collaborators

9 papers

cs.LG2026

Dist2ill: Distributional Distillation for One-Pass Uncertainty Estimation in Large Language Models

Yicong Zhao, King Yeung Tsang, Harshil Vejendla +9

Large Language Models (LLMs) often exhibit misalignment between the quality of their generated responses and the confidence estimates they assign to them. Bayesian treatments, such…

cs.LG2026

RewriteNets: End-to-End Trainable String-Rewriting for Generative Sequence Modeling

Harshil Vejendla

Dominant sequence models like the Transformer represent structure implicitly through dense attention weights, incurring quadratic complexity. We propose RewriteNets, a novel neural…

cs.LG2025

Teaching by Failure: Counter-Example-Driven Curricula for Transformer Self-Improvement

Harshil Vejendla

Transformer models often exhibit brittle extrapolation, failing on inputs that are longer or structurally more complex than those seen during training. We introduce Counter-Example…

cs.LG2025

LATTA: Langevin-Anchored Test-Time Adaptation for Enhanced Robustness and Stability

Harshil Vejendla

Test-time adaptation (TTA) aims to adapt a pretrained model to distribution shifts using only unlabeled test data. While promising, existing methods like Tent suffer from instabili…

cs.CL2025

H1B-KV: Hybrid One-Bit Caches for Memory-Efficient Large Language Model Inference

Harshil Vejendla

Autoregressive decoding in large language models (LLMs) requires caching a growing list of past key-value (KV) pairs, making long-context inference a memory-bound problem. While re…

cs.LG2025

Learning to Predict Chaos: Curriculum-Driven Training for Robust Forecasting of Chaotic Dynamics

Harshil Vejendla

Forecasting chaotic systems is a cornerstone challenge in many scientific fields, complicated by the exponential amplification of even infinitesimal prediction errors. Modern machi…