The Analyst in the Prompt: Role, Retrieval, and Memory Biases in LLM Financial Analysis

Ahmed Asaad; Amr Mohamed; Yang Zhang; Omneya Abdelsalam

The Analyst in the Prompt: Role, Retrieval, and Memory Biases in LLM Financial Analysis

Ahmed Asaad, Amr Mohamed, Yang Zhang, Omneya Abdelsalam

Correct Metadata for

Use this form to create a GitHub issue with structured data describing the correction. You will need a GitHub account. Once you create that issue, the correction will be reviewed by a staff member.

⚠️ Mobile Users: Submitting this form to create a new issue will only work with github.com, not the GitHub Mobile app.

Important: The Anthology treat PDFs as authoritative. Please use this form only to correct data that is out of line with the PDF. See our corrections guidelines if you need to change the PDF.

Title Adjust the title. Retain tags such as <fixed-case>.

Authors Adjust author names and order to match the PDF.

Abstract Correct abstract if needed. Retain XML formatting tags such as <tex-math>. You may use ... for bold, ... for italic, ... for underline, <sc>...</sc> for small-caps, <tt>...<tt> for typewriter text, <url>...</url> for URLs, <a href=...> for hyperlinks, and <par/> for paragraph breaks.

Verification against PDF Ensure that the new title/authors match the snapshot below. (If there is no snapshot or it is too small, consult the PDF.)

Authors concatenated from the text boxes above:

ALL author names match the snapshot above—including middle initials, hyphens, and accents.

Abstract

Large Language Models (LLMs) use persistent user context such as memory, profiles, and role prompts. This enables personalization, but can destabilize evidence-based judgment: the same evidence may yield different conclusions under different user contexts. Finance offers a high-stakes long-document setting where evidence reading is central. We test this on 3,575 SEC filings across twelve models. By comparing persona-conditioned retrieval, neutral retrieval, and memory-framed context, we find that bias comes from how models interpret identical evidence under role-style context, not from which evidence they retrieve. As mitigation, user-memory framing of the same mindset and dual-output scoring reduce the effect, with model-specific effectiveness across models.

Anthology ID:: 2026.emnlp-main.2268
Volume:: Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
Month:: October
Year:: 2026
Address:: Budapest, Hungary
Editors:: Hung-Yi Lee, Sunipa Dev, Jessy Li, Desmond Elliott
Venue:: EMNLP
SIG:: SIGDAT
Publisher:: Association for Computational Linguistics
Note:
Pages:: 51572–51597
Language:
URL:: https://aclanthology.org/2026.emnlp-main.2268/
DOI:
Bibkey:
Cite (ACL):: Ahmed Asaad, Amr Mohamed, Yang Zhang, and Omneya Abdelsalam. 2026. The Analyst in the Prompt: Role, Retrieval, and Memory Biases in LLM Financial Analysis. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 51572–51597, Budapest, Hungary. Association for Computational Linguistics.
Cite (Informal):: The Analyst in the Prompt: Role, Retrieval, and Memory Biases in LLM Financial Analysis (Asaad et al., EMNLP 2026)
Copy Citation:
PDF:: https://aclanthology.org/2026.emnlp-main.2268.pdf

PDF Cite Search Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@inproceedings{asaad-etal-2026-analyst,
    title = "The Analyst in the Prompt: Role, Retrieval, and Memory Biases in {LLM} Financial Analysis",
    author = "Asaad, Ahmed  and
      Mohamed, Amr  and
      Zhang, Yang  and
      Abdelsalam, Omneya",
    editor = "Lee, Hung-Yi  and
      Dev, Sunipa  and
      Li, Jessy  and
      Elliott, Desmond",
    booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
    month = oct,
    year = "2026",
    address = "Budapest, Hungary",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.emnlp-main.2268/",
    pages = "51572--51597",
    ISBN = "979-8-89176-459-0",
    abstract = "Large Language Models (LLMs) use persistent user context such as memory, profiles, and role prompts. This enables personalization, but can destabilize evidence-based judgment: the same evidence may yield different conclusions under different user contexts. Finance offers a high-stakes long-document setting where evidence reading is central. We test this on 3,575 SEC filings across twelve models. By comparing persona-conditioned retrieval, neutral retrieval, and memory-framed context, we find that bias comes from how models interpret identical evidence under role-style context, not from which evidence they retrieve. As mitigation, user-memory framing of the same mindset and dual-output scoring reduce the effect, with model-specific effectiveness across models."
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="asaad-etal-2026-analyst">
    <titleInfo>
        <title>The Analyst in the Prompt: Role, Retrieval, and Memory Biases in LLM Financial Analysis</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Ahmed</namePart>
        <namePart type="family">Asaad</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Amr</namePart>
        <namePart type="family">Mohamed</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yang</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Omneya</namePart>
        <namePart type="family">Abdelsalam</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>Large Language Models (LLMs) use persistent user context such as memory, profiles, and role prompts. This enables personalization, but can destabilize evidence-based judgment: the same evidence may yield different conclusions under different user contexts. Finance offers a high-stakes long-document setting where evidence reading is central. We test this on 3,575 SEC filings across twelve models. By comparing persona-conditioned retrieval, neutral retrieval, and memory-framed context, we find that bias comes from how models interpret identical evidence under role-style context, not from which evidence they retrieve. As mitigation, user-memory framing of the same mindset and dual-output scoring reduce the effect, with model-specific effectiveness across models.</abstract>
    <identifier type="citekey">asaad-etal-2026-analyst</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.2268/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>51572</start>
            <end>51597</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Conference Proceedings
%T The Analyst in the Prompt: Role, Retrieval, and Memory Biases in LLM Financial Analysis
%A Asaad, Ahmed
%A Mohamed, Amr
%A Zhang, Yang
%A Abdelsalam, Omneya
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F asaad-etal-2026-analyst
%X Large Language Models (LLMs) use persistent user context such as memory, profiles, and role prompts. This enables personalization, but can destabilize evidence-based judgment: the same evidence may yield different conclusions under different user contexts. Finance offers a high-stakes long-document setting where evidence reading is central. We test this on 3,575 SEC filings across twelve models. By comparing persona-conditioned retrieval, neutral retrieval, and memory-framed context, we find that bias comes from how models interpret identical evidence under role-style context, not from which evidence they retrieve. As mitigation, user-memory framing of the same mindset and dual-output scoring reduce the effect, with model-specific effectiveness across models.
%U https://aclanthology.org/2026.emnlp-main.2268/
%P 51572-51597

Download as File

Markdown (Informal)

[The Analyst in the Prompt: Role, Retrieval, and Memory Biases in LLM Financial Analysis](https://aclanthology.org/2026.emnlp-main.2268/) (Asaad et al., EMNLP 2026)

The Analyst in the Prompt: Role, Retrieval, and Memory Biases in LLM Financial Analysis (Asaad et al., EMNLP 2026)

ACL

Ahmed Asaad, Amr Mohamed, Yang Zhang, and Omneya Abdelsalam. 2026. The Analyst in the Prompt: Role, Retrieval, and Memory Biases in LLM Financial Analysis. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 51572–51597, Budapest, Hungary. Association for Computational Linguistics.