AI-powered systems for scholarly search and content production
Generative AI systems can be trained to perform useful research tasks such as quickly summarizing research findings, generating visual representations of scientific content, and simplifying technical jargon. Tailoring language, format, tone, and examples can make research more accessible, understand...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | text |
| 出版: |
Institutional Knowledge at Singapore Management University
2024
|
| 主題: | |
| 在線閱讀: | https://ink.library.smu.edu.sg/ai_research_week/Programme/Programme/8 https://ink.library.smu.edu.sg/context/ai_research_week/article/1007/type/native/viewcontent/AI_Powered_Systems_for_Scholarly_Search_and_Content_Production___AI_for_Research_Week_2024__1_.mp4 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 機構: | Singapore Management University |
| 總結: | Generative AI systems can be trained to perform useful research tasks such as quickly summarizing research findings, generating visual representations of scientific content, and simplifying technical jargon. Tailoring language, format, tone, and examples can make research more accessible, understandable, engaging, and useful for different audiences. However, these uses of AI also raise questions about credit and attribution, informational provenance, the responsibilities of authorship, control over science communication, and more.In this talk, I will introduce several AI-powered prototype systems that assist with difficult tasks such as describing the content of complex scientific figures or synthesizing research threads across hundreds or thousands of papers. I encourage attendees to interact with live demonstration systems and will facilitate a discussion around the potential uses and limitations of these tools, as well as ways of addressing these limitations, by incorporating human feedback in the generative process and mitigating the production of false or misleading information. |
|---|