@inproceedings{zhou-etal-2026-uir,
title = "uir-cis at {S}em{E}val-2026 Task 12: Mitigating Prior-Induced Hallucinations in Retrieval-Augmented Reasoning via Precision-Oriented Decoding",
author = "Zhou, Chiyao and
Wang, Zebing and
Deng, Kexin and
Zhao, Yaru and
Deng, Lin and
Li, Binyang",
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-1.415/",
pages = "3337--3342",
ISBN = "979-8-89176-414-9",
abstract = "This paper describes our system for the SemEval-2026 Task 12 on Abductive Event Reasoning (AER). We systematically address the ``over-selection'' hallucination pathology in Instruction-tuned Large Language Models (LLMs), where models erroneously align distractors with semantic priors rather than retrieved evidence. Our framework utilizes a 32-billion parameter Qwen2.5 foundational model adapted via Low-Rank Adaptation (LoRA) and evaluated under a Zero-shot Chain-of-Thought (CoT) setting. To mitigate epistemic noise, we propose a Precision-Oriented Decoding (POD) strategy that couples low-temperature sampling (T=0.45) with scaled majority voting (K=9). Following a three-stage empirical evolution{---}from baseline diagnosis to precision optimization and ensemble analysis{---}our system achieved a score of 0.802 on the official test set. Our findings demonstrate that in causal reasoning tasks with strict penalization for incorrect predictions, epistemic noise suppression is strictly superior to heuristic recall compensation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhou-etal-2026-uir">
<titleInfo>
<title>uir-cis at SemEval-2026 Task 12: Mitigating Prior-Induced Hallucinations in Retrieval-Augmented Reasoning via Precision-Oriented Decoding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chiyao</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zebing</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kexin</namePart>
<namePart type="family">Deng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yaru</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lin</namePart>
<namePart type="family">Deng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Binyang</namePart>
<namePart type="family">Li</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>This paper describes our system for the SemEval-2026 Task 12 on Abductive Event Reasoning (AER). We systematically address the “over-selection” hallucination pathology in Instruction-tuned Large Language Models (LLMs), where models erroneously align distractors with semantic priors rather than retrieved evidence. Our framework utilizes a 32-billion parameter Qwen2.5 foundational model adapted via Low-Rank Adaptation (LoRA) and evaluated under a Zero-shot Chain-of-Thought (CoT) setting. To mitigate epistemic noise, we propose a Precision-Oriented Decoding (POD) strategy that couples low-temperature sampling (T=0.45) with scaled majority voting (K=9). Following a three-stage empirical evolution—from baseline diagnosis to precision optimization and ensemble analysis—our system achieved a score of 0.802 on the official test set. Our findings demonstrate that in causal reasoning tasks with strict penalization for incorrect predictions, epistemic noise suppression is strictly superior to heuristic recall compensation.</abstract>
<identifier type="citekey">zhou-etal-2026-uir</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.415/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>3337</start>
<end>3342</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T uir-cis at SemEval-2026 Task 12: Mitigating Prior-Induced Hallucinations in Retrieval-Augmented Reasoning via Precision-Oriented Decoding
%A Zhou, Chiyao
%A Wang, Zebing
%A Deng, Kexin
%A Zhao, Yaru
%A Deng, Lin
%A Li, Binyang
%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 zhou-etal-2026-uir
%X This paper describes our system for the SemEval-2026 Task 12 on Abductive Event Reasoning (AER). We systematically address the “over-selection” hallucination pathology in Instruction-tuned Large Language Models (LLMs), where models erroneously align distractors with semantic priors rather than retrieved evidence. Our framework utilizes a 32-billion parameter Qwen2.5 foundational model adapted via Low-Rank Adaptation (LoRA) and evaluated under a Zero-shot Chain-of-Thought (CoT) setting. To mitigate epistemic noise, we propose a Precision-Oriented Decoding (POD) strategy that couples low-temperature sampling (T=0.45) with scaled majority voting (K=9). Following a three-stage empirical evolution—from baseline diagnosis to precision optimization and ensemble analysis—our system achieved a score of 0.802 on the official test set. Our findings demonstrate that in causal reasoning tasks with strict penalization for incorrect predictions, epistemic noise suppression is strictly superior to heuristic recall compensation.
%U https://aclanthology.org/2026.semeval-1.415/
%P 3337-3342
Markdown (Informal)
[uir-cis at SemEval-2026 Task 12: Mitigating Prior-Induced Hallucinations in Retrieval-Augmented Reasoning via Precision-Oriented Decoding](https://aclanthology.org/2026.semeval-1.415/) (Zhou et al., SemEval 2026)
ACL