Pranjal Awasthi

From Style to Facts: Mapping the Boundaries of Knowledge Injection with Finetuning

Length Generalization via Auxiliary Tasks

Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verification

Semantic Routing via Autoregressive Modeling

Position Coupling: Improving Length Generalization of Arithmetic Transformers Using Task Structure

Routing with Rich Text Queries via Next-Vertex Prediction Models

Improving length generalization in transformers via task hinting