Role overview Computer science specialists are needed to evaluate AI-generated graphical abstracts against original research papers as part of an AI verification project. The work centres on judging whether machine-produced summaries accurately and faithfully represent complex technical publications. It is an ad hoc, project-based opportunity where expert judgement directly shapes how AI systems learn to condense scientific literature. Responsibilities Review computer science research papers alongside LLM-produced graphical abstracts Identify factual errors, misrepresentations, or structural inconsistencies in the AI-generated visuals Verify that technical concepts are summarised correctly in both image and supporting text Flag any divergence between the source paper and its AI-rendered abstract Navigate dense scientific literature and apply domain expertise to ambiguous outputs Requirements Master's or PhD in Computer Science, or a closely related STEM field Specialised domain knowledge across computer science and adjacent technical areasDemonstrated analytical skills with prior experience in fact-checking or critical review Strong attention to detail when reading and interpreting technical content Ability to work independently and exercise consistent professional judgement Benefits and work setup Fully remote with flexible hours and ad hoc project scheduling Competitive per-task pay rates Short qualification assessment prior to onboarding; ongoing work released as projects become available #J-18808-Ljbffr
AI Trainer - Computer Science Expert - Graphical Abstract - Fully Remote - (CA) in workfromhome at Unknown Company
This position is listed as full time and able to be worked remotely.