@inproceedings{terao-tachioka-2026-itlab,
title = "d-itlab at {S}em{E}val-2026 Task 12: Per-Option Surprisal and Multi-Stage Gating for Precision-Oriented Causal Reasoning",
author = "Terao, Yasunori and
Tachioka, Yuuki",
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-1.228/",
pages = "1791--1801",
ISBN = "979-8-89176-414-9",
abstract = "We describe the system submitted by d-itlab to SemEval-2026 Task{\textasciitilde}12 (Abductive Event Reasoning), which requires selecting the most plausible direct cause(s) of an observed event from candidate options grounded in reference documents. Our approach combines (i) per-option multi-stage LLM inference that evaluates each option independently with progressively stricter verification, (ii) surprisal-based features obtained by teacher-forcing candidate sentences and measuring token-level negative log-likelihood, and (iii) an XGBoost ensemble trained on these heterogeneous features to produce a precision-oriented final prediction. In the official test set, our system scored 0.91, ranking third among 116 participating teams."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="terao-tachioka-2026-itlab">
<titleInfo>
<title>d-itlab at SemEval-2026 Task 12: Per-Option Surprisal and Multi-Stage Gating for Precision-Oriented Causal Reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yasunori</namePart>
<namePart type="family">Terao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuuki</namePart>
<namePart type="family">Tachioka</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>We describe the system submitted by d-itlab to SemEval-2026 Task~12 (Abductive Event Reasoning), which requires selecting the most plausible direct cause(s) of an observed event from candidate options grounded in reference documents. Our approach combines (i) per-option multi-stage LLM inference that evaluates each option independently with progressively stricter verification, (ii) surprisal-based features obtained by teacher-forcing candidate sentences and measuring token-level negative log-likelihood, and (iii) an XGBoost ensemble trained on these heterogeneous features to produce a precision-oriented final prediction. In the official test set, our system scored 0.91, ranking third among 116 participating teams.</abstract>
<identifier type="citekey">terao-tachioka-2026-itlab</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.228/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>1791</start>
<end>1801</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T d-itlab at SemEval-2026 Task 12: Per-Option Surprisal and Multi-Stage Gating for Precision-Oriented Causal Reasoning
%A Terao, Yasunori
%A Tachioka, Yuuki
%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 terao-tachioka-2026-itlab
%X We describe the system submitted by d-itlab to SemEval-2026 Task~12 (Abductive Event Reasoning), which requires selecting the most plausible direct cause(s) of an observed event from candidate options grounded in reference documents. Our approach combines (i) per-option multi-stage LLM inference that evaluates each option independently with progressively stricter verification, (ii) surprisal-based features obtained by teacher-forcing candidate sentences and measuring token-level negative log-likelihood, and (iii) an XGBoost ensemble trained on these heterogeneous features to produce a precision-oriented final prediction. In the official test set, our system scored 0.91, ranking third among 116 participating teams.
%U https://aclanthology.org/2026.semeval-1.228/
%P 1791-1801
Markdown (Informal)
[d-itlab at SemEval-2026 Task 12: Per-Option Surprisal and Multi-Stage Gating for Precision-Oriented Causal Reasoning](https://aclanthology.org/2026.semeval-1.228/) (Terao & Tachioka, SemEval 2026)
ACL