BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Human–AI Collaboration-Driven Cosmology Research
DTSTART;VALUE=DATE-TIME:20261107T083000Z
DTEND;VALUE=DATE-TIME:20261107T090000Z
DTSTAMP;VALUE=DATE-TIME:20261023T133051Z
UID:indico-contribution-2971@indico.itp.ac.cn
DESCRIPTION:Speakers: Jiajun Zhang (Shanghai Astronomical Observatory)\nIn
  modern cosmological research\, using cosmological numerical simulations t
 o study the evolution of the large-scale structure of the Universe has pro
 ven to be a highly effective technique. However\, cosmological numerical s
 imulations beyond the standard cosmological model require numerous modific
 ations to existing cosmological simulation frameworks. They are technicall
 y complex and have a high entry barrier\, so a large number of non-standar
 d cosmological models cannot be effectively validated. In recent years\, t
 he development of artificial intelligence technologies represented by larg
 e language models has brought profound changes to scientific research\; fo
 r example\, large language models such as DeepSeek have rapidly spread acr
 oss various areas of life. Based on the DeepSeek large language model\, we
  developed an agent for cosmological numerical simulations. This agent use
 s multi-agent collaboration technology and integrates functions such as li
 terature review\, code analysis\, and report generation\, allowing users t
 o generate complex cosmological numerical simulation software with a singl
 e sentence. After evaluation\, we found that the agent can effectively avo
 id generating hallucinations and produce work guides with real productive 
 capacity. Using these work guides\, we automatically developed cosmologica
 l numerical simulation software for models including SIDM and WDM. In term
 s of development efficiency\, compared with experienced developers\, the a
 gent is two orders of magnitude more efficient. The application of this ag
 ent will bring about a major revolution in numerical-simulation-based cosm
 ology research. Hundreds or thousands of models that previously could not 
 be tested by simulations will become testable\, and the most time- and exp
 ertise-consuming part of this field has been overcome. We believe that the
  human–AI collaboration method developed in this study can be extended t
 o a broader range of research fields and may bring about a profound transf
 ormation of research paradigms.\n\nhttps://indico.itp.ac.cn/event/417/cont
 ributions/2971/
LOCATION:
URL:https://indico.itp.ac.cn/event/417/contributions/2971/
END:VEVENT
END:VCALENDAR