activity
20242026
most citedUnveiling the Secret Recipe: A Guide For Supervised Fine-Tuning Small LLMs

3 citations · 3 across the 9 of their papers we have counts for

collaborators
Showing cs.LGShow all

7 papers · 1 filter

cs.LG2026

Correcting Stochastic Update Bias in Preconditioned Language Model Optimizers

Nikhil Nayak, Julia White, Urchade Zaratiana +7

Preconditioned optimizers are central to language model training, but their stochastic update rules are usually treated as direct approximations to population preconditioned descen…

cs.LG2025

Mitigating Premature Exploitation in Particle-based Monte Carlo for Inference-Time Scaling

Giorgio Giannone, Guangxuan Xu, Nikhil Shivakumar Nayak +4

Inference-Time Scaling (ITS) improves language models by allocating more computation at generation time. Particle Filtering (PF) has emerged as a strong ITS method for complex math…

cs.LG2025

Learning Task Mixtures from Task Affinities: A Probabilistic Graphical Model for Supervised Fine-Tuning

Prateek Chanda, Saral Sureka, Parth Pratim Chatterjee +3

Supervised fine-tuning performance for large language models depends strongly on how training budget is distributed across a heterogeneous set of tasks. In practice, mixtures are o…

cs.LG2025

Sculpting Subspaces: Constrained Full Fine-Tuning in LLMs for Continual Learning

Nikhil Shivakumar Nayak, Krishnateja Killamsetty, Ligong Han +8

Continual learning in large language models (LLMs) is prone to catastrophic forgetting, where adapting to new tasks significantly degrades performance on previously learned ones. E…

cs.LG2025

Graph Attention for Heterogeneous Graphs with Positional Encoding

Nikhil Shivakumar Nayak

Graph Neural Networks (GNNs) have emerged as the de facto standard for modeling graph data, with attention mechanisms and transformers significantly enhancing their performance on…

cs.LG2025

Towards Interpretable Soft Prompts

Oam Patel, Jason Wang, Nikhil Shivakumar Nayak +2

Soft prompts have been popularized as a cheap and easy way to improve task-specific LLM performance beyond few-shot prompts. Despite their origin as an automated prompting method,…