@inproceedings{han-etal-2026-oscbench,
title = "{OSCB}ench: Benchmarking Object State Change in Text-to-Video Generation",
author = "Han, Xianjing and
Zhu, Bin and
Hu, Shiqi and
Li, Franklin Mingzhe and
Carrington, Patrick and
Zimmermann, Roger and
Chen, Jingjing",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.1425/",
pages = "30867--30884",
ISBN = "979-8-89176-390-6",
abstract = "Text-to-video (T2V) generation models have made rapid progress in producing visually high-quality and temporally coherent videos. However, existing benchmarks primarily focus on perceptual quality, text{--}video alignment, or physical plausibility, leaving a critical aspect of action understanding largely unexplored: object state change (OSC) explicitly specified in the text prompt. OSC refers to the transformation of an object{'}s state induced by an action, such as peeling a potato or slicing a lemon. In this paper, we introduce OSCBench, a benchmark specifically designed to assess OSC performance in T2V models. OSCBench is constructed from instructional cooking data and systematically organizes action{--}object interactions into regular, novel, and compositional scenarios to probe both in-distribution performance and generalization. We evaluate six representative open-source and proprietary T2V models using both human user study and multimodal large language model (MLLM){--}based automatic evaluation. Our results show that, despite strong performance on semantic and scene alignment, current T2V models consistently struggle with accurate and temporally consistent object state changes, especially in novel and compositional settings. These findings position OSC as a key bottleneck in text-to-video generation and establish OSCBench as a diagnostic benchmark for advancing state-aware video generation models. Project page: https://hanxjing.github.io/OSCBench."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="han-etal-2026-oscbench">
<titleInfo>
<title>OSCBench: Benchmarking Object State Change in Text-to-Video Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xianjing</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bin</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shiqi</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Franklin</namePart>
<namePart type="given">Mingzhe</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Patrick</namePart>
<namePart type="family">Carrington</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roger</namePart>
<namePart type="family">Zimmermann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingjing</namePart>
<namePart type="family">Chen</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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</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, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>Text-to-video (T2V) generation models have made rapid progress in producing visually high-quality and temporally coherent videos. However, existing benchmarks primarily focus on perceptual quality, text–video alignment, or physical plausibility, leaving a critical aspect of action understanding largely unexplored: object state change (OSC) explicitly specified in the text prompt. OSC refers to the transformation of an object’s state induced by an action, such as peeling a potato or slicing a lemon. In this paper, we introduce OSCBench, a benchmark specifically designed to assess OSC performance in T2V models. OSCBench is constructed from instructional cooking data and systematically organizes action–object interactions into regular, novel, and compositional scenarios to probe both in-distribution performance and generalization. We evaluate six representative open-source and proprietary T2V models using both human user study and multimodal large language model (MLLM)–based automatic evaluation. Our results show that, despite strong performance on semantic and scene alignment, current T2V models consistently struggle with accurate and temporally consistent object state changes, especially in novel and compositional settings. These findings position OSC as a key bottleneck in text-to-video generation and establish OSCBench as a diagnostic benchmark for advancing state-aware video generation models. Project page: https://hanxjing.github.io/OSCBench.</abstract>
<identifier type="citekey">han-etal-2026-oscbench</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.1425/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>30867</start>
<end>30884</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T OSCBench: Benchmarking Object State Change in Text-to-Video Generation
%A Han, Xianjing
%A Zhu, Bin
%A Hu, Shiqi
%A Li, Franklin Mingzhe
%A Carrington, Patrick
%A Zimmermann, Roger
%A Chen, Jingjing
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F han-etal-2026-oscbench
%X Text-to-video (T2V) generation models have made rapid progress in producing visually high-quality and temporally coherent videos. However, existing benchmarks primarily focus on perceptual quality, text–video alignment, or physical plausibility, leaving a critical aspect of action understanding largely unexplored: object state change (OSC) explicitly specified in the text prompt. OSC refers to the transformation of an object’s state induced by an action, such as peeling a potato or slicing a lemon. In this paper, we introduce OSCBench, a benchmark specifically designed to assess OSC performance in T2V models. OSCBench is constructed from instructional cooking data and systematically organizes action–object interactions into regular, novel, and compositional scenarios to probe both in-distribution performance and generalization. We evaluate six representative open-source and proprietary T2V models using both human user study and multimodal large language model (MLLM)–based automatic evaluation. Our results show that, despite strong performance on semantic and scene alignment, current T2V models consistently struggle with accurate and temporally consistent object state changes, especially in novel and compositional settings. These findings position OSC as a key bottleneck in text-to-video generation and establish OSCBench as a diagnostic benchmark for advancing state-aware video generation models. Project page: https://hanxjing.github.io/OSCBench.
%U https://aclanthology.org/2026.acl-long.1425/
%P 30867-30884
Markdown (Informal)
[OSCBench: Benchmarking Object State Change in Text-to-Video Generation](https://aclanthology.org/2026.acl-long.1425/) (Han et al., ACL 2026)
ACL
- Xianjing Han, Bin Zhu, Shiqi Hu, Franklin Mingzhe Li, Patrick Carrington, Roger Zimmermann, and Jingjing Chen. 2026. OSCBench: Benchmarking Object State Change in Text-to-Video Generation. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 30867–30884, San Diego, California, United States. Association for Computational Linguistics.