@inproceedings{sun-etal-2026-toolsmith,
title = "{T}ool{S}mith: An Automatic Tool-Generation Framework for Multimodal Search Agents",
author = "Sun, Rui and
Huang, Junhua and
Zancato, Luca and
Islam, Md Mofijul and
Chang, Kai-Wei",
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.1699/",
pages = "36305--36324",
ISBN = "979-8-89176-462-0",
abstract = "Recent multimodal search agents use reasoning, vision, and live web search across many turns to answer questions beyond parametric knowledge by orchestrating a small library of hand-crafted search and visual tools. However, the library is laborious to create and maintain, and frozen when new query types, evidence formats, and failure modes are encountered in the wild. Human practitioners, by contrast, cope with insufficient tools by fashioning new ones from familiar parts and reusing them across tasks. Inspired by this, we propose ToolSmith, an automatic tool-generation framework that grows and reuses modules on top of a lean base tool library. We evaluate ToolSmith on three in-domain multimodal search benchmarks and ten out-of-domain benchmarks under a VLM-as-a-Judge protocol. ToolSmith consistently improves over the base library across deployed agents of varying scales, with larger backbones reaping greater gains, demonstrating the scalability of the approach. Our code and data are available at https://threesr.github.io/toolsmith/."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sun-etal-2026-toolsmith">
<titleInfo>
<title>ToolSmith: An Automatic Tool-Generation Framework for Multimodal Search Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junhua</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Luca</namePart>
<namePart type="family">Zancato</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Md</namePart>
<namePart type="given">Mofijul</namePart>
<namePart type="family">Islam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai-Wei</namePart>
<namePart type="family">Chang</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>Recent multimodal search agents use reasoning, vision, and live web search across many turns to answer questions beyond parametric knowledge by orchestrating a small library of hand-crafted search and visual tools. However, the library is laborious to create and maintain, and frozen when new query types, evidence formats, and failure modes are encountered in the wild. Human practitioners, by contrast, cope with insufficient tools by fashioning new ones from familiar parts and reusing them across tasks. Inspired by this, we propose ToolSmith, an automatic tool-generation framework that grows and reuses modules on top of a lean base tool library. We evaluate ToolSmith on three in-domain multimodal search benchmarks and ten out-of-domain benchmarks under a VLM-as-a-Judge protocol. ToolSmith consistently improves over the base library across deployed agents of varying scales, with larger backbones reaping greater gains, demonstrating the scalability of the approach. Our code and data are available at https://threesr.github.io/toolsmith/.</abstract>
<identifier type="citekey">sun-etal-2026-toolsmith</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1699/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>36305</start>
<end>36324</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ToolSmith: An Automatic Tool-Generation Framework for Multimodal Search Agents
%A Sun, Rui
%A Huang, Junhua
%A Zancato, Luca
%A Islam, Md Mofijul
%A Chang, Kai-Wei
%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 sun-etal-2026-toolsmith
%X Recent multimodal search agents use reasoning, vision, and live web search across many turns to answer questions beyond parametric knowledge by orchestrating a small library of hand-crafted search and visual tools. However, the library is laborious to create and maintain, and frozen when new query types, evidence formats, and failure modes are encountered in the wild. Human practitioners, by contrast, cope with insufficient tools by fashioning new ones from familiar parts and reusing them across tasks. Inspired by this, we propose ToolSmith, an automatic tool-generation framework that grows and reuses modules on top of a lean base tool library. We evaluate ToolSmith on three in-domain multimodal search benchmarks and ten out-of-domain benchmarks under a VLM-as-a-Judge protocol. ToolSmith consistently improves over the base library across deployed agents of varying scales, with larger backbones reaping greater gains, demonstrating the scalability of the approach. Our code and data are available at https://threesr.github.io/toolsmith/.
%U https://aclanthology.org/2026.findings-emnlp.1699/
%P 36305-36324
Markdown (Informal)
[ToolSmith: An Automatic Tool-Generation Framework for Multimodal Search Agents](https://aclanthology.org/2026.findings-emnlp.1699/) (Sun et al., Findings 2026)
ACL