@inproceedings{tian-etal-2026-devil,
title = "The Devil is in the Distributions: Explicit Modeling of Scene Content is Key in Zero-Shot Video Captioning",
author = "Tian, Mingkai and
Li, Guorong and
Qi, Yuankai and
Hengel, Anton Van Den and
Huang, Qingming",
editor = "Demberg, Vera and
Inui, Kentaro and
Marquez, Llu{\'i}s",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EACL} 2026",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-eacl.98/",
pages = "1916--1929",
ISBN = "979-8-89176-386-9",
abstract = "Zero-shot video captioning requires that a model generate high-quality captions without human-annotated video-text pairs for training. State-of-the-art approaches to the problem leverage CLIP to extract video-informed text prompts to guide language models in generating captions. However, by using representations at a single granularity (e.g., noun phrases or full sentences), these methods tend to focus on one key aspect of the scene and build a caption that ignores the rest of the visual input. To address this issue, and generate more accurate and complete captions, we propose a novel progressive multi-granularity textual prompting strategy for zero-shot video captioning. Our approach constructs three distinct memory banks, encompassing noun phrases, scene graphs of noun phrases, and entire sentences. Moreover, we introduce a category-aware retrieval mechanism that models the distribution of natural language surrounding the specific topics, to promote prompt diversity while ensuring visual relevance. Extensive experiments on both in-domain and cross-domain settings demonstrate that the proposed method consistently outperforms state-of-the-art approaches."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tian-etal-2026-devil">
<titleInfo>
<title>The Devil is in the Distributions: Explicit Modeling of Scene Content is Key in Zero-Shot Video Captioning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mingkai</namePart>
<namePart type="family">Tian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guorong</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuankai</namePart>
<namePart type="family">Qi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anton</namePart>
<namePart type="given">Van</namePart>
<namePart type="given">Den</namePart>
<namePart type="family">Hengel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qingming</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vera</namePart>
<namePart type="family">Demberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lluís</namePart>
<namePart type="family">Marquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Rabat, Morocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-386-9</identifier>
</relatedItem>
<abstract>Zero-shot video captioning requires that a model generate high-quality captions without human-annotated video-text pairs for training. State-of-the-art approaches to the problem leverage CLIP to extract video-informed text prompts to guide language models in generating captions. However, by using representations at a single granularity (e.g., noun phrases or full sentences), these methods tend to focus on one key aspect of the scene and build a caption that ignores the rest of the visual input. To address this issue, and generate more accurate and complete captions, we propose a novel progressive multi-granularity textual prompting strategy for zero-shot video captioning. Our approach constructs three distinct memory banks, encompassing noun phrases, scene graphs of noun phrases, and entire sentences. Moreover, we introduce a category-aware retrieval mechanism that models the distribution of natural language surrounding the specific topics, to promote prompt diversity while ensuring visual relevance. Extensive experiments on both in-domain and cross-domain settings demonstrate that the proposed method consistently outperforms state-of-the-art approaches.</abstract>
<identifier type="citekey">tian-etal-2026-devil</identifier>
<location>
<url>https://aclanthology.org/2026.findings-eacl.98/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>1916</start>
<end>1929</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The Devil is in the Distributions: Explicit Modeling of Scene Content is Key in Zero-Shot Video Captioning
%A Tian, Mingkai
%A Li, Guorong
%A Qi, Yuankai
%A Hengel, Anton Van Den
%A Huang, Qingming
%Y Demberg, Vera
%Y Inui, Kentaro
%Y Marquez, Lluís
%S Findings of the Association for Computational Linguistics: EACL 2026
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-386-9
%F tian-etal-2026-devil
%X Zero-shot video captioning requires that a model generate high-quality captions without human-annotated video-text pairs for training. State-of-the-art approaches to the problem leverage CLIP to extract video-informed text prompts to guide language models in generating captions. However, by using representations at a single granularity (e.g., noun phrases or full sentences), these methods tend to focus on one key aspect of the scene and build a caption that ignores the rest of the visual input. To address this issue, and generate more accurate and complete captions, we propose a novel progressive multi-granularity textual prompting strategy for zero-shot video captioning. Our approach constructs three distinct memory banks, encompassing noun phrases, scene graphs of noun phrases, and entire sentences. Moreover, we introduce a category-aware retrieval mechanism that models the distribution of natural language surrounding the specific topics, to promote prompt diversity while ensuring visual relevance. Extensive experiments on both in-domain and cross-domain settings demonstrate that the proposed method consistently outperforms state-of-the-art approaches.
%U https://aclanthology.org/2026.findings-eacl.98/
%P 1916-1929
Markdown (Informal)
[The Devil is in the Distributions: Explicit Modeling of Scene Content is Key in Zero-Shot Video Captioning](https://aclanthology.org/2026.findings-eacl.98/) (Tian et al., Findings 2026)
ACL