<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-zhang-2026-serialdistill">
    <titleInfo>
        <title>SerialDistill: Retrieval-Conditioned On-Policy Distillation for Data-Scarce Serial Subtitle Translation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yiran</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ziqi</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Thirteenth Workshop on Asian Translation (WAT 2026)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Toshiaki</namePart>
            <namePart type="family">Nakazawa</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Isao</namePart>
            <namePart type="family">Goto</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Hengqin, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-446-0</identifier>
    </relatedItem>
    <abstract>Serial subtitle translation requires work-specific terminology and translation conventions to be maintained across episodes, despite scarce bilingual supervision. Previously finalized subtitles provide both adaptation data and translation memory, but our experiments show that the benefit of combining supervised adaptation and retrieval varies across series. We introduce SerialDistill, which combines LoRA-based domain adaptation with retrieval-conditioned on-policy distributional distillation (OPD). A Qwen3-8B student generates translation trajectories, while a frozen Qwen3-14B teacher provides token-level supervision under the same bilingual context retrieved exclusively from previously finalized episodes. On Japanese-to-Chinese translation across two anime series, SerialDistill improves over the non-distilled LoRA-8B+RAG baseline by up to 3.80 chrF and 4.15 BLEU, measured on pooled held-out episodes. The student also outperforms the larger retrieval-augmented teacher, which has not undergone domain adaptation. These results support retrieval-conditioned distillation as an additional adaptation stage for data-scarce serial subtitle translation.</abstract>
    <identifier type="citekey">zhang-zhang-2026-serialdistill</identifier>
    <location>
        <url>https://aclanthology.org/2026.wat-1.5/</url>
    </location>
    <part>
        <date>2026-11</date>
        <extent unit="page">
            <start>71</start>
            <end>83</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T SerialDistill: Retrieval-Conditioned On-Policy Distillation for Data-Scarce Serial Subtitle Translation
%A Zhang, Yiran
%A Zhang, Ziqi
%Y Nakazawa, Toshiaki
%Y Goto, Isao
%S Proceedings of the Thirteenth Workshop on Asian Translation (WAT 2026)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-446-0
%F zhang-zhang-2026-serialdistill
%X Serial subtitle translation requires work-specific terminology and translation conventions to be maintained across episodes, despite scarce bilingual supervision. Previously finalized subtitles provide both adaptation data and translation memory, but our experiments show that the benefit of combining supervised adaptation and retrieval varies across series. We introduce SerialDistill, which combines LoRA-based domain adaptation with retrieval-conditioned on-policy distributional distillation (OPD). A Qwen3-8B student generates translation trajectories, while a frozen Qwen3-14B teacher provides token-level supervision under the same bilingual context retrieved exclusively from previously finalized episodes. On Japanese-to-Chinese translation across two anime series, SerialDistill improves over the non-distilled LoRA-8B+RAG baseline by up to 3.80 chrF and 4.15 BLEU, measured on pooled held-out episodes. The student also outperforms the larger retrieval-augmented teacher, which has not undergone domain adaptation. These results support retrieval-conditioned distillation as an additional adaptation stage for data-scarce serial subtitle translation.
%U https://aclanthology.org/2026.wat-1.5/
%P 71-83

Markdown (Informal)

[SerialDistill: Retrieval-Conditioned On-Policy Distillation for Data-Scarce Serial Subtitle Translation](https://aclanthology.org/2026.wat-1.5/) (Zhang & Zhang, WAT 2026)

• SerialDistill: Retrieval-Conditioned On-Policy Distillation for Data-Scarce Serial Subtitle Translation (Zhang & Zhang, WAT 2026)

ACL

• Yiran Zhang and Ziqi Zhang. 2026. SerialDistill: Retrieval-Conditioned On-Policy Distillation for Data-Scarce Serial Subtitle Translation. In Proceedings of the Thirteenth Workshop on Asian Translation (WAT 2026), pages 71–83, Hengqin, China. Association for Computational Linguistics.