@inproceedings{kohankhaki-etal-2026-latent,
title = "Latent Prediction Needs Alignment: A Controlled Study of Joint-Embedding Predictive Vision{--}Language Learning",
author = "Kohankhaki, Mohammad and
Kusuma, Daniel and
Salehi, Shirin and
Kamp, Carsten and
Brell-Cokcan, Sigrid and
Schmeink, Anke",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-long.232/",
pages = "5210--5222",
ISBN = "979-8-89176-439-2",
abstract = "Predictive vision{--}language learning aims to learn semantic representations by predicting latent targets rather than reconstructing pixels or generating text. We ask whether latent prediction alone induces the global image{--}text geometry required for zero-shot matching, and how the location of alignment pressure changes this outcome. In a controlled joint-embedding predictive model family with matched training and evaluation protocols, non-contrastive latent prediction retains information recoverable by supervised adapters and token-level probes but does not form a usable zero-shot image{--}text space under pooled cosine scoring. A target-contrastive design improves matching, but remains weaker than the directly aligned variant. Our results show that latent prediction is underconstrained at the global embedding level in this setting. Direct global alignment supplies the constraint that makes predictive representations usable for zero-shot scoring. With only a modest direct global alignment loss, latent prediction becomes competitive with matched CLIP and SigLIP baselines and improves several language-sensitive diagnostics."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kohankhaki-etal-2026-latent">
<titleInfo>
<title>Latent Prediction Needs Alignment: A Controlled Study of Joint-Embedding Predictive Vision–Language Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="family">Kohankhaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Kusuma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shirin</namePart>
<namePart type="family">Salehi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carsten</namePart>
<namePart type="family">Kamp</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sigrid</namePart>
<namePart type="family">Brell-Cokcan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anke</namePart>
<namePart type="family">Schmeink</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 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</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">979-8-89176-439-2</identifier>
</relatedItem>
<abstract>Predictive vision–language learning aims to learn semantic representations by predicting latent targets rather than reconstructing pixels or generating text. We ask whether latent prediction alone induces the global image–text geometry required for zero-shot matching, and how the location of alignment pressure changes this outcome. In a controlled joint-embedding predictive model family with matched training and evaluation protocols, non-contrastive latent prediction retains information recoverable by supervised adapters and token-level probes but does not form a usable zero-shot image–text space under pooled cosine scoring. A target-contrastive design improves matching, but remains weaker than the directly aligned variant. Our results show that latent prediction is underconstrained at the global embedding level in this setting. Direct global alignment supplies the constraint that makes predictive representations usable for zero-shot scoring. With only a modest direct global alignment loss, latent prediction becomes competitive with matched CLIP and SigLIP baselines and improves several language-sensitive diagnostics.</abstract>
<identifier type="citekey">kohankhaki-etal-2026-latent</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-long.232/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>5210</start>
<end>5222</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Latent Prediction Needs Alignment: A Controlled Study of Joint-Embedding Predictive Vision–Language Learning
%A Kohankhaki, Mohammad
%A Kusuma, Daniel
%A Salehi, Shirin
%A Kamp, Carsten
%A Brell-Cokcan, Sigrid
%A Schmeink, Anke
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-439-2
%F kohankhaki-etal-2026-latent
%X Predictive vision–language learning aims to learn semantic representations by predicting latent targets rather than reconstructing pixels or generating text. We ask whether latent prediction alone induces the global image–text geometry required for zero-shot matching, and how the location of alignment pressure changes this outcome. In a controlled joint-embedding predictive model family with matched training and evaluation protocols, non-contrastive latent prediction retains information recoverable by supervised adapters and token-level probes but does not form a usable zero-shot image–text space under pooled cosine scoring. A target-contrastive design improves matching, but remains weaker than the directly aligned variant. Our results show that latent prediction is underconstrained at the global embedding level in this setting. Direct global alignment supplies the constraint that makes predictive representations usable for zero-shot scoring. With only a modest direct global alignment loss, latent prediction becomes competitive with matched CLIP and SigLIP baselines and improves several language-sensitive diagnostics.
%U https://aclanthology.org/2026.aacl-long.232/
%P 5210-5222
Markdown (Informal)
[Latent Prediction Needs Alignment: A Controlled Study of Joint-Embedding Predictive Vision–Language Learning](https://aclanthology.org/2026.aacl-long.232/) (Kohankhaki et al., AACL-IJCNLP 2026)
ACL