1
天工团队回应机器人捂脸跑夺冠：没有找小姑娘让它学，是它自己想出来的
天工AI一站式智能助手，能聊天、能搜索、能写作，快速高效，是你数字生活的好帮手！
IT之家 8 月 24 日消息，第二届世界人形机器人运动会 8 月 22 日在国家速滑馆开幕，在 400 米（小型组）决赛中，以“捂脸跑”的跑姿完赛的天工在该项目中以 45.66 秒的成绩夺冠。 据红星资本局报道，赛后，天工 Omni 400 米小型组研发团队负责人、北京人形机器人创新中心运动控制算…
2
Benchmarking Patent Drafting from Inventor-Style Disclosures
科研新基准：从发明人式披露生成专利文本，直击大模型专业写作能力短板。
arXiv:2608.21249v1 Announce Type: new Abstract: While recent large language models (LLMs) have achieved promising results on individual patent draftin…
3
Forking Fast: Efficiently Estimating Uncertainty Dynamics in Text Generation
快速估计文本生成不确定性动态的新方法，让模型自我评估更高效精准。
arXiv:2608.19611v1 Announce Type: cross Abstract: LLM reasoning is stochastic, and so understanding a model requires grappling with the distribution o…
4
Can Language Models Imagine Without Seeing? Ekphrasis: Measuring Visual Creative Ideation in Text-Only LLMs
不看图也能想象？这项研究用新基准Ekphrasis测出纯文本大模型的视觉创意生成能力，揭示AI“脑内画图”的边界。
arXiv:2608.06967v1 Announce Type: new Abstract: Current evaluations do not isolate whether text-only language models can originate visual concepts bef…
5
RH-RAG: Trustworthy Long-Form Generation for Privacy-Constrained Settings
隐私受限环境下用RH-RAG实现可信长文本生成，直击RAG应用的核心痛点。
arXiv:2608.01311v1 Announce Type: new Abstract: Generating long-form content from extensive internal reports remains challenging for organizations ope…
6
On Improving Faithfulness of Podcasts from Documents
arXiv新研究聚焦文档转播客（Doc2Podcast）的忠实度问题，提出提升内容一致性的方法，减少AI生成幻觉。
arXiv:2607.21961v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to generate long-form conversational content such a…
7
Evaluating LLM Uncertainty in Long-Form Generation Using Deterministic Ground Truth
被ICML 2026接收，提出用确定性真实值评估大模型长文本生成的不确定性，为LLM可靠性研究提供新工具。
arXiv:2607.03870v1 Announce Type: new Abstract: As LLMs generate increasingly long outputs, effective uncertainty estimation must identify errors at f…
8
Text AI watermarks will always be trivial to remove
欧盟强制AI水印看似严格，但文本水印始终可被轻易剥离，技术漏洞比想象更致命。
Article URL: https://www.seangoedecke.com/text-ai-watermarks/ Comments URL: https://news.ycombinator.com/item?id=48757466 Points: 3 # Comments: 0
9
On AI Text Detection
AI文本检测与生成技术的猫鼠游戏，剖析检测工具的局限与背后博弈
Article URL: https://ethansmith2000.substack.com/p/ai-text-detection-arms-dealers-in Comments URL: https://news.ycombinator.com/item?id=48756891 Point…
10
Multi-Block Diffusion Language Models
多块扩散语言模型实现高质量文本生成，模块化设计提升可控性和多样性，适合AI研究与创作。
arXiv:2606.29215v1 Announce Type: new Abstract: Block Diffusion Language Models (BD-LMs) improve diffusion-based text generation with KV caching and f…
11
I Built a Free Cursive Text Generator — Here's How It Works
免费在线生成草书字体，无需注册秒速加载，适合社交媒体文案装饰。
Have you ever wanted to write something in cursive style for your Instagram bio, WhatsApp status, or just for fun — but didn't want to download any ap…
12
Improved Large Language Diffusion Models
改进大型语言扩散模型，高效生成高质量文本，值得所有AI开发者一读的新方法。
arXiv:2606.25331v1 Announce Type: cross Abstract: Modern large language models are predominantly trained with autoregressive factorization and causal …
13
Ensemble Learning for Large Language Models in Text and Code Generation: A Survey
权威综述，系统梳理LLM集成学习在文本与代码生成中的方法、挑战与未来方向。
arXiv:2503.13505v3 Announce Type: replace-cross Abstract: Generative Pretrained Transformers (GPTs) are foundational Large Language Models (LLMs) for …
14
IV-CoT: Implicit Visual Chain-of-Thought for Structure-Aware Text-to-Image Generation
把隐式视觉思维链引入文生图，让模型按结构逻辑逐层生成，画面布局更稳更精确
arXiv:2606.24849v1 Announce Type: cross Abstract: Unified multi-modal large language models (MLLMs) have achieved strong text-to-image generation qual…
15
Retrieval-Augmented Anatomical Guidance for Text-to-CT Generation
输入文字即可生成CT影像，检索增强解剖知识让生成结果更贴合人体结构，医学影像AI的新进展。
arXiv:2603.08305v2 Announce Type: replace-cross Abstract: Text-conditioned generative models for volumetric medical imaging provide semantic control b…
16
Fine-Grained Uncertainty Quantification for Long-Form Language Model Outputs: A Comparative Study
长文本生成中，模型何时在“胡编”？这项细粒度不确定性量化比较研究，为幻觉检测提供新视角。
arXiv:2602.17431v2 Announce Type: replace-cross Abstract: Uncertainty quantification has emerged as an effective approach to closed-book hallucination…
17
Customizing Video Portraits via Identity-ActionDecoupling
如何让AI生成的视频既保留人物身份又精准控制表情动作？这篇论文提出身份-动作解耦新框架，解决视频定制核心难题。
arXiv:2606.22347v1 Announce Type: new Abstract: Identity-Preserving Text-to-Video Generation (IPT2V) seeks to synthesize a temporally coherent video f…
18
StegoStylo: Squelching Stylometric Scrutiny through Steganographic Stitching
用隐写术对抗文体指纹，StegoStylo让AI写作身份无从追踪。
arXiv:2601.09056v5 Announce Type: replace-cross Abstract: Stylometry -- the identification of an author through analysis of a text's style (i.e., auth…
19
Exposing the Unsaid: Visualizing Hidden LLM Bias through Stochastic Path Aggregation
用随机路径聚合可视化LLM生成中的隐藏偏见，揭秘文本背后的系统性偏差。
arXiv:2606.19344v1 Announce Type: cross Abstract: Large Language Models (LLMs) exhibit representational and syntactic biases that are difficult to eva…
20
Sumi: Open Uniform Diffusion Language Model from Scratch
基于均匀扩散的语言模型，支持任意位置同时更新token，实现比自回归模型更灵活的文本生成
arXiv:2606.19005v1 Announce Type: cross Abstract: Diffusion models have become a promising alternative to autoregressive models. Among these, uniform …