One-Pass to Reason: Token Duplication and Block-Sparse Mask for Efficient Fine-Tuning on Multi-Turn Reasoning
Year: 2025
Type: preprint
Abstract: Fine-tuning Large Language Models (LLMs) on multi-turn reasoning datasets requires N (number of turns) separate forward passes per conversation due to reasoning token visibility constraints, as reasoning tokens for a turn are discarded in subsequent turns. We propose duplicating response tokens along with a custom attention mask to enable single-pass processing of entire conversations. We prove our method produces identical losses to the N-pass approach while reducing time complexity from $O\bigl(N^{3}\bigl)$ to $O\bigl(N^{2}\bigl)$ and maintaining the same memory complexity for a transformer based model. Our approach achieves significant training speedup while preserving accuracy. Our implementation is available online (https://github.com/devrev/One-Pass-to-Reason).
Source: arXiv (Cornell University)
Authors: Goru, Ritesh, Mehta, Shanay, Jain, Prateek
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6979314659
Cites:
Cited by:
Related to: