From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents

Swathi Shree Narashiman; Ishan Jindal; Mahesh Chandran

From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents

Swathi Shree Narashiman, Ishan Jindal, Mahesh Chandran

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

Procedural memory for language agents typically accumulates as trajectories, scripts, or task-specific skills, so memory grows with the number of interactions rather than the number of distinct procedures learned. We propose a workflow-centric procedural memory framework that continually consolidates repeated experience into a bounded set of natural-language workflows. These workflows are merged online under outcome-conditioned update policies. On ALFWorld and ScienceWorld, our memory is competitive with trajectory-level procedural memory on task success while using 3.5× fewer memory tokens on average (1.1-9.1× across settings), showing that representation besides construction, retrieval and updating is also a key design axis for scalable procedural memory.

Anthology ID:: 2026.realm-1.78
Volume:: Proceedings of the Second Workshop for REsearch on Agent Language Models (REALM 2026)
Month:: October
Year:: 2026
Address:: Budapest, Hungary
Editors:: Ehsan Kamalloo, Nicolas Gontier, Nikolai Rozanov, Maryam Hashemzadeh, Akhil Arora
Venues:: REALM | WS
SIG:
Publisher:: Association for Computational Linguistics
Note:
Pages:: 1144–1159
Language:
URL:: https://aclanthology.org/2026.realm-1.78/
DOI:
Bibkey:
Cite (ACL):: Swathi Shree Narashiman, Ishan Jindal, and Mahesh Chandran. 2026. From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents. In Proceedings of the Second Workshop for REsearch on Agent Language Models (REALM 2026), pages 1144–1159, Budapest, Hungary. Association for Computational Linguistics.
Cite (Informal):: From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents (Narashiman et al., REALM 2026)
Copy Citation:
PDF:: https://aclanthology.org/2026.realm-1.78.pdf

PDF Cite Search Fix data

Export citation

BibTeX
MODS XML
Endnote
Preformatted

@inproceedings{narashiman-etal-2026-trajectories,
    title = "From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents",
    author = "Narashiman, Swathi Shree  and
      Jindal, Ishan  and
      Chandran, Mahesh",
    editor = "Kamalloo, Ehsan  and
      Gontier, Nicolas  and
      Rozanov, Nikolai  and
      Hashemzadeh, Maryam  and
      Arora, Akhil",
    booktitle = "Proceedings of the Second Workshop for {RE}search on Agent Language Models ({REALM} 2026)",
    month = oct,
    year = "2026",
    address = "Budapest, Hungary",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.realm-1.78/",
    pages = "1144--1159",
    ISBN = "979-8-89176-485-9",
    abstract = "Procedural memory for language agents typically accumulates as trajectories, scripts, or task-specific skills, so memory grows with the number of interactions rather than the number of distinct procedures learned. We propose a workflow-centric procedural memory framework that continually consolidates repeated experience into a bounded set of natural-language workflows. These workflows are merged online under outcome-conditioned update policies. On ALFWorld and ScienceWorld, our memory is competitive with trajectory-level procedural memory on task success while using $3.5\times$ fewer memory tokens on average (1.1-$9.1\times$ across settings), showing that representation besides construction, retrieval and updating is also a key design axis for scalable procedural memory."
}

Download as File

<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="narashiman-etal-2026-trajectories">
    <titleInfo>
        <title>From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Swathi</namePart>
        <namePart type="given">Shree</namePart>
        <namePart type="family">Narashiman</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ishan</namePart>
        <namePart type="family">Jindal</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Mahesh</namePart>
        <namePart type="family">Chandran</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 Second Workshop for REsearch on Agent Language Models (REALM 2026)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Ehsan</namePart>
            <namePart type="family">Kamalloo</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nicolas</namePart>
            <namePart type="family">Gontier</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Nikolai</namePart>
            <namePart type="family">Rozanov</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Maryam</namePart>
            <namePart type="family">Hashemzadeh</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Akhil</namePart>
            <namePart type="family">Arora</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">979-8-89176-485-9</identifier>
    </relatedItem>
    <abstract>Procedural memory for language agents typically accumulates as trajectories, scripts, or task-specific skills, so memory grows with the number of interactions rather than the number of distinct procedures learned. We propose a workflow-centric procedural memory framework that continually consolidates repeated experience into a bounded set of natural-language workflows. These workflows are merged online under outcome-conditioned update policies. On ALFWorld and ScienceWorld, our memory is competitive with trajectory-level procedural memory on task success while using 3.5\times fewer memory tokens on average (1.1-9.1\times across settings), showing that representation besides construction, retrieval and updating is also a key design axis for scalable procedural memory.</abstract>
    <identifier type="citekey">narashiman-etal-2026-trajectories</identifier>
    <location>
        <url>https://aclanthology.org/2026.realm-1.78/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>1144</start>
            <end>1159</end>
        </extent>
    </part>
</mods>
</modsCollection>

Download as File

%0 Conference Proceedings
%T From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents
%A Narashiman, Swathi Shree
%A Jindal, Ishan
%A Chandran, Mahesh
%Y Kamalloo, Ehsan
%Y Gontier, Nicolas
%Y Rozanov, Nikolai
%Y Hashemzadeh, Maryam
%Y Arora, Akhil
%S Proceedings of the Second Workshop for REsearch on Agent Language Models (REALM 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-485-9
%F narashiman-etal-2026-trajectories
%X Procedural memory for language agents typically accumulates as trajectories, scripts, or task-specific skills, so memory grows with the number of interactions rather than the number of distinct procedures learned. We propose a workflow-centric procedural memory framework that continually consolidates repeated experience into a bounded set of natural-language workflows. These workflows are merged online under outcome-conditioned update policies. On ALFWorld and ScienceWorld, our memory is competitive with trajectory-level procedural memory on task success while using 3.5\times fewer memory tokens on average (1.1-9.1\times across settings), showing that representation besides construction, retrieval and updating is also a key design axis for scalable procedural memory.
%U https://aclanthology.org/2026.realm-1.78/
%P 1144-1159

Download as File

Markdown (Informal)

[From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents](https://aclanthology.org/2026.realm-1.78/) (Narashiman et al., REALM 2026)

From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents (Narashiman et al., REALM 2026)

ACL

Swathi Shree Narashiman, Ishan Jindal, and Mahesh Chandran. 2026. From Trajectories to Workflows: Online Procedural Memory Abstraction for Language Agents. In Proceedings of the Second Workshop for REsearch on Agent Language Models (REALM 2026), pages 1144–1159, Budapest, Hungary. Association for Computational Linguistics.