PulseAugur
实时 21:26:03

AI代理发现新的密度泛函,准确性提高9%

一个AI代理发现了一种新的交换-关联泛函SAFS26-a,其在量子化学中的准确性比之前的标准提高了9%。另外,一个名为BALAR(用于主动推理的贝叶斯代理循环)的新算法,通过主动寻求缺失信息,增强了大型语言模型进行多轮对话的能力,从而带来了显著的准确性提升。 AI

影响 这些在代理发现和主动推理方面的进展可能会加速科学突破,并提高LLM在复杂任务中的性能。

排序理由 该集群描述了AI驱动的量子化学和LLM推理方面的新研究发现。

在 Mastodon — mastodon.social 阅读 →

AI 生成摘要 · Google Gemini · 来自 5 个来源。 我们如何撰写摘要 →

AI代理发现新的密度泛函,准确性提高9%

报道来源 [5]

  1. arXiv cs.AI TIER_1 English(EN) · Titouan Duston, Jiashu Liang, Yuanheng Wang, Weihao Gao, Xuelan Wen, Nan Sheng, Weiluo Ren, Yang Sun, Yixiao Chen ·

    Agentic Discovery of Exchange-Correlation Density Functionals

    arXiv:2605.05460v1 Announce Type: new Abstract: The development of accurate exchange-correlation (XC) functionals remains a longstanding challenge in density functional theory (DFT). The vast majority of XC functionals have been hand designed by human researchers combining physic…

  2. Mastodon — mastodon.social TIER_1 English(EN) · aihaberleri ·

    📰 Agentic Discovery Creates 9% More Accurate Exchange-Correlation Functional in 2026 Agentic discovery of exchange-correlation density functionals has achieved

    📰 Agentic Discovery Creates 9% More Accurate Exchange-Correlation Functional in 2026 Agentic discovery of exchange-correlation density functionals has achieved a 9% improvement over the gold-standard ωB97M-V, marking a milestone in AI-driven quantum chemistry. The breakthrough, h…

  3. Mastodon — mastodon.social TIER_1 Türkçe(TR) · aihaberleri ·

    📰 AI Discovers 60-Year-Old DFT Exchange-Correlation Functional: %9 Accuracy with SAFS26-a... An AI agent has discovered a 60-year-old exchange-correlation functional in density functional theory

    📰 Yapay Zekâ, 60 Yıllık DFT Exchange-Correlation Fonksiyonunu Keşfetti: SAFS26-a ile %9 Doğruluk Ar... Bir yapay zekâ ajanı, yoğunluk fonksiyonel teorisinde 60 yıldır çözülemeyen bir sorunu çözdü: exchange-correlation fonksiyonlarını otomatik olarak keşfetti. Bu keşif, kimya ve m…

  4. Mastodon — mastodon.social TIER_1 English(EN) · aihaberleri ·

    📰 BALAR: 38.5% Accuracy Boost in LLMs with Bayesian Agentic Loop (2026) BALAR, a Bayesian Agentic Loop for Active Reasoning, transforms how large language model

    📰 BALAR: 38.5% Accuracy Boost in LLMs with Bayesian Agentic Loop (2026) BALAR, a Bayesian Agentic Loop for Active Reasoning, transforms how large language models engage in multi-turn dialogues by proactively identifying missing information. The algorithm achieves unprecedented ac…

  5. Mastodon — mastodon.social TIER_1 Türkçe(TR) · aihaberleri ·

    📰 BALAR: The Active Thinking Power of AI with Bayesian Agentic Loop BALAR, which enables AI to actively ask questions instead of remaining passive in dialogues, provides information

    📰 BALAR: Bayesian Agentic Loop ile Yapay Zekânın Aktif Düşünme Gücü Yapay zekânın diyaloglarda pasif kalması yerine aktif soru sormasını sağlayan BALAR, bilgi eksikliklerini kendi başına analiz ederek insan gibi akıl yürütüyor. Bu yeni algoritma, AI'nın nasıl anladığını kökten de…