@inproceedings{askari-etal-2026-avoidance,
title = "Avoidance Learning of Failure Modes Produces Robust Tool Planners",
author = "Askari, Hadi and
Singh, Japinder and
Huang, Tenghao and
Jung, Dongwon and
Chhabra, Anshuman and
Chen, Muhao",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1486/",
pages = "31641--31658",
ISBN = "979-8-89176-462-0",
abstract = "Multi-step tool planning is vulnerable to structural errors that propagate across dependent calls: a wrong argument, broken dependency, or mis-ordered operation can invalidate an otherwise plausible trajectory. We introduce FORTE, a framework that makes these failure modes explicit training targets. FORTE defines a nine-error taxonomy for multi-step plans, constructs taxonomy-grounded contrastive corpora by injecting minimally edited failures into gold trajectories, and trains a lightweight judge to recognize and score these errors. The same frozen judge then provides dense reward for GRPO planner optimization and verification for Best-of-N inference. Across ToolHop and NESTFUL, FORTE achieves the highest Judge Success Rate across all six open-weight backbone-dataset configurations and the highest observed end-task execution accuracy among the evaluated methods in all six settings. With Qwen2.5-7B, FORTE reaches 96.6 JSR and 99.5 dependency accuracy on ToolHop, and 98.6 JSR and 89.0 dependency accuracy on NESTFUL. Under a matched five-sample selection budget, FORTE-Qwen2.5-7B also outperforms GPT-5.5 across JSR, functional match, parameter accuracy, and dependency accuracy on both benchmarks. Ablations show that taxonomy-grounded contrastive negatives are essential for both judge discrimination and effective inference-time selection. These results demonstrate that explicitly learning which planning failures to avoid provides a strong supervision signal for robust multi-step tool planning."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="askari-etal-2026-avoidance">
<titleInfo>
<title>Avoidance Learning of Failure Modes Produces Robust Tool Planners</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hadi</namePart>
<namePart type="family">Askari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Japinder</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tenghao</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dongwon</namePart>
<namePart type="family">Jung</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anshuman</namePart>
<namePart type="family">Chhabra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Muhao</namePart>
<namePart type="family">Chen</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Multi-step tool planning is vulnerable to structural errors that propagate across dependent calls: a wrong argument, broken dependency, or mis-ordered operation can invalidate an otherwise plausible trajectory. We introduce FORTE, a framework that makes these failure modes explicit training targets. FORTE defines a nine-error taxonomy for multi-step plans, constructs taxonomy-grounded contrastive corpora by injecting minimally edited failures into gold trajectories, and trains a lightweight judge to recognize and score these errors. The same frozen judge then provides dense reward for GRPO planner optimization and verification for Best-of-N inference. Across ToolHop and NESTFUL, FORTE achieves the highest Judge Success Rate across all six open-weight backbone-dataset configurations and the highest observed end-task execution accuracy among the evaluated methods in all six settings. With Qwen2.5-7B, FORTE reaches 96.6 JSR and 99.5 dependency accuracy on ToolHop, and 98.6 JSR and 89.0 dependency accuracy on NESTFUL. Under a matched five-sample selection budget, FORTE-Qwen2.5-7B also outperforms GPT-5.5 across JSR, functional match, parameter accuracy, and dependency accuracy on both benchmarks. Ablations show that taxonomy-grounded contrastive negatives are essential for both judge discrimination and effective inference-time selection. These results demonstrate that explicitly learning which planning failures to avoid provides a strong supervision signal for robust multi-step tool planning.</abstract>
<identifier type="citekey">askari-etal-2026-avoidance</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1486/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>31641</start>
<end>31658</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Avoidance Learning of Failure Modes Produces Robust Tool Planners
%A Askari, Hadi
%A Singh, Japinder
%A Huang, Tenghao
%A Jung, Dongwon
%A Chhabra, Anshuman
%A Chen, Muhao
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F askari-etal-2026-avoidance
%X Multi-step tool planning is vulnerable to structural errors that propagate across dependent calls: a wrong argument, broken dependency, or mis-ordered operation can invalidate an otherwise plausible trajectory. We introduce FORTE, a framework that makes these failure modes explicit training targets. FORTE defines a nine-error taxonomy for multi-step plans, constructs taxonomy-grounded contrastive corpora by injecting minimally edited failures into gold trajectories, and trains a lightweight judge to recognize and score these errors. The same frozen judge then provides dense reward for GRPO planner optimization and verification for Best-of-N inference. Across ToolHop and NESTFUL, FORTE achieves the highest Judge Success Rate across all six open-weight backbone-dataset configurations and the highest observed end-task execution accuracy among the evaluated methods in all six settings. With Qwen2.5-7B, FORTE reaches 96.6 JSR and 99.5 dependency accuracy on ToolHop, and 98.6 JSR and 89.0 dependency accuracy on NESTFUL. Under a matched five-sample selection budget, FORTE-Qwen2.5-7B also outperforms GPT-5.5 across JSR, functional match, parameter accuracy, and dependency accuracy on both benchmarks. Ablations show that taxonomy-grounded contrastive negatives are essential for both judge discrimination and effective inference-time selection. These results demonstrate that explicitly learning which planning failures to avoid provides a strong supervision signal for robust multi-step tool planning.
%U https://aclanthology.org/2026.findings-emnlp.1486/
%P 31641-31658
Markdown (Informal)
[Avoidance Learning of Failure Modes Produces Robust Tool Planners](https://aclanthology.org/2026.findings-emnlp.1486/) (Askari et al., Findings 2026)
ACL