@inproceedings{ballal-v-r-2026-grounding,
title = "Grounding First: Auditing a Structured Recovery Wrapper for Local Open-Weight Reflexion Agents",
author = "Ballal, Nimay and
V R, Badri Prasad",
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-1.52/",
pages = "774--784",
ISBN = "979-8-89176-485-9",
abstract = "We audit how action interfaces and recovery mechanisms shape the evaluation of local open-weight Reflexion agents on 134 ALFWorld valid-unseen tasks. The analysis covers six primary admissible-action conditions and historical free-text runs across three model deployments. We organize the audit around four implementation contracts: action execution, recovery triggering, treatment changes, and outcome construction. Historical free-text runs show substantial invalid-output fallback, while designated admissible-parser fallback remains below 1.1 percent; this contrast does not by itself establish parser faithfulness. Structured Recovery Prompting (SRP) combines a recovery prompt with counter resets and relaxed termination limits. Its implementation-reported task-success differences range from 0.00 to +6.72 percentage points and are smaller than the baseline contrast between the two Qwen model sizes. The available evidence supports descriptive comparisons rather than an isolated recovery-prompt effect. The audit identifies incomplete repeat detection and termination-based success labeling, highlighting the importance of trigger coverage and outcome definitions when interpreting recovery results. The contribution is a focused diagnostic audit that connects aggregate performance to implementation behavior and provides concrete checks for evaluating recovery in the studied agents."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ballal-v-r-2026-grounding">
<titleInfo>
<title>Grounding First: Auditing a Structured Recovery Wrapper for Local Open-Weight Reflexion Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nimay</namePart>
<namePart type="family">Ballal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Badri</namePart>
<namePart type="given">Prasad</namePart>
<namePart type="family">V R</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>We audit how action interfaces and recovery mechanisms shape the evaluation of local open-weight Reflexion agents on 134 ALFWorld valid-unseen tasks. The analysis covers six primary admissible-action conditions and historical free-text runs across three model deployments. We organize the audit around four implementation contracts: action execution, recovery triggering, treatment changes, and outcome construction. Historical free-text runs show substantial invalid-output fallback, while designated admissible-parser fallback remains below 1.1 percent; this contrast does not by itself establish parser faithfulness. Structured Recovery Prompting (SRP) combines a recovery prompt with counter resets and relaxed termination limits. Its implementation-reported task-success differences range from 0.00 to +6.72 percentage points and are smaller than the baseline contrast between the two Qwen model sizes. The available evidence supports descriptive comparisons rather than an isolated recovery-prompt effect. The audit identifies incomplete repeat detection and termination-based success labeling, highlighting the importance of trigger coverage and outcome definitions when interpreting recovery results. The contribution is a focused diagnostic audit that connects aggregate performance to implementation behavior and provides concrete checks for evaluating recovery in the studied agents.</abstract>
<identifier type="citekey">ballal-v-r-2026-grounding</identifier>
<location>
<url>https://aclanthology.org/2026.realm-1.52/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>774</start>
<end>784</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Grounding First: Auditing a Structured Recovery Wrapper for Local Open-Weight Reflexion Agents
%A Ballal, Nimay
%A V R, Badri Prasad
%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 ballal-v-r-2026-grounding
%X We audit how action interfaces and recovery mechanisms shape the evaluation of local open-weight Reflexion agents on 134 ALFWorld valid-unseen tasks. The analysis covers six primary admissible-action conditions and historical free-text runs across three model deployments. We organize the audit around four implementation contracts: action execution, recovery triggering, treatment changes, and outcome construction. Historical free-text runs show substantial invalid-output fallback, while designated admissible-parser fallback remains below 1.1 percent; this contrast does not by itself establish parser faithfulness. Structured Recovery Prompting (SRP) combines a recovery prompt with counter resets and relaxed termination limits. Its implementation-reported task-success differences range from 0.00 to +6.72 percentage points and are smaller than the baseline contrast between the two Qwen model sizes. The available evidence supports descriptive comparisons rather than an isolated recovery-prompt effect. The audit identifies incomplete repeat detection and termination-based success labeling, highlighting the importance of trigger coverage and outcome definitions when interpreting recovery results. The contribution is a focused diagnostic audit that connects aggregate performance to implementation behavior and provides concrete checks for evaluating recovery in the studied agents.
%U https://aclanthology.org/2026.realm-1.52/
%P 774-784
Markdown (Informal)
[Grounding First: Auditing a Structured Recovery Wrapper for Local Open-Weight Reflexion Agents](https://aclanthology.org/2026.realm-1.52/) (Ballal & V R, REALM 2026)
ACL