<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="he-etal-2025-delta">
    <titleInfo>
        <title>DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning Ability</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yunzhen</namePart>
        <namePart type="family">He</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yusuke</namePart>
        <namePart type="family">Takase</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hidetoshi</namePart>
        <namePart type="family">Shimodaira</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2nd Workshop on Uncertainty-Aware NLP (UncertaiNLP 2025)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Noidea</namePart>
            <namePart type="family">Noidea</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-349-4</identifier>
    </relatedItem>
    <abstract>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.</abstract>
    <identifier type="citekey">he-etal-2025-delta</identifier>
    <location>
        <url>https://aclanthology.org/2025.uncertainlp-main.26/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>309</start>
            <end>321</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning Ability
%A He, Yunzhen
%A Takase, Yusuke
%A Shimodaira, Hidetoshi
%Y Noidea, Noidea
%S Proceedings of the 2nd Workshop on Uncertainty-Aware NLP (UncertaiNLP 2025)
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-349-4
%F he-etal-2025-delta
%X 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.
%U https://aclanthology.org/2025.uncertainlp-main.26/
%P 309-321

Markdown (Informal)

[DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning Ability](https://aclanthology.org/2025.uncertainlp-main.26/) (He et al., UncertaiNLP 2025)

• DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning Ability (He et al., UncertaiNLP 2025)

ACL

• Yunzhen He, Yusuke Takase, and Hidetoshi Shimodaira. 2025. DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning Ability. In Proceedings of the 2nd Workshop on Uncertainty-Aware NLP (UncertaiNLP 2025), pages 309–321, Suzhou, China. Association for Computational Linguistics.