TZUTC: -0500   
   MSGID: 359.consprcy@1:2320/105 2c2ee494   
   PID: Synchronet 3.20a-Linux master/acc19483f Apr 26 202 GCC 12.2.0   
   TID: SBBSecho 3.20-Linux master/acc19483f Apr 26 2024 23:04 GCC 12.2.0   
   BBSID: CAPCITY2   
   CHRS: ASCII 1   
   'Simulating scientists': A new AI tool wants to make serendipitous scientific   
   discovery less human   
      
   Date:   
   Wed, 05 Mar 2025 18:34:00 +0000   
      
   Description:   
   Scientists develop new AI model designed to aid scientific discovery.   
      
   FULL STORY   
      
   An Australian research team led by Monash University has come up with a   
   generative AI tool designed to speed up scientific discoveries. Called LLM4SD   
   (Large Language Model 4 Scientific Discovery), the open source tool retrieves   
   information, analyzes the data, and then generates hypotheses from it.    
      
   While LLMs are used in natural sciences, their role in scientific discovery   
   remains largely unexplored, and unlike many validation tools, LLM4SD explains   
   its reasoning, making its predictions more transparent (and hopefully cutting   
   down on hallucinations).    
      
   PhD candidate Yizhen Zheng from Monash Universitys Department of Data Science   
   and AI explains, Just like ChatGPT writes essays or solves math problems, our   
   LLM4SD tool reads decades of scientific literature and analyses lab data to   
   predict how molecules behave - answering questions like, Can this drug cross   
   the brains protective barrier? or Will this compound dissolve in water?   
      
   Simulating scientists    
      
   LLM4SD was tested over 58 research tasks across physiology, physical   
   chemistry, biophysics, and quantum mechanics, and outperformed leading   
   scientific models, improving accuracy by up to 48% in predicting quantum   
   properties crucial for materials design. Zheng said, Apart from outperforming   
   current validation tools that operate like a black box, this system can   
   explain its analysis process, predictions and results using simple rules,   
   which can help scientists trust and act on its insights.    
      
   PhD candidate Jiaxin Ju from Griffith University said, Rather than replacing   
   traditional machine learning models, LLM4SD enhances them by synthesizing   
   knowledge and generating interpretable explanations.    
      
   The team views the tool as essentially simulating scientists. Professor Geoff   
   Webb from Monash University stressed the importance of AIs role in research.   
   We are already fully immersed in the age of generative AI and we need to    
   start harnessing this as much as possible to advance science, while ensuring   
   we are developing it ethically, he said.    
      
   The research, published in Nature Machine Intelligence and available to view   
   on the arXiv pre-print server , was a collaboration between Monash    
   Universitys Faculty of Information Technology, Monash Institute of   
   Pharmaceutical Sciences, and Griffith University.   
      
   ======================================================================   
   Link to news story:   
   https://www.techradar.com/pro/simulating-scientists-a-new-ai-tool-wants-to-mak   
   e-serendipitous-scientific-discovery-less-human   
      
   $$   
   --- SBBSecho 3.20-Linux   
    * Origin: capitolcityonline.net * Telnet/SSH:2022/HTTP (1:2320/105)   
   SEEN-BY: 105/81 106/201 128/187 129/305 153/7715 154/110 218/700 226/30   
   SEEN-BY: 227/114 229/110 111 114 206 300 307 317 400 426 428 470 664   
   SEEN-BY: 229/700 705 266/512 291/111 320/219 322/757 342/200 396/45   
   SEEN-BY: 460/58 712/848 902/26 2320/0 105 3634/12 5075/35   
   PATH: 2320/105 229/426