揭示低资源语言在向量空间中的几何结构，为多语言模型优化提供全新视角。
arXiv:2608.23358v1 Announce Type: new Abstract: The performance gap between low- and high-resource languages in LLMs is widely known, but it remains u…
内存压缩8倍还比FAISS快，免训练免调参的Rust向量索引，值得一试
Article URL: https://github.com/RyanCodrai/turbovec Comments URL: https://news.ycombinator.com/item?id=49349898 Points: 289 # Comments: 34
DynamoDB原生向量搜索上线，单毫秒级延迟+万亿规模，传统向量库要变天？
Originally published on Build With AWS . Subscribe for weekly AWS builds. On August 5, 2026, AWS made vector search generally available in Amazon Dyna…
别让长文本淹没重点，用OpenSearch为AI智能体打造选择性记忆，精准调用所需信息。
Imagine handing an LLM the full text of Anna Karenina and asking what Levin thinks about farming. The model has the information somewhere in its windo…
抖音多模态嵌入模型技术报告，揭秘工业级搜索推荐背后的向量表示学习。
arXiv:2608.02148v1 Announce Type: cross Abstract: Multimodal representation learning is a cornerstone of modern AI. By encoding multimodal queries and…
GitHub工程师详解如何用向量化把源码大小写折叠提速到内存极限，性能优化实战范例。
How a branch-free loop and byte-space arithmetic let GitHub case-fold every byte of code search at >45 GiB/s on a single core. The post Don’t st…
降维向量能否讲清大模型训练数据？这篇论文给出了实证对比，值得关注。
arXiv:2601.16651v3 Announce Type: replace Abstract: Gradient-based methods for instance-based explanation for large language models (LLMs) are hindere…
LuaJIT 新版本原生支持 SSE/AVX 向量指令，性能大幅提升，典型测试提速超 17 倍。
Article URL: https://github.com/TheLuaOSProject/LuaJITMT/releases/tag/1.0.0-simd Comments URL: https://news.ycombinator.com/item?id=49063010 Points: 1…
Guard Vector通过任务向量组合与流式感知前缀SFT，突破英语限制实现多语言LLM安全护栏，提供可扩展的防护新范式。
arXiv:2509.23381v2 Announce Type: replace Abstract: We introduce Guard Vector, a safety task vector computed as the parameter difference between a gua…
向量化Trie结构加速LLM约束解码，专为生成式检索设计，在GPU等加速器上实现高效推理。
arXiv:2602.22647v2 Announce Type: replace-cross Abstract: Generative retrieval has emerged as a powerful paradigm for LLM-based recommendation. Howeve…
一门将大语言模型融入多变量微积分课程的创新学习资源，互动讲解向量基础。
Article URL: https://calculus.academa.ai/ Comments URL: https://news.ycombinator.com/item?id=48964585 Points: 23 # Comments: 14
单一向量让大模型同时记住新旧答案，攻克时间事实编辑难题
arXiv:2607.11327v1 Announce Type: cross Abstract: Model editing keeps large language models (LLMs) up to date without retraining, but temporal facts e…
免API、免GPU，800行Python代码让任意LLM拥有随时间自动衰减的长期记忆
PVM: Give any LLM long-term memory. No API keys, no GPU, ~800 lines Python. Every LLM you've ever used is an amnesiac. ChatGPT forgets your name betwe…
基于谷歌开源LLM打造的专用向量化AI助手，论文详解模型设计与应用场景。
arXiv:2607.07846v1 Announce Type: new Abstract: VectorizationLLM is a specialized Large Language Model based on Google open-weight LLMs. The model is …
用任务向量让大模型生成既功能正确又安全无漏洞的代码，新方法直击LLM代码生成痛点
arXiv:2607.07881v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used for code generation, but they struggle to generat…
提出一种属性知识RAG模式，通过本地与生产环境的混合检索架构，让LLM在受控条件下选择字段，避免幻觉。
Article URL: https://superml.dev/attribute-knowledge-rag-pattern-llm-governed-attributes Comments URL: https://news.ycombinator.com/item?id=48810859 P…
Qdrant用户福音：类Alembic的声明式YAML迁移，一条命令搞定向量库schema版本管理
Hey HN! I'm Diego. I've been extensively using Qdrant for hybrid search, but there's no standard way to track and manage schema migrations. I've used …
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 …
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
用LLM聚类实现实时难负样本采样，直击双塔检索训练瓶颈，规模越大收益越明显。
arXiv:2607.00448v1 Announce Type: cross Abstract: The two-tower model has been widely used for large-scale recommendation systems, particularly in the…