@inproceedings{osei-brefo-liang-2026-oseibrefo,
title = "{O}sei{B}refo-Liang at {S}em{E}val-2026 Task 12: Hybrid Causal Knowledge Graphs and Neural-Symbolic Policy Optimisation for Abductive Event Reasoning",
author = "Osei-Brefo, Emmanuel and
Liang, Huizhi(elly)",
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-1.287/",
pages = "2268--2274",
ISBN = "979-8-89176-414-9",
abstract = "Abductive Event Reasoning (AER) requires selecting plausible causal explanations for observed events from incomplete and noisy textual evidence. Unlike deductive reasoning, abductive inference proceeds from effects to candidate causes and is highly sensitive to distractor information and implicit multi-hop relationships. We present a hybrid neural-symbolic framework that models abductive reasoning as structured causal validation rather than unconstrained generation. Our framework integrates hybrid retrieval, micro-level evidence grounding, concept-level causal abstraction, reinforcement learning-based decision calibration, and structured Theorem-of-Thought verification. Experiments on SemEval-2026 Task 12 show that LLM reasoning constrained by structured causal graphs achieves the strongest development performance of 0.5288 and a leaderboard score of 0.61 on the test set, substantially outperforming symbolic-only and policy-only variants. These findings indicate that explicit causal modelling improves robustness in document-grounded abduction tasks."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="osei-brefo-liang-2026-oseibrefo">
<titleInfo>
<title>OseiBrefo-Liang at SemEval-2026 Task 12: Hybrid Causal Knowledge Graphs and Neural-Symbolic Policy Optimisation for Abductive Event Reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Emmanuel</namePart>
<namePart type="family">Osei-Brefo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Huizhi(elly)</namePart>
<namePart type="family">Liang</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>Abductive Event Reasoning (AER) requires selecting plausible causal explanations for observed events from incomplete and noisy textual evidence. Unlike deductive reasoning, abductive inference proceeds from effects to candidate causes and is highly sensitive to distractor information and implicit multi-hop relationships. We present a hybrid neural-symbolic framework that models abductive reasoning as structured causal validation rather than unconstrained generation. Our framework integrates hybrid retrieval, micro-level evidence grounding, concept-level causal abstraction, reinforcement learning-based decision calibration, and structured Theorem-of-Thought verification. Experiments on SemEval-2026 Task 12 show that LLM reasoning constrained by structured causal graphs achieves the strongest development performance of 0.5288 and a leaderboard score of 0.61 on the test set, substantially outperforming symbolic-only and policy-only variants. These findings indicate that explicit causal modelling improves robustness in document-grounded abduction tasks.</abstract>
<identifier type="citekey">osei-brefo-liang-2026-oseibrefo</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.287/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>2268</start>
<end>2274</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T OseiBrefo-Liang at SemEval-2026 Task 12: Hybrid Causal Knowledge Graphs and Neural-Symbolic Policy Optimisation for Abductive Event Reasoning
%A Osei-Brefo, Emmanuel
%A Liang, Huizhi(elly)
%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 osei-brefo-liang-2026-oseibrefo
%X Abductive Event Reasoning (AER) requires selecting plausible causal explanations for observed events from incomplete and noisy textual evidence. Unlike deductive reasoning, abductive inference proceeds from effects to candidate causes and is highly sensitive to distractor information and implicit multi-hop relationships. We present a hybrid neural-symbolic framework that models abductive reasoning as structured causal validation rather than unconstrained generation. Our framework integrates hybrid retrieval, micro-level evidence grounding, concept-level causal abstraction, reinforcement learning-based decision calibration, and structured Theorem-of-Thought verification. Experiments on SemEval-2026 Task 12 show that LLM reasoning constrained by structured causal graphs achieves the strongest development performance of 0.5288 and a leaderboard score of 0.61 on the test set, substantially outperforming symbolic-only and policy-only variants. These findings indicate that explicit causal modelling improves robustness in document-grounded abduction tasks.
%U https://aclanthology.org/2026.semeval-1.287/
%P 2268-2274
Markdown (Informal)
[OseiBrefo-Liang at SemEval-2026 Task 12: Hybrid Causal Knowledge Graphs and Neural-Symbolic Policy Optimisation for Abductive Event Reasoning](https://aclanthology.org/2026.semeval-1.287/) (Osei-Brefo & Liang, SemEval 2026)
ACL