Darius Peteleaza — accepted AI-conference papers

2025

ConciseRL: Conciseness-Guided Reinforcement Learning for Efficient Reasoning Models

Razvan-Gabriel Dumitru, Darius Peteleaza, Vikas Yadav, Liangming Pan

EMNLP 2025

Large language models excel at complex tasks by breaking down problems into structured reasoning steps. However, reasoning traces often extend beyond reaching a correct answer, causing wasted computation, reduced readability, and hallucinations. To address this, we introduce a novel hyperparameter-f

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗

2024

Change Is the Only Constant: Dynamic LLM Slicing based on Layer Redundancy

Razvan-Gabriel Dumitru, Paul Ioan Clotan, Vikas Yadav, Darius Peteleaza, Mihai Surdeanu

EMNLP 2024finding

This paper introduces a novel model compression approach through dynamic layer-specific pruning in Large Language Models (LLMs), enhancing the traditional methodology established by SliceGPT. By transitioning from constant to dynamic slicing, our method leverages the newly proposed Layer Redundancy…

Cited by 7 Source ↗PDF ↗code ↗Scholar ↗