Connecting the Dots: LLMs can Infer and Verbalize Latent Structure from Disparate Training Data.
Johannes TreutleinDami ChoiJan BetleyCem AnilSamuel MarksRoger Baker GrosseOwain EvansPublished in: CoRR (2024)
Keyphrases
- latent structure
- training data
- multidimensional scaling
- discriminative learning
- negative matrix factorization
- latent variables
- generative model
- learning algorithm
- document classification
- latent variable models
- topic modeling
- training set
- group activities
- low dimensional
- data sets
- nonnegative matrix factorization
- decision trees
- supervised learning
- prior knowledge
- maximum margin
- generalization error
- topic models
- unlabeled data
- random variables
- latent semantic indexing
- support vector machine
- hidden markov models
- machine learning