Histórias consolidadas

Anthropic firma parceria com OpenEvidence para IA médica CNN Brasil

Centros de Dados de IA da Google tornam‑se parceiros da Grid datacenterknowledge.com

EXCLUSIVO: Anthropic, OpenEvidence se unem para levar IA médica ao redor do mundo reuters.com

Anthropic lançou a versão mais recente de seu modelo de médio alcance, ostentando tempos de resposta mais rápidos e menor consumo de tokens.

Os dois trailers da PlusAI com marcações de parceiro da Nvidia foram aparentemente deixados do lado de fora do armazém da startup, tornando-os um alvo atraente para criminosos que buscam ganhar dinheiro fácil com GPUs de IA. No entanto, eles foram deixados deliberadamente do lado de fora, pois os trailers continham apenas areia para simular cargas reais de caminhões.

URL do artigo: https://www.f-secure.com/en/partners/insights/can-ai-shopping-agents-be-trusted-we-built-one-to-find-out URL dos comentários: https://news.ycombinator.com/item?id=49853522 Pontos: 4 # Comentários: 0

Compre um passe para o TechCrunch Disrupt 2026 e ganhe 50% de desconto no segundo do mesmo tipo de ingresso. Inscreva‑se antes do início do evento em 13 de outubro, às 8h PT.

Google Beam expands with new regions, partners, and customers blog.google

Novo Nordisk se associa à Anthropic para aprofundar a IA na descoberta de medicamentos Yahoo Finance

Biblioteca Pública de San Diego faz parceria com o Google para oferecer bolsas de estudo para certificados de carreira Inside San Diego

Synopsys e TSMC se unem para acelerar a inovação em sistemas de IA com IA agente e design avançado Synopsys

Meaning identity (whether two sentences say the same thing after wording changes) is treated in retrieval and RAG as a geometric fact about independently encoded sentence vectors. We show that, for frozen off-the-shelf encoders and language models, it is not: identity is computed when both sentences share one forward pass, and is not a property of the embedding geometry those systems ship. On overlap-matched PAWS-X, purpose-built encoders (BGE, E5, GTE, MiniLM, E5-Mistral-7B) reach English confirm AUC only 0.55-0.65 (dense peak 0.70). Independently encoded last-token states of Llama 3, Mistral, and Qwen do no better; late fusion of the two vectors stays near chance. The same probe on a joint forward pass reaches 0.90-0.96 from 1.5B to 32B, collapses under partner shuffle, is mid-depth, saturates near 0.94 by 3B, and appears more weakly in GPT-2 XL (0.76). The gap holds beyond Llama-styl…

No AI Day Singapore, NVIDIA e parceiros apresentam avanços de IA em toda o Sudeste Asiático NVIDIA Blog

This work presents the task and results of the CHiME-9 challenge for Enhancing Conversations to address Hearing Impairment. The challenge considers the scenario of four-party conversations in a noisy, cafeteria-style environment with interfering speech sources and sound effects. Participants are provided with audio recordings made with Meta Aria glasses and hearing aid microphones, and clean speech samples of the conversation participants. The task is to extract the speech of the conversation partners from the noisy multi-channel recordings with the goal of improving the intelligibility and quality of the speech, evaluated using objective metrics and subjective listening tests. This paper reviews submissions from seven teams and ranks them on a combination of subjective intelligibility and quality. Results show that while the objective metrics do not reflect listener performance, the to…