@inproceedings{ganguli-etal-2026-lexmachina,
title = "{L}ex{M}achina at {S}em{E}val-2026 Task 2: Predicting Variation in Emotional Valence and Arousal over Time from Ecological Essays",
author = "Ganguli, Somdev and
Dutta, Vibhan and
Datta, Romit and
Barman, Amit and
Naskar, Sudip",
editor = "Kochmar, Ekaterina and
Ghosh, Debanjan and
North, Kai and
Komachi, Mamoru",
booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
month = jul,
year = "2026",
address = "San Diego, California, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.semeval-1.79/",
pages = "553--560",
ISBN = "979-8-89176-414-9",
abstract = "Tracking emotional dynamics like valence and arousal is critical for understanding users' affective baselines in ecological text. However, encoder models often struggle to distinguish stable user traits from dynamic shifts, leading to poor generalization. This paper presents LexMachina, our system for SemEval-2026 Task 2, addressing ``domain shift'' and ``regression to the mean.'' LexMachina utilizes a DeBERTa-v3-Base backbone with a bifurcated strategy: post-hoc Isotonic Regression for valence calibration and a Domain Adversarial Neural Network (DANN) to mitigate user-bias in arousal. LexMachina achieved composite scores of r=0.645 (Valence) and r=0.434 (Arousal), demonstrating that adversarial disentanglement effectively captures nuances in longitudinal affective data."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ganguli-etal-2026-lexmachina">
<titleInfo>
<title>LexMachina at SemEval-2026 Task 2: Predicting Variation in Emotional Valence and Arousal over Time from Ecological Essays</title>
</titleInfo>
<name type="personal">
<namePart type="given">Somdev</namePart>
<namePart type="family">Ganguli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vibhan</namePart>
<namePart type="family">Dutta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Romit</namePart>
<namePart type="family">Datta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Amit</namePart>
<namePart type="family">Barman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sudip</namePart>
<namePart type="family">Naskar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 20th International Workshop on Semantic Evaluation (2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Kochmar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Debanjan</namePart>
<namePart type="family">Ghosh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">North</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mamoru</namePart>
<namePart type="family">Komachi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-414-9</identifier>
</relatedItem>
<abstract>Tracking emotional dynamics like valence and arousal is critical for understanding users’ affective baselines in ecological text. However, encoder models often struggle to distinguish stable user traits from dynamic shifts, leading to poor generalization. This paper presents LexMachina, our system for SemEval-2026 Task 2, addressing “domain shift” and “regression to the mean.” LexMachina utilizes a DeBERTa-v3-Base backbone with a bifurcated strategy: post-hoc Isotonic Regression for valence calibration and a Domain Adversarial Neural Network (DANN) to mitigate user-bias in arousal. LexMachina achieved composite scores of r=0.645 (Valence) and r=0.434 (Arousal), demonstrating that adversarial disentanglement effectively captures nuances in longitudinal affective data.</abstract>
<identifier type="citekey">ganguli-etal-2026-lexmachina</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.79/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>553</start>
<end>560</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LexMachina at SemEval-2026 Task 2: Predicting Variation in Emotional Valence and Arousal over Time from Ecological Essays
%A Ganguli, Somdev
%A Dutta, Vibhan
%A Datta, Romit
%A Barman, Amit
%A Naskar, Sudip
%Y Kochmar, Ekaterina
%Y Ghosh, Debanjan
%Y North, Kai
%Y Komachi, Mamoru
%S Proceedings of the 20th International Workshop on Semantic Evaluation (2026)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, USA
%@ 979-8-89176-414-9
%F ganguli-etal-2026-lexmachina
%X Tracking emotional dynamics like valence and arousal is critical for understanding users’ affective baselines in ecological text. However, encoder models often struggle to distinguish stable user traits from dynamic shifts, leading to poor generalization. This paper presents LexMachina, our system for SemEval-2026 Task 2, addressing “domain shift” and “regression to the mean.” LexMachina utilizes a DeBERTa-v3-Base backbone with a bifurcated strategy: post-hoc Isotonic Regression for valence calibration and a Domain Adversarial Neural Network (DANN) to mitigate user-bias in arousal. LexMachina achieved composite scores of r=0.645 (Valence) and r=0.434 (Arousal), demonstrating that adversarial disentanglement effectively captures nuances in longitudinal affective data.
%U https://aclanthology.org/2026.semeval-1.79/
%P 553-560
Markdown (Informal)
[LexMachina at SemEval-2026 Task 2: Predicting Variation in Emotional Valence and Arousal over Time from Ecological Essays](https://aclanthology.org/2026.semeval-1.79/) (Ganguli et al., SemEval 2026)
ACL