LLM作为评判者自身结果不一致？揭秘模型版本漂移与提示模糊性的根源与解法
Same outputs, same judge, two runs, two scores. The gate flickered red then green on a branch with zero code changes, and that flapping cost me more t…
用Python脚本同时向5个AI检测器发送同一篇文章，82%与31%的巨大差异揭露检测工具的可信度危机。
Last semester I pasted the same essay into two detectors back to back and got 82% on one, 31% on the other. No edits. Same paste. Within the same hour…
别手动扒ChatGPT界面了！用官方API每月$0.05就能追踪产品被AI推荐的频率，轻松出周报趋势。
Ask ChatGPT or Perplexity "what's the best note-taking app" and you get a shortlist of three to five names. Either you're on it or you don't exist in …
UKRI资助的Metascience项目，系统对比GPT-4o、Mistral与自研算法在基金提案中提取研究实体的表现。
arXiv:2606.30304v1 Announce Type: cross Abstract: This paper presents preliminary findings from a UKRI-funded Metascience project comparing three LLM-…
硬编码系统提示的惨痛教训：一个小改动影响1.8亿用户，生产环境中的反模式需警惕。
A support engineer pasted the system prompt into Slack to ask why the bot had started agreeing with everything. Three people replied with different ve…
实测DeepSeek V4 Pro与GPT-4o在SQL优化、素数证明等硬核任务上的性能对决，谁更胜一筹？
DeepSeek V4 Pro vs GPT-4o: Real Benchmark Comparison (June 2026) I ran both models through 20 coding, math, and reasoning tests. Here are the raw numb…
颠覆传统计费模式：让用户直接为LLM API付费，开发者零成本集成。5分钟接入Claude Sonnet等模型。
Hey HN! I recently built halupedia - a funny site that hit 300k+ users in few weeks and #1 on hacker news. It was amazing project, which costed me $30…
多模态AI自我评估存在系统性偏差，一种新发现的“偏好崩溃”现象加剧了模型反馈循环中的错误。
arXiv:2606.16682v1 Announce Type: new Abstract: When AI agents use language models to evaluate their own outputs in a feedback loop, systematic biases…
用GitHub Actions和OpenAI GPT-4o自动化代码审查，教你配置工作流、获取PR diff并编写提示词，提升代码审查效率。
Manual code reviews are a bottleneck in fast-moving development teams. Automating initial feedback loops ensures consistency and allows human reviewer…
花了400美元才学会的GPT-4o API省钱血泪教训，3分钟掌握避坑方法
Last year I deployed a GPT-4o powered support chatbot for a small SaaS. Traffic was modest — maybe 500 active users. I checked the OpenAI bill at the …
AI解读泰米尔古老诗篇，轻松探索桑嘎姆文学精髓的搜索引擎
Article URL: http://sangam.tamilnlp.com/mp/json/ai.html Comments URL: https://news.ycombinator.com/item?id=48398995 Points: 1 # Comments: 0
从Audio-LLM token直接生成实时面部动画，让对话头像告别机械表情，更自然流畅。
arXiv:2605.31294v1 Announce Type: new Abstract: Recent advances in Audio-LLMs like GPT-4o have ushered in an era of conversational interaction with la…
一次改写质量审计揭露了数学推理基准中语义漏洞，排名翻盘只需移除3.1%错误项，单一模型评估的盲区由此暴露。
arXiv:2605.29001v1 Announce Type: cross Abstract: A paraphrase-quality audit of MathCheck (ICLR 2025) detected 4 semantically incorrect paraphrases in…
揭示GPT-4o mini安全过滤器的盲点：多模态仇恨言论检测中为何退化为单模态瓶颈
arXiv:2509.13608v2 Announce Type: replace Abstract: As Large Multimodal Models (LMMs) become integral to daily digital life, understanding their safet…
OpenAI官方发布GPT-4o系统卡，详解模型能力与安全评估
OpenAI官方揭秘如何用GPT-4o视觉微调打造更智能的地图应用。
Building smarter maps with GPT-4o vision fine-tuning
OpenAI官方宣布Operator底层模型升级为o3，API版本保持4o不变，关注新模型应用动态。
We are replacing the existing GPT-4o-based model for Operator with a version based on OpenAI o3. The API version will remain based on 4o.
看Mercari如何用GPT-4系列模型简化销售流程，智能提升商品列表，助力卖家销量增长。
Mercari leverages GPT-4o mini and GPT-4 to streamline selling, enhance product listings, and boost sales, transforming the online marketplace with fea…
GPT-4o赋能Altera，开创人类与AI协作新范式。
Altera uses GPT-4o to build a new area of human collaboration
OpenAI 回滚了GPT-4o的谄媚式更新，恢复更平衡的对话行为。
We have rolled back last week’s GPT‑4o update in ChatGPT so people are now using an earlier version with more balanced behavior. The update we removed…