Saved in:
Bibliographic Details
Main Authors: Jan, Aviv, Tahory, Dean, Talmi, Omer, Mokh, Omar Abo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.00483
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Multi-hop questions still stump large language models (LLMs), which struggle to link information across multiple reasoning steps. We introduce Auto-Patch, a novel method that dynamically patches hidden states during inference to enhance multi-hop reasoning in LLMs. Building on the PatchScopes framework, Auto-Patch selectively modifies internal representations using a learned classifier. Evaluated on the MuSiQue dataset, Auto-Patch improves the solve rate from 18.45\% (baseline) to 23.63~$\pm$~0.7\% (3 runs), narrowing the gap to Chain-of-Thought prompting (27.44\%). Our results highlight the potential of dynamic hidden state interventions for advancing complex reasoning in LLMs.