Harshil Vejendla

14 papers A* 2Journal 9Unranked 3

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	RewriteNets: End-to-End Trainable String-Rewriting for Generative Sequence Modeling. CoRR Harshil Vejendla
2025	A*	`conf`	Drift-Adapter: A Practical Approach to Near Zero-Downtime Embedding Model Upgrades in Vector Databases. EMNLP Harshil Vejendla
2025	J	`jnl`	Drift-Adapter: A Practical Approach to Near Zero-Downtime Embedding Model Upgrades in Vector Databases. CoRR Harshil Vejendla
2025	J	`jnl`	Efficient Uncertainty Estimation via Distillation of Bayesian Large Language Models. CoRR Harshil Vejendla, Haizhou Shi, Yibin Wang, Tunyu Zhang, Huan Zhang, Hao Wang
2025	J	`jnl`	H1B-KV: Hybrid One-Bit Caches for Memory-Efficient Large Language Model Inference. CoRR Harshil Vejendla
2025	J	`jnl`	LATTA: Langevin-Anchored Test-Time Adaptation for Enhanced Robustness and Stability. CoRR Harshil Vejendla
2025	J	`jnl`	Learning to Predict Chaos: Curriculum-Driven Training for Robust Forecasting of Chaotic Dynamics. CoRR Harshil Vejendla
2025	—	`conf`	RewriteNets: End-to-End Trainable String-Rewriting for Generative Sequence Modeling. IJCNLP-AACL (Findings) Harshil Vejendla
2025	A*	`conf`	SliceMoE: Routing Embedding Slices Instead of Tokens for Fine-Grained and Balanced Transformer Scaling. EMNLP Harshil Vejendla
2025	J	`jnl`	SliceMoE: Routing Embedding Slices Instead of Tokens for Fine-Grained and Balanced Transformer Scaling. CoRR Harshil Vejendla
2025	—	`conf`	Teaching by Failure: Counter-Example-Driven Curricula for Transformer Self-Improvement. IJCNLP-AACL (Findings) Harshil Vejendla
2025	J	`jnl`	Teaching by Failure: Counter-Example-Driven Curricula for Transformer Self-Improvement. CoRR Harshil Vejendla
2025	—	`conf`	Wave-PDE Nets: Trainable Wave-Equation Layers as an Alternative to Attention. PRICAI (4) Harshil Vejendla
2025	J	`jnl`	Wave-PDE Nets: Trainable Wave-Equation Layers as an Alternative to Attention. CoRR Harshil Vejendla