@inproceedings{tuli-etal-2026-mindflayer-semeval,
title = "{M}ind{F}layer at {S}em{E}val-2026 Task 13:{LACR}-{ENS}: Calibration-Aware Ensemble Routing for Cross-Language {AI}-Generated Code Detection",
author = "Tuli, Jerin Romijah and
Naem, Talukder Naemul Hasan and
Pritom, Md. Sartaj Alam",
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-1.294/",
pages = "2322--2329",
ISBN = "979-8-89176-414-9",
abstract = "This paper presents LACR-ENS, a calibration-aware ensemble system for detecting AI-generated code across eight programming languages (SemEval-2026 Task 13). We identify a severe asymmetric out-of-distribution (OOD) failure in fine-tuned code transformers: Expected Calibration Error doubles from 0.09 (seen languages) to 0.18 (unseen languages), and high-confidence predictions (p0.80) are wrong 39{\%} of the time on OOD inputs. We propose Language-Aware Confidence Routing (LACR), formally equivalent to implicit per-language temperature scaling, which reduces OOD ECE to 0.11 and improves macro-F1 by +0.013 over fixed-threshold ensembling. A language-family proximity analysis reveals that syntactic distance to training languages predicts OOD F1 with Pearson r=+0.94, providing a principled, label-free signal for deployment risk assessment and motivating a continuous routing extension. Our system combines UniXCoder and GraphCodeBERT via weighted logit-level fusion and achieves macro-F1 0.538 , outperforming comparable encoder-only systems. We additionally document a HuggingFace label inversion pitfall that suppressed our initial score by approximately 0.29 F1."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tuli-etal-2026-mindflayer-semeval">
<titleInfo>
<title>MindFlayer at SemEval-2026 Task 13:LACR-ENS: Calibration-Aware Ensemble Routing for Cross-Language AI-Generated Code Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jerin</namePart>
<namePart type="given">Romijah</namePart>
<namePart type="family">Tuli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Talukder</namePart>
<namePart type="given">Naemul</namePart>
<namePart type="given">Hasan</namePart>
<namePart type="family">Naem</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Md.</namePart>
<namePart type="given">Sartaj</namePart>
<namePart type="given">Alam</namePart>
<namePart type="family">Pritom</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 presents LACR-ENS, a calibration-aware ensemble system for detecting AI-generated code across eight programming languages (SemEval-2026 Task 13). We identify a severe asymmetric out-of-distribution (OOD) failure in fine-tuned code transformers: Expected Calibration Error doubles from 0.09 (seen languages) to 0.18 (unseen languages), and high-confidence predictions (p0.80) are wrong 39% of the time on OOD inputs. We propose Language-Aware Confidence Routing (LACR), formally equivalent to implicit per-language temperature scaling, which reduces OOD ECE to 0.11 and improves macro-F1 by +0.013 over fixed-threshold ensembling. A language-family proximity analysis reveals that syntactic distance to training languages predicts OOD F1 with Pearson r=+0.94, providing a principled, label-free signal for deployment risk assessment and motivating a continuous routing extension. Our system combines UniXCoder and GraphCodeBERT via weighted logit-level fusion and achieves macro-F1 0.538 , outperforming comparable encoder-only systems. We additionally document a HuggingFace label inversion pitfall that suppressed our initial score by approximately 0.29 F1.</abstract>
<identifier type="citekey">tuli-etal-2026-mindflayer-semeval</identifier>
<location>
<url>https://aclanthology.org/2026.semeval-1.294/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>2322</start>
<end>2329</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T MindFlayer at SemEval-2026 Task 13:LACR-ENS: Calibration-Aware Ensemble Routing for Cross-Language AI-Generated Code Detection
%A Tuli, Jerin Romijah
%A Naem, Talukder Naemul Hasan
%A Pritom, Md. Sartaj Alam
%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 tuli-etal-2026-mindflayer-semeval
%X This paper presents LACR-ENS, a calibration-aware ensemble system for detecting AI-generated code across eight programming languages (SemEval-2026 Task 13). We identify a severe asymmetric out-of-distribution (OOD) failure in fine-tuned code transformers: Expected Calibration Error doubles from 0.09 (seen languages) to 0.18 (unseen languages), and high-confidence predictions (p0.80) are wrong 39% of the time on OOD inputs. We propose Language-Aware Confidence Routing (LACR), formally equivalent to implicit per-language temperature scaling, which reduces OOD ECE to 0.11 and improves macro-F1 by +0.013 over fixed-threshold ensembling. A language-family proximity analysis reveals that syntactic distance to training languages predicts OOD F1 with Pearson r=+0.94, providing a principled, label-free signal for deployment risk assessment and motivating a continuous routing extension. Our system combines UniXCoder and GraphCodeBERT via weighted logit-level fusion and achieves macro-F1 0.538 , outperforming comparable encoder-only systems. We additionally document a HuggingFace label inversion pitfall that suppressed our initial score by approximately 0.29 F1.
%U https://aclanthology.org/2026.semeval-1.294/
%P 2322-2329
Markdown (Informal)
[MindFlayer at SemEval-2026 Task 13:LACR-ENS: Calibration-Aware Ensemble Routing for Cross-Language AI-Generated Code Detection](https://aclanthology.org/2026.semeval-1.294/) (Tuli et al., SemEval 2026)
ACL