ChatGPT as an inventor: Eliciting the strengths and weaknesses of current large language models against humans in engineering design.
Daniel Nygård EgeHenrik H. ØvrebøVegar StubberudMartin Francis BergChrister ElverumMartin SteinertHåvard VestadPublished in: CoRR (2024)
Keyphrases
- strengths and weaknesses
- engineering design
- language model
- language modeling
- advantages and disadvantages
- document retrieval
- conceptual design
- probabilistic model
- speech recognition
- n gram
- test collection
- query expansion
- design process
- artificial intelligence
- language modelling
- retrieval model
- information retrieval
- statistical language models
- language models for information retrieval
- relative strengths and weaknesses
- ad hoc information retrieval
- engineering problems
- engineering education
- translation model
- relevance model
- vector space model
- document ranking
- smoothing methods
- context sensitive
- usability testing
- bayesian networks
- hidden markov models
- co occurrence
- query terms