Enhancing LLM-Based Test Generation by Eliminating Covered Code
阅读文章 ARXIV
ArXiv
大语言模型
CHRONOLOGICAL LEDGER
按时间浏览
2026 年
二月
50 ENTRIESCxMP: A Linguistic Minimal-Pair Benchmark for Evaluating Constructional Understanding in Language Models
阅读文章 ARXIV
ArXiv
大语言模型
Efficiently serve dozens of fine-tuned models with vLLM on Amazon SageMaker AI and Amazon Bedrock | Amazon Web Services
阅读文章 BLOGS_PODCASTS
博客与播客
Why Pass@k Optimization Can Degrade Pass@1: Prompt Interference in LLM Post-training
阅读文章 ARXIV
ArXiv
大语言模型
The First Fully General Computer Action Model
阅读文章 HACKER_NEWS
Hacker News
On Data Engineering for Scaling LLM Terminal Capabilities
阅读文章 ARXIV
ArXiv
AI Agent
Learning from Trials and Errors: Reflective Test-Time Planning for Embodied LLMs
阅读文章 ARXIV
ArXiv
AI Agent
Aletheia tackles FirstProof autonomously
阅读文章 ARXIV
ArXiv
AI Agent
PA bench: Evaluating web agents on real world personal assistant workflows
阅读文章 HACKER_NEWS
Hacker News
AI Agent
Making MCP cheaper via CLI
阅读文章 HACKER_NEWS
Hacker News
MCP
Learnings from 4 months of Image-Video VAE experiments
阅读文章 HACKER_NEWS
Hacker News
Large-Scale Online Deanonymization with LLMs
阅读文章 HACKER_NEWS
Hacker News
大语言模型
Building intelligent event agents using Amazon Bedrock AgentCore and Amazon Bedrock Knowledge Bases | Amazon Web Services
阅读文章 BLOGS_PODCASTS
博客与播客
RAG
AI to help researchers see the bigger picture in cell biology
阅读文章 BLOGS_PODCASTS
博客与播客
机器学习
Show HN: Sgai – Goal-driven multi-agent software dev (GOAL.md → working code)
阅读文章 HACKER_NEWS
Hacker News
AI Agent
Disrupting malicious uses of AI
阅读文章 BLOGS_PODCASTS
博客与播客
Launch HN: TeamOut (YC W22) – AI agent for planning company retreats
阅读文章 HACKER_NEWS
Hacker News
AI Agent
Claude Code 构建完全指南：十大核心功能深度解析
阅读文章 JUEJIN
掘金
MCP
AI Agent框架探秘：拆解 OpenHands（8）--- CodeActAgent
阅读文章 JUEJIN
掘金
MCP
Show HN: A real-time strategy game that AI agents can play
阅读文章 HACKER_NEWS
Hacker News
AI Agent
Claude Code Remote Control
阅读文章 HACKER_NEWS
Hacker News
Sovereignty in a System Prompt
阅读文章 HACKER_NEWS
Hacker News
Show HN: Context Mode – 315 KB of MCP output becomes 5.4 KB in Claude Code
阅读文章 HACKER_NEWS
Hacker News
MCP
Tool Building as a Path to "Superintelligence"
阅读文章 ARXIV
ArXiv
大语言模型
Motivation is Something You Need
阅读文章 ARXIV
ArXiv
Beyond the Star Rating: A Scalable Framework for Aspect-Based Sentiment Analysis Using LLMs and Text Classification
阅读文章 ARXIV
ArXiv
大语言模型
Anthropic drops flagship safety pledge
阅读文章 HACKER_NEWS
Hacker News
An Expert Schema for Evaluating Large Language Model Errors in Scholarly Question-Answering Systems
阅读文章 ARXIV
ArXiv
大语言模型
Mercury 2: Fast reasoning LLM powered by diffusion
阅读文章 HACKER_NEWS
Hacker News
大语言模型
什么？oh-my-opencode 太重了？那试试 oh-my-opencode-slim
阅读文章 JUEJIN
掘金
AI Agent
Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks
阅读文章 ARXIV
ArXiv
AI Agent
Show HN: Moonshine Open-Weights STT models – higher accuracy than WhisperLargev3
阅读文章 HACKER_NEWS
Hacker News
RCCLX: Innovating GPU Communications on AMD Platforms
阅读文章 BLOGS_PODCASTS
博客与播客
大语言模型
KNIGHT: Knowledge Graph-Driven Multiple-Choice Question Generation with Adaptive Hardness Calibration
阅读文章 ARXIV
ArXiv
RAG
How we rebuilt Next.js with AI in one week
阅读文章 HACKER_NEWS
Hacker News
AI 视觉连载5：传统 CV 之均值滤波
阅读文章 JUEJIN
掘金
深度学习