Aug 2026
RINet: an attention-based transformer with the RIT-3V benchmark for semantic role identification in human activities
This work establishes the first transformer-based benchmark for role identification in human activity descriptions and provides a scalable foundation for advanced activity reasoning and knowledge extraction.
Vivek Tiwari, Anam Arshad, Mayank Lovanshi et al.
· Neural computing & applicati... · 0 citations