5 papers
Joint Flow Matching for Generator-Consistent Classification
Hayden McAlister, Lech Szymanski
We introduce Joint Flow Matching (JFM), a training framework for continuous normalising flows over multiple variables. Standard flow matching transports variables from noise to dat…
Faithful Evaluation of Semantic-ID Tokenizers for Generative Recommendation
Qian Zhang, Lech Szymanski, Haibo Zhang +1
Generative recommendation based on Semantic IDs (SIDs) represents each item as a discrete sequence of SIDs and is conventionally evaluated by matching the generated SID sequence ag…
CAST: Modeling Semantic-Level Transitions for Complementary-Aware Sequential Recommendation
Qian Zhang, Lech Szymanski, Haibo Zhang +1
Sequential Recommendation (SR) aims to predict the next interaction of a user based on their behavior sequence, where complementary relations often provide essential signals for pr…
Classifying States of the Hopfield Network with Improved Accuracy, Generalization, and Interpretability
Hayden McAlister, Anthony Robins, Lech Szymanski
We extend the existing work on Hopfield network state classification, employing more complex models that remain interpretable, such as densely-connected feed-forward deep neural ne…
Sequential Learning in the Dense Associative Memory
Hayden McAlister, Anthony Robins, Lech Szymanski
Sequential learning involves learning tasks in a sequence, and proves challenging for most neural networks. Biological neural networks regularly conquer the sequential learning cha…