1
Show HN: Ty-extended – a Ty fork with plugin support
Ty语言分支，引入WASM插件机制，为语义检查器带来可扩展能力。
Hi HN! I'm a Ty frequent user on all my Python projects, however, lately I've been needing to go back to mypy due to the lack of plugin support on Ty.…
2
Production RAG at Scale: Lessons from Processing 10,000+ Listings Daily
每天处理10000+列表的RAG系统实战经验，揭示固定分块如何丢失语义边界，优化策略直击痛点。
I spent time debugging a RAG pipeline that looked correct in staging and behaved differently under real load. The job board I built processes thousand…
3
Code-MUE: Measuring Code LLMs' Uncertainty through Execution-based Semantic Interaction Graphs
论文提出通过执行语义交互图测量代码LLM的不确定性，为模型评估提供了新思路。
arXiv:2607.12273v1 Announce Type: cross Abstract: As Code Large Language Models (LLMs) become central to modern software engineering, their inherent s…
4
Comparing Semantic Navigation in Humans and Large Language Models using Natural Language Processing
用NLP对比人类与LLM的语义导航能力，揭秘AI认知机制的新视角。
arXiv:2607.12195v1 Announce Type: new Abstract: Semantic memory retrieval can be conceptualized as navigation through conceptual space. We compared se…
5
Token-Flow Firewall: Semantic Runtime Auditing for Persistent AI Agents
提出Token流语义防火墙，为持久AI代理提供运行时审计，保障Agent安全。
arXiv:2607.08395v1 Announce Type: cross Abstract: Persistent AI agents extend large language models (LLMs) beyond single-turn interaction into long-li…
6
智能体互联网关键系统技术验证工程启动，致力于打通跨平台、跨语义交互
启动智能体互联网技术验证工程，打通跨平台跨语义交互，解决异构智能体调度难题。
IT之家 7 月 9 日消息，7 月 8 日，由中国互联网协会、中国信通院联合主办的“中国互联网大会 · 智能体互联网论坛”在北京召开。会上，启动了智能体互联网关键系统技术验证工程。 据介绍，智能体互联网是以智能体为主体， 通过统一标准协议规则 ，实现智能体自主通信、交互协作和代理服务的新一代互联网…
7
LLM-Guided Task-Semantic Field Factorization for Industrial Process Forecasting
用大模型引导任务语义场分解，革新工业过程预测的精度与可解释性
arXiv:2607.06623v1 Announce Type: cross Abstract: Process industries rely on time-series forecasting and soft sensing to estimate quality variables th…
8
InfluMatch: Frontier-Quality KOL Search at 4B-Model Cost
只需4B参数模型，就能实现前沿KOL搜索的语义匹配，大幅降低成本而不牺牲精度。
arXiv:2607.05968v1 Announce Type: cross Abstract: Matching influencers (KOLs) to free-form, multi-part Thai marketing criteria is today served either …
9
Segmentation before Answering: Pixel Grounding for MLLM Visual Reasoning
提出「先分割再回答」新范式，让多模态大模型在视觉推理中实现像素级精准定位
arXiv:2607.05798v1 Announce Type: cross Abstract: Recent advancements in Multimodal Large Language Models (MLLMs) have evolved from static perception …
10
Risk-Constrained Freshness-Aware Semantic Caching for Open-Web Retrieval-Augmented LLMs
创新语义缓存方法，兼顾风险与新鲜度，提升开放网络RAG系统的效率与可靠性
arXiv:2607.04281v1 Announce Type: cross Abstract: Semantic caching reduces the latency and cost of retrieval-augmented generation (RAG) by serving cac…
11
Catching AI Red-Handed in Financial Data
精准捕捉金融数据中AI的幻觉，用向量数据库对比和LLM验证双把关，让财务审计不再被语义相似误导
When I was building security auditing tools like Git Secret Scanner, the rules were binary: a vulnerability exists, or it doesn't. But when you start …
12
Fidx – local semantic search in one SQLite file, no LLM at query
用SQLite实现本地语义搜索，查询时无需大模型，轻量高效的开源方案。
Article URL: https://github.com/williamliu-ai/fidx Comments URL: https://news.ycombinator.com/item?id=48791740 Points: 1 # Comments: 0
13
Evergreen: Efficient Claim Verification for Semantic Aggregates
提出高效声明验证方法Evergreen，在保持高F1值的同时将成本降低3.1倍，占据成本质量帕累托前沿。
arXiv:2604.26180v2 Announce Type: replace-cross Abstract: With recent semantic query processing engines, semantic aggregation has become a primitive o…
14
ExPerT: Personalizing LLM Responses to Users' Domain Expertise via Query-Wise Semantic and Keystroke Behavioral Cues
利用用户击键行为与语义线索，实现大模型对用户领域专业度的个性化响应
arXiv:2607.01242v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used by end users, yet existing personalization method…
15
Show HN: PurRDF-High performance, cross platform RDF1.2
高性能跨平台 RDF 1.2 工具包，支持 SPARQL、SHACL、ShEx 并通过 W3C 测试套件，语义网开发利器
Article URL: https://github.com/Blackcat-Informatics/purrdf/ Comments URL: https://news.ycombinator.com/item?id=48766069 Points: 1 # Comments: 1
16
A Hybrid Framework for Song Lyric Annotation Based on Human-LLM Alignment
人类智慧与AI大模型协同，突破歌词标注的语义理解瓶颈
arXiv:2606.29273v1 Announce Type: new Abstract: Emotion recognition of song lyrics is a challenging task since lyrics may not necessarily align with t…
17
Metaphors are a Source of Cross-Domain Misalignment of Large Reasoning Models
大型推理模型在跨域隐喻理解上存在对齐偏差，揭示隐喻是模型跨域误对齐的关键来源，为提升AI语义鲁棒性提供新视角。
arXiv:2601.03388v3 Announce Type: replace-cross Abstract: Earlier research has shown that metaphors influence human decision-making, raising the quest…
18
OracleAnalyser: Analysing Implicit Semantics of Oracle Bone Scripts through MLLMs with Post-training
首次将多模态大模型用于甲骨文语义分析，突破传统文字识别局限，开启古文字研究新范式。
arXiv:2606.25906v1 Announce Type: new Abstract: With the advancement of artificial intelligence, research on oracle bone scripts has entered a new era…
19
Real-Time Voice AI Hears but Does Not Listen
评估四大主流实时语音AI在“言外之意”上的表现，揭示模型“听得到词却听不懂调”的盲点
arXiv:2606.26083v1 Announce Type: new Abstract: Speech conveys information through both words and vocal delivery. We evaluate four leading production …
20
SemChunk-C: Semantic Segmentation for C Code
提出基于LLM的C代码语义分块方法，定义分块类别并训练模型识别边界与功能属性，值得技术研究者关注。
arXiv:2606.23697v1 Announce Type: cross Abstract: Semantic segmentation of code written in a C-family language remains a challenging problem, due to t…