The MUltiplexed Survey Telescope (MUST) is an ambitious 6.5-m telescope for large-scale cosmological surveys. This presentation will introduce the MUST initiative and delve into its pivotal role in the forthcoming Stage-V cosmological survey. As we venture into this new era of cosmic exploration, one of the most daunting challenges we face is handling the huge amount of data these surveys will produce. We will discuss the complexities of scientific prediction, survey planning, and cosmological inference for MUST and Stage-V surveys. Central to our strategy is the adoption of advanced data science techniques and intelligent computing. Their applications are not just ancillary but fundamental to unlocking new understandings in cosmology. Join us as we explore the nexus of observational cosmology and cutting-edge computational methods, where the MUST project serves as a platform for the future of our understanding of the universe. 个人简介：Song Huang is an Assistant Professor at the Department of Astronomy in Tsinghua University. He has conducted postdoctoral research at the University of Tokyo in Japan, the University of California, Santa Cruz, and Princeton University in the United States. He is a member of the Subaru Strategic Program (SSP) using the Hyper-Suprime Camera (HSC) on the 8.2-meter Subaru Telescope and the Merian Survey. He has published over 60 research papers in SCI-indexed journals. Currently, he is leading the scientific preparation of the 6.5-m MUltiplexed Survey Telescope (MUST) for the next generation of spectroscopic surveys.

Now a days, artificial intelligence algorithms are broadly applied in many fields of astronomy, especially in processing and analyzing observed data. It takes great advantage when the observed data volume is extremely large and the analysis is computational consuming. Here, I report a series of studies, mostly based on CSST survey, which is a large multi-band photometric and slitless spectroscopic survey will be operated by Chinese Space Survey Telescope (CSST), to reduce data and make analysis on the data using AI algorithms. These studies can well demonstrate how AI algorithms help in a large survey containing tens million images and around 10 billion celestial objects.

Molecules, minerals, ices and organics begin to form in the diffuse medium and keep evolving through a complex journey until they are incorporated into planetary bodies. Cold cores are a key stage in this cosmic course, as the composition of ice and gas in these stellar nurseries determines further evolutionary sequences and, eventually, the initial conditions for building planets, atmospheres and the first bricks of life. New observatories (JWST, ALMA and NOEMA) allow us to probe the interstellar medium (ISM) with unprecedented spectral coverage, spatial and spectral resolutions. To interpret these observations, a new generation of sophisticated chemical models are built based on laboratory astrophysics and coupled with the dynamical evolution of the interstellar matter. Overall, both from the observational and modeling points of view, a large number of data are obtained/computed and we are developing methods to treat, visualize and interpret them. In this presentation, we will show some of these methods together with the results that we can derive from them. We will focus on the chemical composition of cold dense cores as they are the most "simple" sources, abundant in molecules, on which astrochemical models can be tested.

In recent years, AI, notably through advancements in Large Language Models (LLMs) such as ChatGPT, has garnered significant attention both within academia and the broader public sphere. However, these general-purpose LLMs have been criticized for their tendency to produce spurious or 'hallucinated' information when grappling with specialized or technical domains. To address this limitation, we introduce the UniverseTBD Consortium—an international collaboration, comprising a diverse team of 30 active contributors from computer science and astronomy. Our mission is to democratize the field of astronomy by developing public-facing, AI-driven large language model tools specialized for this discipline. Our research presents the first astronomy-centric LLM, AstroLLaMa, that can produces text completion and embedding that outperform GPT models. We also show that LLMs can generate scientific hypotheses of a complexity comparable to those produced by human experts through techniques such as in-context prompting and fine-tuning on domain-specific literature. Moreover, we posit that these specialized foundational models can revolutionize the methods we employ for literature searches and the tracing of intellectual developments within the field. We argue that the domain of physical sciences, particularly astronomy, serves as an ideal test bed for investigating the potential of modern LLMs. This inquiry stands to fundamentally reshape our understanding of both artificial and human intelligence and the boundaries of accumulated knowledge.

In this lecture I will describe the history of searching the radio sky for fast radio transients and how this led to the discovery and deciphering of the Fast Radio Burst phenomenon.