1
消息称预测市场 Polymarket 游说欧洲各国监管机构，意图摘掉博彩帽子
IT之家 9 月 22 日消息，据《金融时报》报道，预测市场 Polymarket 近期一直在和欧盟和英国监管机构沟通， 希望对方将其认定为金融服务机构 ， 而非博彩网站 。 IT之家注：预测市场指的是允许押注现实世界事件结果的平台， 例如人们可以在 Polymarket 上预测 OpenAI 今年…
2
When Do LLMs Replace Fine-Tuned NLU? A Decision Framework for Intent Detection in Production Conversational Systems
零样本大模型能否替代精调NLU？这项研究用ATIS和CLINC150实测，给出基于意图空间的决策框架，生产环境选型必看。
arXiv:2608.20371v1 Announce Type: cross Abstract: A common claim is that zero-shot large language models (LLMs) can replace fine-tuned NLU classifiers…
3
英伟达砸 60 美元投资 Poolside，意图打造中国开源 AI 竞争对手
英伟达豪掷60亿美元入股Poolside，目标直指DeepSeek和Kimi，开源AI竞争格局再添变数。
IT之家 8 月 23 日消息，据《华尔街日报》今天报道，英伟达同意向人工智能初创企业 Poolside 支付 60 亿美元 （IT之家注：现汇率约合 404.58 亿元人民币） ，换取该公司的 AI 模型软件授权，同时将向该公司 109 名员工提供工作机会， 旨在开发一款能够与 DeepSeek …
4
How I Built Intent Detection for a Chat-Based AI Comic Tool
聊天式AI漫画工具，用智能意图识别兜底模糊输入，角色一致性靠状态管理，5次免费体验免绑卡。
Most AI comic generators are single-shot: prompt in, image out. That works fine until you need the same character to show up consistently across five …
5
6
Toward Meaningful Transparency for AI Chatbots: Disclosing Persuasive Intent Reduces Persuasion
内置说服意图披露功能的AI聊天机器人，主动告知劝说目的，有效提升对话透明度，减少诱导操控感。
arXiv:2608.11794v1 Announce Type: cross Abstract: The growing role of AI-generated content and AI-enabled systems in public communication has led regu…
7
Building Tactile UX: Honoring Intentional Design With Lottie
用Lottie打造触感体验，从细节动效见设计诚意。181帧挤压动画揭露微交互的用心。
When tasked with building a highly interactive, tactile web experience, the architecture must serve the art direction. In this article, Alexey Kopytin…
8
DevIntent: How Much Does LLM-Generated Code Violate Developer Intent?
LLM代码通过测试就够了吗？新基准揭示隐藏的意图违背问题。
arXiv:2608.07614v1 Announce Type: cross Abstract: Code generated by LLMs can violate a developer's implicit intentions when given an ambiguous prompt,…
9
Training-Free versus Training-Based Intent Classification in LLMs: Accuracy, Robustness, and Failure Modes
大模型意图分类新研究：对比零样本与微调路线，揭示精度、鲁棒性与失败边界，值得关注。
arXiv:2608.02415v1 Announce Type: new Abstract: Intent classification in Large Language Models (LLMs) involves categorizing user prompts into predefin…
10
Show HN: Product analytics (and evals) for agent sessions on your MCP
用 MCP 深度分析 AI Agent 会话，自动识别用户意图和未被满足的需求，连 200 响应背后的隐性失败也无处遁形
Hi HN! We’re Theodore and Louis, founders of Armature (YC P26). We reconstruct the entire session behind the MCP tool calls you receive, including wha…
11
MIND: Multimodal Intent-Driven Network via Diffusion Transformers for Medical Image Fusion
医学影像融合迎来意图驱动新思路：扩散Transformer多模态网络，ACM MM 2026收录，技术细节扎实，值得算法研究者细读。
arXiv:2607.28565v1 Announce Type: new Abstract: Medical image fusion aims to integrate complementary information from diverse imaging modalities to su…
12
Neural Dynamics of AI-attributed Irony Reveal a Partial Intentional Stance
从神经科学角度揭示人类如何解读AI的讽刺意图，探索大语言模型幽默背后的认知机制。
arXiv:2510.17168v2 Announce Type: replace Abstract: As Large Language Models (LLMs) are increasingly deployed as social agents and trained to produce …
13
AI PR reviewer is looking at the diff, not the intent
AI代码审查虽能揪出语法错误，却致命地忽视了代码意图与业务背景
AI code review tools excel at analyzing syntax within a single file or a pull request. They can catch a missing semicolon a mile away. But they suffer…
14
LLMs Get Lost in Evolving User Intent
研究发现大型语言模型在用户意图动态演变时容易“迷失”，揭示了多轮对话中推理与记忆的脆弱性。
arXiv:2607.20734v1 Announce Type: new Abstract: As LLMs become more capable, they are increasingly deployed as collaborative agents, taking on user-de…
15
Robust Critics: Defending LLMs Against Multi-Turn Attacks
让大模型不再“被套路”：多轮攻击通过动态推断用户意图实现防御。
arXiv:2607.20472v1 Announce Type: new Abstract: When a user asks a language model something harmful, is it a genuine attack or a misunderstood but wel…
16
小米澎程新车布局示意图发布，雷军透露 N90 转弯半径很小
IT之家 7 月 23 日消息，小米创办人、董事长兼 CEO 雷军今日在微博分享了小米澎程新车的布局示意图。从画面可以看到，澎程布局有所优化， 三排座椅的空间相较传统 SUV 更方正、更宽敞 。 不少网友对澎程的轮拱去哪里感到好奇，雷军表示：“ 下了很大功夫，重新设计了技术架构。 ”还有网友询问澎程…
17
When JSON Is Not Enough: Semantic Reliability of Schema-Constrained LLM Ordering Agents
当JSON不够用时，这项研究揭示了LLM排序代理在模式约束下的语义可靠性问题，值得关注。
arXiv:2607.18261v1 Announce Type: new Abstract: LLM agents are increasingly used as transaction compilers: a user states an intent in natural language…
18
Vibecoding: When the AI writes the code and you manage the intent
AI时代编程新范式：用精确语言管理意图，让AI负责代码实现，程序员从语法细节转向消除歧义。
What is actually happening here For a decade, coding was about knowing where the semicolon goes and how the memory is allocated. Now, we are entering …
19
美初创公司 Reflect Orbital 获 FCC 批准发射轨道反光卫星，意图“按需”照亮地球
美初创公司获FCC批准，计划发射反光卫星实现按需照亮地球，科幻成真。
IT之家 7 月 13 日消息，美国加州初创公司 Reflect Orbital 日前获得了美国联邦通信委员会（FCC）的批准，可发射其首颗轨道反光卫星 Eärendil-1。 据IT之家了解，这家公司将自己定位为“阳光公司（The Sunlight Company）”，目标是让“清洁、充足的能源能…
20
Nika – Intent as Code for AI Workflows
一个文件、四个动词，像SQL一样定义AI工作流，让有用AI不再消失在聊天里。
Article URL: https://github.com/supernovae-st/nika Comments URL: https://news.ycombinator.com/item?id=48837692 Points: 1 # Comments: 0