@inproceedings{silchenko-2026-thylinao-childsafeads,
title = "thylinao at {C}hild{S}afe{A}ds 2026: A Six-Member Ensemble and What Each Level of Data Access Provided",
author = "Silchenko, Maksim",
editor = "Aletras, Nikolaos and
Chalkidis, Ilias and
Barrett, Leslie and
Goanț{\u{a}}, C{\u{a}}t{\u{a}}lina and
Preoțiuc-Pietro, Daniel and
Spanakis, Gerasimos and
Xu, Shanshan",
booktitle = "Proceedings of the Natural Legal Language Processing Workshop 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.nllp-1.1/",
pages = "1--11",
ISBN = "979-8-89176-494-1",
abstract = "This paper describes the thylinao system for the ChildSafeAds shared task on monitoring sponsored segments in child-facing YouTube videos. The system is a six-member ensemble: two fine-tuned encoders, a zero-shot prompted Qwen3-32B, a LoRA-tuned Qwen3-14B used for ST2 only, a TF-IDF baseline, and a level-2 stacker for the compliance sub-task, blended at equal weight behind a decision layer whose every parameter is fitted on channel-grouped out-of-fold predictions. The system scores mean macro-F1 0.6537 on the evaluation split; inference over the 503 test instances is dominated by one measured 14-minute GPU pass of the 32B member. The zero-shot member is the most valuable and nearly the cheapest, while removing the fine-tuned DeBERTa changes nothing measurable. An ablation over the task{'}s four data access levels, run on the zero-shot member, finds the channel name adds nothing and the product-page crawl helps product identification but not, measurably, compliance. Validating these changes also surfaced properties of the evaluation, reported as feedback: the metric{'}s divisor dominates score variance and is recoverable in closed form from a published baseline row, and the benchmark is channel-disjoint but substantially advertiser-overlapping. The final submitted entry additionally contained three labels set by hand, which the task Terms prohibit; this was disclosed to the organisers and is documented in Section 7. The system-only score is reported throughout and no rank is claimed."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="silchenko-2026-thylinao-childsafeads">
<titleInfo>
<title>thylinao at ChildSafeAds 2026: A Six-Member Ensemble and What Each Level of Data Access Provided</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maksim</namePart>
<namePart type="family">Silchenko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Natural Legal Language Processing Workshop 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nikolaos</namePart>
<namePart type="family">Aletras</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ilias</namePart>
<namePart type="family">Chalkidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leslie</namePart>
<namePart type="family">Barrett</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cătălina</namePart>
<namePart type="family">Goanță</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Preoțiuc-Pietro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gerasimos</namePart>
<namePart type="family">Spanakis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shanshan</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-494-1</identifier>
</relatedItem>
<abstract>This paper describes the thylinao system for the ChildSafeAds shared task on monitoring sponsored segments in child-facing YouTube videos. The system is a six-member ensemble: two fine-tuned encoders, a zero-shot prompted Qwen3-32B, a LoRA-tuned Qwen3-14B used for ST2 only, a TF-IDF baseline, and a level-2 stacker for the compliance sub-task, blended at equal weight behind a decision layer whose every parameter is fitted on channel-grouped out-of-fold predictions. The system scores mean macro-F1 0.6537 on the evaluation split; inference over the 503 test instances is dominated by one measured 14-minute GPU pass of the 32B member. The zero-shot member is the most valuable and nearly the cheapest, while removing the fine-tuned DeBERTa changes nothing measurable. An ablation over the task’s four data access levels, run on the zero-shot member, finds the channel name adds nothing and the product-page crawl helps product identification but not, measurably, compliance. Validating these changes also surfaced properties of the evaluation, reported as feedback: the metric’s divisor dominates score variance and is recoverable in closed form from a published baseline row, and the benchmark is channel-disjoint but substantially advertiser-overlapping. The final submitted entry additionally contained three labels set by hand, which the task Terms prohibit; this was disclosed to the organisers and is documented in Section 7. The system-only score is reported throughout and no rank is claimed.</abstract>
<identifier type="citekey">silchenko-2026-thylinao-childsafeads</identifier>
<location>
<url>https://aclanthology.org/2026.nllp-1.1/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1</start>
<end>11</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T thylinao at ChildSafeAds 2026: A Six-Member Ensemble and What Each Level of Data Access Provided
%A Silchenko, Maksim
%Y Aletras, Nikolaos
%Y Chalkidis, Ilias
%Y Barrett, Leslie
%Y Goanță, Cătălina
%Y Preoțiuc-Pietro, Daniel
%Y Spanakis, Gerasimos
%Y Xu, Shanshan
%S Proceedings of the Natural Legal Language Processing Workshop 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-494-1
%F silchenko-2026-thylinao-childsafeads
%X This paper describes the thylinao system for the ChildSafeAds shared task on monitoring sponsored segments in child-facing YouTube videos. The system is a six-member ensemble: two fine-tuned encoders, a zero-shot prompted Qwen3-32B, a LoRA-tuned Qwen3-14B used for ST2 only, a TF-IDF baseline, and a level-2 stacker for the compliance sub-task, blended at equal weight behind a decision layer whose every parameter is fitted on channel-grouped out-of-fold predictions. The system scores mean macro-F1 0.6537 on the evaluation split; inference over the 503 test instances is dominated by one measured 14-minute GPU pass of the 32B member. The zero-shot member is the most valuable and nearly the cheapest, while removing the fine-tuned DeBERTa changes nothing measurable. An ablation over the task’s four data access levels, run on the zero-shot member, finds the channel name adds nothing and the product-page crawl helps product identification but not, measurably, compliance. Validating these changes also surfaced properties of the evaluation, reported as feedback: the metric’s divisor dominates score variance and is recoverable in closed form from a published baseline row, and the benchmark is channel-disjoint but substantially advertiser-overlapping. The final submitted entry additionally contained three labels set by hand, which the task Terms prohibit; this was disclosed to the organisers and is documented in Section 7. The system-only score is reported throughout and no rank is claimed.
%U https://aclanthology.org/2026.nllp-1.1/
%P 1-11
Markdown (Informal)
[thylinao at ChildSafeAds 2026: A Six-Member Ensemble and What Each Level of Data Access Provided](https://aclanthology.org/2026.nllp-1.1/) (Silchenko, NLLP 2026)
ACL