DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning Ability

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: He, Yunzhen, Takase, Yusuke, Ishibashi, Yoichi, Shimodaira, Hidetoshi
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915181475397632
author He, Yunzhen
Takase, Yusuke
Ishibashi, Yoichi
Shimodaira, Hidetoshi
author_facet He, Yunzhen
Takase, Yusuke
Ishibashi, Yoichi
Shimodaira, Hidetoshi
contents Large Language Models (LLMs) are increasingly being used in real-world applications. However, concerns about the reliability of the content they generate persist, as it frequently deviates from factual correctness or exhibits deficiencies in logical reasoning. This paper proposes a novel decoding strategy aimed at enhancing both factual accuracy and inferential reasoning without requiring any modifications to the architecture or pre-trained parameters of LLMs. Our approach adjusts next-token probabilities by analyzing the trajectory of logits from lower to higher layers in Transformers and applying linear regression. We find that this Decoding by Logit Trajectory-based approach (DeLTa) effectively reinforces factuality and reasoning while mitigating incorrect generation. Experiments on TruthfulQA demonstrate that DeLTa attains up to a 4.9% improvement over the baseline. Furthermore, it enhances performance by up to 8.1% on StrategyQA and 7.3% on GSM8K, both of which demand strong reasoning capabilities.
format Preprint
id arxiv_https___arxiv_org_abs_2503_02343
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning Ability
He, Yunzhen
Takase, Yusuke
Ishibashi, Yoichi
Shimodaira, Hidetoshi
Computation and Language
Machine Learning
Large Language Models (LLMs) are increasingly being used in real-world applications. However, concerns about the reliability of the content they generate persist, as it frequently deviates from factual correctness or exhibits deficiencies in logical reasoning. This paper proposes a novel decoding strategy aimed at enhancing both factual accuracy and inferential reasoning without requiring any modifications to the architecture or pre-trained parameters of LLMs. Our approach adjusts next-token probabilities by analyzing the trajectory of logits from lower to higher layers in Transformers and applying linear regression. We find that this Decoding by Logit Trajectory-based approach (DeLTa) effectively reinforces factuality and reasoning while mitigating incorrect generation. Experiments on TruthfulQA demonstrate that DeLTa attains up to a 4.9% improvement over the baseline. Furthermore, it enhances performance by up to 8.1% on StrategyQA and 7.3% on GSM8K, both of which demand strong reasoning capabilities.
title DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning Ability
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2503.02343