@inproceedings{fadeev-etal-2025-latte,
title = "{LATTE}: Learning Aligned Transactions and Textual Embeddings for Bank Clients",
author = "Fadeev, Egor and
Mollaev, Dzhambulat and
Shestov, Aleksei and
Korolev, Dima and
Zoloev, Omar and
Kireev, Ivan A and
Savchenko, Andrey and
Makarenko, Maksim",
editor = "Potdar, Saloni and
Rojas-Barahona, Lina and
Montella, Sebastien",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = nov,
year = "2025",
address = "Suzhou (China)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-industry.179/",
pages = "2635--2647",
ISBN = "979-8-89176-333-3",
abstract = "Learning clients embeddings from sequences of their historic communications is central to financial applications. While large language models (LLMs) offer general world knowledge, their direct use on long event sequences is computationally expensive and impractical in real-world pipelines. In this paper, we propose , a contrastive learning framework that aligns raw event embeddings with description-based semantic embeddings from frozen LLMs. Behavioral features based on statistical user descriptions are summarized into short prompts, embedded by the LLM, and used as supervision via contrastive loss. The proposed approach significantly reduces inference cost and input size compared to the conventional processing of complete sequences by LLM. We experimentally show that our method outperforms state-of-the-art techniques for learning event sequence representations on real-world financial datasets while remaining deployable in latency-sensitive environments."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="fadeev-etal-2025-latte">
<titleInfo>
<title>LATTE: Learning Aligned Transactions and Textual Embeddings for Bank Clients</title>
</titleInfo>
<name type="personal">
<namePart type="given">Egor</namePart>
<namePart type="family">Fadeev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dzhambulat</namePart>
<namePart type="family">Mollaev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aleksei</namePart>
<namePart type="family">Shestov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dima</namePart>
<namePart type="family">Korolev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Omar</namePart>
<namePart type="family">Zoloev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="given">A</namePart>
<namePart type="family">Kireev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrey</namePart>
<namePart type="family">Savchenko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maksim</namePart>
<namePart type="family">Makarenko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Saloni</namePart>
<namePart type="family">Potdar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lina</namePart>
<namePart type="family">Rojas-Barahona</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sebastien</namePart>
<namePart type="family">Montella</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou (China)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-333-3</identifier>
</relatedItem>
<abstract>Learning clients embeddings from sequences of their historic communications is central to financial applications. While large language models (LLMs) offer general world knowledge, their direct use on long event sequences is computationally expensive and impractical in real-world pipelines. In this paper, we propose , a contrastive learning framework that aligns raw event embeddings with description-based semantic embeddings from frozen LLMs. Behavioral features based on statistical user descriptions are summarized into short prompts, embedded by the LLM, and used as supervision via contrastive loss. The proposed approach significantly reduces inference cost and input size compared to the conventional processing of complete sequences by LLM. We experimentally show that our method outperforms state-of-the-art techniques for learning event sequence representations on real-world financial datasets while remaining deployable in latency-sensitive environments.</abstract>
<identifier type="citekey">fadeev-etal-2025-latte</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-industry.179/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>2635</start>
<end>2647</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LATTE: Learning Aligned Transactions and Textual Embeddings for Bank Clients
%A Fadeev, Egor
%A Mollaev, Dzhambulat
%A Shestov, Aleksei
%A Korolev, Dima
%A Zoloev, Omar
%A Kireev, Ivan A.
%A Savchenko, Andrey
%A Makarenko, Maksim
%Y Potdar, Saloni
%Y Rojas-Barahona, Lina
%Y Montella, Sebastien
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou (China)
%@ 979-8-89176-333-3
%F fadeev-etal-2025-latte
%X Learning clients embeddings from sequences of their historic communications is central to financial applications. While large language models (LLMs) offer general world knowledge, their direct use on long event sequences is computationally expensive and impractical in real-world pipelines. In this paper, we propose , a contrastive learning framework that aligns raw event embeddings with description-based semantic embeddings from frozen LLMs. Behavioral features based on statistical user descriptions are summarized into short prompts, embedded by the LLM, and used as supervision via contrastive loss. The proposed approach significantly reduces inference cost and input size compared to the conventional processing of complete sequences by LLM. We experimentally show that our method outperforms state-of-the-art techniques for learning event sequence representations on real-world financial datasets while remaining deployable in latency-sensitive environments.
%U https://aclanthology.org/2025.emnlp-industry.179/
%P 2635-2647
Markdown (Informal)
[LATTE: Learning Aligned Transactions and Textual Embeddings for Bank Clients](https://aclanthology.org/2025.emnlp-industry.179/) (Fadeev et al., EMNLP 2025)
ACL
- Egor Fadeev, Dzhambulat Mollaev, Aleksei Shestov, Dima Korolev, Omar Zoloev, Ivan A Kireev, Andrey Savchenko, and Maksim Makarenko. 2025. LATTE: Learning Aligned Transactions and Textual Embeddings for Bank Clients. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 2635–2647, Suzhou (China). Association for Computational Linguistics.