：Java 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
热度：158398⭐
推荐理由：近期活跃且星标领先，值得关注。
链接：https://github.com/Snailclimb/JavaGuide

三 · HuggingFace 热门

📄 Daily Papers 精选

NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness

摘要：Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabilities and converts that evidence into the next round of learning. We present NeoHorse-1, a family of agent-native models developed to explore this path through agentic post-training. Our sys
热度：193⬆
GitHub：https://github.com/TokenRhythm/NeoHorse
链接：https://huggingface.co/papers/2609.08183

AuK Technical Report: An Open-Source Foundational Model for Speech Generation and Editing

摘要：We introduce AuK, an open-source foundational model that unifies speech generation and editing through a common interface of natural-language instructions and audio context. To support this broad capability set, we construct approximately 3.03 billion instruction–audio instances and 1.95 million ho
热度：114⬆
GitHub：https://github.com/Tencent-Hunyuan/AuK
链接：https://huggingface.co/papers/2609.08936

Omni Interaction Agent Technical Report

摘要：In this work, we present Gander, an end-to-end model that unifies omni perception, realtime interaction, and agentic capabilities within a single framework. In contrast to turn-based conventional paradigms, Gander continuously receives streaming inputs across multiple modalities, including video, sp
热度：80⬆
GitHub：https://github.com/Omni-Interaction-Gander/Omni-Interaction-Agent
链接：https://huggingface.co/papers/2609.08977

Eliciting Weak-to-Strong Generalization with On-Policy Reverse Distillation

摘要：Weak-to-strong generalization asks whether stronger models can learn from weaker supervisors and surpass them. This question is particularly important for successive model generations and multi-domain consolidation, where repeating frontier-scale post-training from scratch can be prohibitively expen
热度：69⬆
GitHub：https://github.com/raymin0223/on_policy_reverse_distillation
链接：https://huggingface.co/papers/2609.08798

DriveZero: End-to-End Driving Beyond Human Demonstrations

摘要：Most end-to-end autonomous-driving systems learn by imitating human driving logs, leaving their learned behavior constrained by the quality and behavioral coverage of the recorded trajectories. This report presents DriveZero, an end-to-end system that learns driving behavior beyond human demonstrati
热度：50⬆
GitHub：https://github.com/XiaomiAutoL3/DriveZero
链接：https://huggingface.co/papers/2609.06055

📝 HuggingFace Blog

Safety for Whom? Refusing the Right Subset of a Topic, Not the Whole Topic

摘要：Safety for Whom? Refusing the Right Subset of a Topic, Not the Whole Topic
链接：https://huggingface.co/blog/MultiverseComputingCAI/safety-for-whom

NeoMME: an efficient Multimodal-native and Multilingual Encoder

摘要：NeoMME: an efficient Multimodal-native and Multilingual Encoder
链接：https://huggingface.co/blog/Hcompany/neomme

Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO Steps

摘要：Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO Steps
链接：https://huggingface.co/blog/grpo-with-trl-ifstruct

四 · 行业资讯

具身机器人能搞定超市盘点吗？全球七万门店正在给出答案

内容：从Demo到货架，这两家公司要让具身智能算得过账
推荐理由：来自聚合源，偏产业动态。
来源：量子位

7 篇 ECCV 论文！极佳视界联合顶尖高校，打通空间智能从「看得稳」到「摸得准」再到「决策灵」的落地瓶颈

内容：<img data-src=“https://static.leiphone.com/uploads/new/images/20260909/6aa0c48aa4a68.jpg" class=“rich_pages wxw-img” data-ratio=“0.55” data-s=“300,640” data-type=“jpeg” data-w=“1000” style=“width:100%;display:inline-block;text-align:center;ba
推荐理由：来自聚合源，偏产业动态。
来源：雷锋网 AI

On the Navier–Stokes Millennium Prize Problem

内容：We’re sharing an AI-generated solution to the Navier–Stokes Millennium Prize Problem, including a writeup and a formal proof in Lean.
推荐理由：来自聚合源，偏产业动态。
来源：OpenAI Blog

ECCV 2026 专访：让大模型「忘掉XYZ」，RoboTracer 用 3D 空间感知与度量推理重塑机器人轨迹追踪

内容：<img data-src=“https://static.leiphone.com/uploads/new/images/20260909/6aa0c4e718b71.jpg" class=“rich_pages wxw-img” data-ratio=“0.55” data-s=“300,640” data-type=“jpeg” data-w=“1000” style=“width:100%;display:inline-block;text-align:center;ba
推荐理由：来自聚合源，偏产业动态。
来源：雷锋网 AI

腾讯混元、清华、南洋理工联手，「以小博大」破解空间智能算力与记忆断裂难题 | ECCV 2026

内容：<img class=“rich_pages wxw-img” data-aistatus=“1” data-croporisrc=“https://mmbiz.qpic.cn/sz_mmbiz_jpg/XqAicMdcoiafN9rQFS8mhciaWg9MYXxANNAaZT9W9iaFIbKK8icHh0YJricw8ibF6Hqqgcfz9nBicicI9QicL2COVgsXvxpTatMyOhuaBBLPmgP6372E8/0?wx_fmt=jpeg&from=app
推荐理由：来自聚合源，偏产业动态。
来源：雷锋网 AI

Meta debuts its Muse AI agent. Will consumers trust it?

内容：Meta’s new personal AI agent Muse wants access to users’ email, calendars, payments, health services, and more — making the company’s biggest consumer AI bet yet a major test of whether people still trust Meta with their data.
推荐理由：来自聚合源，偏产业动态。
来源：TechCrunch AI

百度搭子全面接入小度硬件，百度智能体进驻家庭空间

内容：9月8日，在北京举行的百度AI Day小度新品发布会上，小度宣布超能小度完成智能体化升级，并发布多款全新的家庭场景智能体应用。百度搭子作为小度智能体能力的底座，全面落地小度智能屏、闺蜜机、智能摄像机、智能音箱等硬件新品，推动智能体进驻家庭空间。此次升级后，超能小度不再止于接收和响应指令，而是能够自主拆解目标、统筹调度工具、闭环交付任务。面向家庭日程管理场景，家长通过微信一句话即可创建家庭日程与孩子作业，任务自动同步至智能屏并通知家庭成员；儿童陪学成长场景，则支持孩子在小度设备上完成音视频作业打卡提交、电子宠物互动、一句话生成应用等。智能体看护2.0同步升级，用户可直接说出看护需求，超能小度自动拆解为多维度看护任务，并匹配分级提醒机制，同时基于长周期数据生成习惯成长报告。此外，家庭智能体服务已覆盖儿童成长陪伴、家庭健康管理、生活服务等更多家庭场景。搭载智能体
推荐理由：来自聚合源，偏产业动态。
来源：雷锋网 AI

燧原科技发行结果出炉！募资61.19亿元，国产AI芯片龙头即将登陆科创板

内容：9月7日晚间，燧原科技（688801.SH）正式公布发行结果。本次发行价格为142.18元/股，发行数量为4303.5173万股，募集资金总额为61.19亿元。网上投资者认购数量为1031.1626万股，网下投资者认购数量为2409.9639万股。此前，燧原科技网上发行有效申购户数达703.20万户，最终中签率为0.02455315%，市场认购火热。燧原科技长期专注于云端AI芯片及相关产品研发，经过多年技术积累和产品迭代，已形成覆盖AI芯片、AI加速卡及模组、智算系统及集群以及AI计算与编程软件平台的完整产品体系。目前，公司已自主研发迭代四代架构、五款云端AI芯片，并围绕芯片、硬件、软件及系统持续构建全栈技术能力。技术研发是燧原科技持续成长的重要支撑。公司采用自主可控的DSA架构，围绕GCU-CARE计算加速单元、GCU-LARE芯片互联等核心技术持续迭代
推荐理由：来自聚合源，偏产业动态。
来源：雷锋网 AI

本次任务消耗Token统计：脚本化模式（opencode 启动，无独立 token 计量）