Post-training alignment is central to deploying large language models (LLMs), yet practical workflows remain split across backend-specific tools and ad-hoc glue code, making experiments hard to reproduce. We identify backend interference, reward fragmentation, and irreproducible pipelines as key obstacles in alignment research. We introduce AlignTune, a modular toolkit exposing a unified interface for supervised fine-tuning (SFT) and RLHF-style optimization with interchangeable TRL and Unsloth backends. AlignTune standardizes configuration, provides an extensible reward layer (rule-based and learned), and integrates evaluation over standard benchmarks and custom tasks. By isolating backend-specific logic behind a single factory boundary, AlignTune enables controlled comparisons and reproducible alignment experiments.
@article{arxiv.2602.09621,
title = {AlignTune: Modular Toolkit for Post-Training Alignment of Large Language Models},
author = {R E Zera Marveen Lyngkhoi and Chirag Chawla and Pratinav Seth and Utsav Avaiya and Soham Bhattacharjee and Mykola Khandoga and Rui Yuan and Vinay Kumar Sankarapu},
journal= {arXiv preprint arXiv:2602.09621},
year = {2026}
}
Comments
Library opensource and available at https://github.com/Lexsi-Labs/aligntune