뉴스해외IBM
실시간
국내외 뉴스 · 기업 공식 발표 · 논문 · 영상 · 커뮤니티 소식을 한국시간 기준으로 제공합니다. 같은 소식은 하나로 묶고, 기본은 핵심만 보여 드립니다.
모든 수집 소식(구글 뉴스로 들어온 단건 보도 포함).
다른 보도 WIRED
Learning accurate off-policy state-action value functions (Q-functions) from offline datasets becomes harder as the task horizon grows.
Denoising and score estimation are classically linked through Tweedie's formula, which relates the posterior mean under Gaussian noise to the Stein score of…
Weight tying reduces the parameter cost of language models, but the usual tied output head is limited to a single softmax.
Large language models can generate motion graphics from static images and natural language requests, but it remains difficult to evaluate how well models can…
Time-series classifiers deployed in risk-sensitive domains must be (1) accurate, (2) interpretable, and (3) able to abstain when evidence is weak.
Weight-activation quantization can reduce the memory and compute cost of neural-network inference, but recent methods often require expensive optimization and…
How little of a pretrained large language model (LLM) has to change for it to acquire new capabilities?
Diffusion language models (DLMs) generate text through iterative denoising, where decisions about which tokens to commit or remask strongly affect output…
Web-facing AI agents repeatedly access webpages, documents, repositories, and tool responses, yet reading interfaces often expose entire objects when only…
다른 보도 IMDb · Screen Daily
Security researchers found hundreds of servers hosting the penetration testing tool Artex exposed to the internet, allowing anyone access to the hacking tool.
Meta's Muse agents continue to get new features and integrations. Spaces promises to be a new way to interact with the AI helpers.
An experiment in which AI models were asked to break into simulated networks showed the hacking skills of DeepSeek and GLM to be not far behind OpenAI’s most…
목요일 급락장에서 AI 인프라 공급업체의 낙폭이 대형 클라우드 기업보다 컸으며, 나스닥 시가총액은 약 2조 5,000억 달러 줄어듦대형 클라우드 기업은 기존 사업의 막대한 매출을 바탕으로 지출을 조정할 여지가 있지만, 칩과 메모리, 서버, 전력 장비 공급업체는 데이터센터 건설 둔화가…
These are the first 20 pictures. The rest will follow in the comments.
Scrabble 엔진, Ultimate Tic-Tac-Toe 봇, 철학 연구 등 세 가지 사이드 프로젝트에서 AI의 기본적인 연구 판단력이 부족했음. 중요한 문헌을 놓치고, 의미 있는 결론을 얻기 어려운 실험을 진행함자료를 찾는 것과 그 가치를 알아보는 것은 달랐음.
See what Optimizely’s Virtual Teammates can automate for you at This week’s AI news covers OpenAI’s GPT-Image 2.6, Google Gemini Omni 1.2 Flash, and the latest…
Here are three AI product drops you might’ve missed this week.