Voici 2025, Claude est désormais listé comme auteur sur des articles de recherche publiés sur arXiv. Un article "The Illusion of Thinking" vient tout juste de paraître, avec Claude Opus en tant que premier auteur. L'IA entre dans l'arène comme un acteur scientifique à part
LLMS
-
Multimodal LLM Creates Human-Like Internal Object Maps in Study
By
–
Ce que révèle l’étude : Même sans règles explicites, un LLM multimodal parvient à créer une carte interne des objets qui rejoint nos concepts humains — avec ses propres chemins, mais les mêmes destinations. Les humains associent les objets à leur usage, leur texture, leur rôle
-
ChatGPT Search Update: Enhanced Multi-Query Capabilities
By
–
ChatGPT Search feature got updated Interestingly, now it can perform more complex search that implies taking multiple queries underneath. This is a list of changes behind 4o "thinking" that users have been noticing during the past month.
-
Chinese Scientists Confirm AI Can Form Human-Level Cognition Spontaneously
By
–
Des scientifiques chinois confirment qu’une IA est capable de former spontanément une cognition de niveau humain. « Le résultat montre que les modèles de langage ne sont pas de simples “perroquets stochastiques”. Au contraire, ces modèles possèdent une compréhension interne des
-
AI Navigation Platform Dedicated Teachers DeepSeek R1
By
–
发现一个专门为老师制作的 AI 工具导航站，从幼儿专区到高中专区，从英文专区到数学专区 老师帮 – 需要登录 – 提供免费体验 用得是 DeepSeek R1，不仅老师可以用，家长也可以用 亮点
• 汇集教学、备课、出题、批改、成绩分析等多种 AI 工具
• 支持一键生成教案、作业、试题、教学反思、总结等
• -
Can ChatGPT Be Imported? Comparing LLM Integration Options
By
–
Is ChatGPT the only LLM can be imported?
-
Concerns about shallow distance between LLM ethics and real patient harm
By
–
I'd worry that this distance is shallower than the distance between "What are your opinions on medical ethics" and observing what happens when an LLM talks to a susceptible patient.
-
LLMs Struggle Learning From Millions of Games and Tutorials
By
–
there are millions of games and thousands of tutorial on the web that llms likely train on, to no avail.
-
World LLM Systems Key Role in Major Operations
By
–
which is why the world LLM is key in the OP, of course
-
LLMs Produce Plausible but Deeply Wrong Outputs Across Domains
By
–
Terence Tao reporting on the same problem we have seen endlessly in other domains: LLM’s produce output that “looks” correct but is often deeply wrong and even stupid on careful inspection. I know of no domain in which this is NOT the case. And yet people seem surprised over x.com/vitrupo/status…