<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="moon-cohen-2026-lightweight">
    <titleInfo>
        <title>Lightweight and Faithful Visual Condition Checking in Behavior Trees via Expert-Regularized Reinforcement Learning</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Hyosik</namePart>
        <namePart type="family">Moon</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Eldan</namePart>
        <namePart type="family">Cohen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Maria</namePart>
            <namePart type="family">Liakata</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Viviane</namePart>
            <namePart type="given">P</namePart>
            <namePart type="family">Moreira</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jiajun</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">David</namePart>
            <namePart type="family">Jurgens</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">San Diego, California, United States</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-390-6</identifier>
    </relatedItem>
    <abstract>Behavior trees provide a transparent and modular structure for encoding expert-designed policies, enabling interpretable decision-making in complex tasks. Yet, applying behavior trees to high-dimensional perceptual inputs such as images or language is challenging as defining symbolic predicates over raw perceptual data is non-trivial. While state-of-the-art large multimodal models (such as vision-language models) can overcome this issue by utilizing natural language queries over perceptual inputs, they incur high computational cost, making them unsuitable for many applications. Imitation learning offers a way to distill these expert models into compact models, though it requires extensive supervision. In contrast, reinforcement learning reduces the need for costly supervision but risks misalignment of condition nodes with their intended semantics as well as poor credit assignment. To address these challenges, we introduce CERL (Condition-node Expert-regularized Reinforcement Learning), a framework that leverages expert-regularized reinforcement learning to preserve semantic faithfulness, while employing a factorized policy that aggregates sequential condition-node decisions into a single decision unit to alleviate credit assignment challenges. Experiments across seven tasks from the GymCards, FrozenLake, and BabyAIText suites demonstrate that our framework outperforms pure imitation learning or reinforcement learning baselines, retains strong agreement with expert decisions, and achieves substantial gains in inference speed and model size over expert models. Our implementation is available in https://github.com/HyosikMoon/CERL.</abstract>
    <identifier type="citekey">moon-cohen-2026-lightweight</identifier>
    <identifier type="doi">10.18653/v1/2026.acl-long.1935</identifier>
    <location>
        <url>https://aclanthology.org/2026.acl-long.1935/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>41764</start>
            <end>41799</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Lightweight and Faithful Visual Condition Checking in Behavior Trees via Expert-Regularized Reinforcement Learning
%A Moon, Hyosik
%A Cohen, Eldan
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F moon-cohen-2026-lightweight
%X Behavior trees provide a transparent and modular structure for encoding expert-designed policies, enabling interpretable decision-making in complex tasks. Yet, applying behavior trees to high-dimensional perceptual inputs such as images or language is challenging as defining symbolic predicates over raw perceptual data is non-trivial. While state-of-the-art large multimodal models (such as vision-language models) can overcome this issue by utilizing natural language queries over perceptual inputs, they incur high computational cost, making them unsuitable for many applications. Imitation learning offers a way to distill these expert models into compact models, though it requires extensive supervision. In contrast, reinforcement learning reduces the need for costly supervision but risks misalignment of condition nodes with their intended semantics as well as poor credit assignment. To address these challenges, we introduce CERL (Condition-node Expert-regularized Reinforcement Learning), a framework that leverages expert-regularized reinforcement learning to preserve semantic faithfulness, while employing a factorized policy that aggregates sequential condition-node decisions into a single decision unit to alleviate credit assignment challenges. Experiments across seven tasks from the GymCards, FrozenLake, and BabyAIText suites demonstrate that our framework outperforms pure imitation learning or reinforcement learning baselines, retains strong agreement with expert decisions, and achieves substantial gains in inference speed and model size over expert models. Our implementation is available in https://github.com/HyosikMoon/CERL.
%R 10.18653/v1/2026.acl-long.1935
%U https://aclanthology.org/2026.acl-long.1935/
%U https://doi.org/10.18653/v1/2026.acl-long.1935
%P 41764-41799

Markdown (Informal)

[Lightweight and Faithful Visual Condition Checking in Behavior Trees via Expert-Regularized Reinforcement Learning](https://aclanthology.org/2026.acl-long.1935/) (Moon & Cohen, ACL 2026)

• Lightweight and Faithful Visual Condition Checking in Behavior Trees via Expert-Regularized Reinforcement Learning (Moon & Cohen, ACL 2026)

ACL

• Hyosik Moon and Eldan Cohen. 2026. Lightweight and Faithful Visual Condition Checking in Behavior Trees via Expert-Regularized Reinforcement Learning. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 41764–41799, San Diego, California, United States. Association for Computational Linguistics.