@inproceedings{zhu-etal-2026-beyond-ctc,
title = "Beyond {CTC} Spikes: {LLM}-Driven Gloss-Temporal Supervision for Continuous Sign Language Recognition",
author = "Zhu, Junfeng and
Jin, Cheng and
Fang, Yuchun",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.578/",
pages = "13237--13252",
ISBN = "979-8-89176-459-0",
abstract = "Continuous Sign Language Recognition aims to predict gloss sequences from unsegmented sign language videos. Existing methods typically rely on Connectionist Temporal Classification (CTC) loss for implicit alignment, which is the most dominant loss function for current sequence alignment tasks but causes low-entropy alignment and weak visual-semantic correspondence learning. To address this issue, we propose an explicit temporal-semantic supervision framework that provides dense signals beyond CTC. Specifically, we design the Gloss-Temporal Module (GTM) to construct sustained gloss-to-frame interval supervision by using an LLM to integrate three types of information from RGB videos. We further introduce a joint optimization method with frame-level and sequence-level objectives, coupled with a curriculum learning strategy, enabling a transition from local to global alignment and improving the model{'}s learning of visual-semantic features. The results show that dense temporal supervision substantially alleviates alignment collapse and enables even simple RGB-only baseline models to achieve competitive performance among state-of-the-art methods without extra cues, while the recognizer converges significantly faster."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhu-etal-2026-beyond-ctc">
<titleInfo>
<title>Beyond CTC Spikes: LLM-Driven Gloss-Temporal Supervision for Continuous Sign Language Recognition</title>
</titleInfo>
<name type="personal">
<namePart type="given">Junfeng</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cheng</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuchun</namePart>
<namePart type="family">Fang</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 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</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">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Continuous Sign Language Recognition aims to predict gloss sequences from unsegmented sign language videos. Existing methods typically rely on Connectionist Temporal Classification (CTC) loss for implicit alignment, which is the most dominant loss function for current sequence alignment tasks but causes low-entropy alignment and weak visual-semantic correspondence learning. To address this issue, we propose an explicit temporal-semantic supervision framework that provides dense signals beyond CTC. Specifically, we design the Gloss-Temporal Module (GTM) to construct sustained gloss-to-frame interval supervision by using an LLM to integrate three types of information from RGB videos. We further introduce a joint optimization method with frame-level and sequence-level objectives, coupled with a curriculum learning strategy, enabling a transition from local to global alignment and improving the model’s learning of visual-semantic features. The results show that dense temporal supervision substantially alleviates alignment collapse and enables even simple RGB-only baseline models to achieve competitive performance among state-of-the-art methods without extra cues, while the recognizer converges significantly faster.</abstract>
<identifier type="citekey">zhu-etal-2026-beyond-ctc</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.578/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>13237</start>
<end>13252</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Beyond CTC Spikes: LLM-Driven Gloss-Temporal Supervision for Continuous Sign Language Recognition
%A Zhu, Junfeng
%A Jin, Cheng
%A Fang, Yuchun
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F zhu-etal-2026-beyond-ctc
%X Continuous Sign Language Recognition aims to predict gloss sequences from unsegmented sign language videos. Existing methods typically rely on Connectionist Temporal Classification (CTC) loss for implicit alignment, which is the most dominant loss function for current sequence alignment tasks but causes low-entropy alignment and weak visual-semantic correspondence learning. To address this issue, we propose an explicit temporal-semantic supervision framework that provides dense signals beyond CTC. Specifically, we design the Gloss-Temporal Module (GTM) to construct sustained gloss-to-frame interval supervision by using an LLM to integrate three types of information from RGB videos. We further introduce a joint optimization method with frame-level and sequence-level objectives, coupled with a curriculum learning strategy, enabling a transition from local to global alignment and improving the model’s learning of visual-semantic features. The results show that dense temporal supervision substantially alleviates alignment collapse and enables even simple RGB-only baseline models to achieve competitive performance among state-of-the-art methods without extra cues, while the recognizer converges significantly faster.
%U https://aclanthology.org/2026.emnlp-main.578/
%P 13237-13252
Markdown (Informal)
[Beyond CTC Spikes: LLM-Driven Gloss-Temporal Supervision for Continuous Sign Language Recognition](https://aclanthology.org/2026.emnlp-main.578/) (Zhu et al., EMNLP 2026)
ACL