@inproceedings{rahbar-etal-2026-topic,
title = "Topic Matching in the Wild: Benchmark and Lessons from Real-World {ASR} Transcripts",
author = "Rahbar, Saman and
Zhu, Xiliang and
Cardoza, Irvin and
Rossouw, David",
editor = "Bak, JinYeong and
van der Goot, Rob and
Jang, Hyeju and
Buaphet, Weerayut and
Ramponi, Alan and
Xu, Wei and
Ritter, Alan",
booktitle = "Proceedings of the 11th Workshop on Natural User-generated Text",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.wnut-1.23/",
pages = "332--340",
ISBN = "979-8-89176-470-5",
abstract = "In contact centers, real-time agent-assist tools determine, for each of many predefined topics, whether a live customer utterance is relevant and display a coaching card to the agent when it is. The input is noisy and challenging: ASR (Automatic Speech Recognition) transcripts of spontaneous phone conversations, which can be unclear, repetitive, and mostly lack punctuation. To systematically study this real-world task, we curate a human-annotated topic-utterance judgments dataset sourced from real call-center transcripts. We compare three types of matchers: a regex-based baseline, zero-shot sentence-embedding encoders, and Gemini-based LLM matchers. In addition, two types of topic representations are studied in our benchmark: keyphrases and natural language description. Our empirical experiments highlight the superior performance of lightweight LLM matchers when equipped with natural language descriptions."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rahbar-etal-2026-topic">
<titleInfo>
<title>Topic Matching in the Wild: Benchmark and Lessons from Real-World ASR Transcripts</title>
</titleInfo>
<name type="personal">
<namePart type="given">Saman</namePart>
<namePart type="family">Rahbar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiliang</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Irvin</namePart>
<namePart type="family">Cardoza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Rossouw</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 11th Workshop on Natural User-generated Text</title>
</titleInfo>
<name type="personal">
<namePart type="given">JinYeong</namePart>
<namePart type="family">Bak</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rob</namePart>
<namePart type="family">van der Goot</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hyeju</namePart>
<namePart type="family">Jang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weerayut</namePart>
<namePart type="family">Buaphet</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ramponi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</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">979-8-89176-470-5</identifier>
</relatedItem>
<abstract>In contact centers, real-time agent-assist tools determine, for each of many predefined topics, whether a live customer utterance is relevant and display a coaching card to the agent when it is. The input is noisy and challenging: ASR (Automatic Speech Recognition) transcripts of spontaneous phone conversations, which can be unclear, repetitive, and mostly lack punctuation. To systematically study this real-world task, we curate a human-annotated topic-utterance judgments dataset sourced from real call-center transcripts. We compare three types of matchers: a regex-based baseline, zero-shot sentence-embedding encoders, and Gemini-based LLM matchers. In addition, two types of topic representations are studied in our benchmark: keyphrases and natural language description. Our empirical experiments highlight the superior performance of lightweight LLM matchers when equipped with natural language descriptions.</abstract>
<identifier type="citekey">rahbar-etal-2026-topic</identifier>
<location>
<url>https://aclanthology.org/2026.wnut-1.23/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>332</start>
<end>340</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Topic Matching in the Wild: Benchmark and Lessons from Real-World ASR Transcripts
%A Rahbar, Saman
%A Zhu, Xiliang
%A Cardoza, Irvin
%A Rossouw, David
%Y Bak, JinYeong
%Y van der Goot, Rob
%Y Jang, Hyeju
%Y Buaphet, Weerayut
%Y Ramponi, Alan
%Y Xu, Wei
%Y Ritter, Alan
%S Proceedings of the 11th Workshop on Natural User-generated Text
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-470-5
%F rahbar-etal-2026-topic
%X In contact centers, real-time agent-assist tools determine, for each of many predefined topics, whether a live customer utterance is relevant and display a coaching card to the agent when it is. The input is noisy and challenging: ASR (Automatic Speech Recognition) transcripts of spontaneous phone conversations, which can be unclear, repetitive, and mostly lack punctuation. To systematically study this real-world task, we curate a human-annotated topic-utterance judgments dataset sourced from real call-center transcripts. We compare three types of matchers: a regex-based baseline, zero-shot sentence-embedding encoders, and Gemini-based LLM matchers. In addition, two types of topic representations are studied in our benchmark: keyphrases and natural language description. Our empirical experiments highlight the superior performance of lightweight LLM matchers when equipped with natural language descriptions.
%U https://aclanthology.org/2026.wnut-1.23/
%P 332-340
Markdown (Informal)
[Topic Matching in the Wild: Benchmark and Lessons from Real-World ASR Transcripts](https://aclanthology.org/2026.wnut-1.23/) (Rahbar et al., WNUT 2026)
ACL