ENTRAP-VL: A Taxonomic Probe for Dual Contextual Entrainment in Vision-Language Models

Hugging Face Daily Papers Papers
vision-language-models contextual-entrainment evaluation-benchmark taxonomy dataset multimodal

Summary

Introduces ENTRAP-VL, a taxonomically structured dataset of 1,500 items to probe contextual entrainment in vision-language models, examining how textual and visual context independently influence model outputs.

Contextual entrainment is the tendency of a model to let auxiliary context in its input pull its output, independently of whether that context is relevant, true, or even meaningful. Recently, it has been identified and given a mechanistic account in unimodal language models. Whether and how it manifests in vision-language models (VLMs) is, by contrast, largely unexamined, and the field lacks a purpose-built instrument with which to investigate it. We take the position that studying contextual entrainment in VLMs requires more than porting an existing text-only benchmark to the multimodal setting: it requires a taxonomically structured, dual-modality instrument whose conditions are constructed around the item at hand (the depicted image in the textual stream, the textual query in the visual stream). We argue that the move to VLMs is substantive rather than incremental. It makes entrainment a dual phenomenon, drivable independently by textual and by visual context, and it opens a veracity distinction (context that is false of the depicted scene yet possible in the world) that has no counterpart in the unimodal, world-knowledge-only formulation of prior work. To make this position concrete and actionable, we introduce ENTRAP-VL (ENTRainment Assessment Probe for Vision and Language), a manually curated dataset of 1,500 items across eight categories, organized by a taxonomy that spans two axes, i.e., the association of context with the item and its relationship to truth, and split into a textual-entrainment stream (eight context conditions) and a visual-entrainment stream (three context conditions). We do not claim to measure entrainment in any particular model; we provide the instrument, the taxonomy that motivates it, and the evaluation protocols it enables, so that the community can investigate the phenomenon rigorously. We will release the dataset and its documentation publicly.
Original Article
View Cached Full Text

Cached at: 07/24/26, 05:06 AM

Paper page - ENTRAP-VL: A Taxonomic Probe for Dual Contextual Entrainment in Vision-Language Models

Source: https://huggingface.co/papers/2607.20092

Abstract

Contextualentrainmentisthetendencyofamodeltoletauxiliarycontextinitsinputpullitsoutput,independentlyofwhetherthatcontextisrelevant,true,orevenmeaningful.Recently,ithasbeenidentifiedandgivenamechanisticaccountinunimodallanguagemodels.Whetherandhowitmanifestsinvision-languagemodels(VLMs)is,bycontrast,largelyunexamined,andthefieldlacksapurpose-builtinstrumentwithwhichtoinvestigateit.WetakethepositionthatstudyingcontextualentrainmentinVLMsrequiresmorethanportinganexistingtext-onlybenchmarktothemultimodalsetting:itrequiresataxonomicallystructured,dual-modalityinstrumentwhoseconditionsareconstructedaroundtheitemathand(thedepictedimageinthetextualstream,thetextualqueryinthevisualstream).WearguethatthemovetoVLMsissubstantiveratherthanincremental.Itmakesentrainmentadualphenomenon,drivableindependentlybytextualandbyvisualcontext,anditopensaveracitydistinction(contextthatisfalseofthedepictedsceneyetpossibleintheworld)thathasnocounterpartintheunimodal,world-knowledge-onlyformulationofpriorwork.Tomakethispositionconcreteandactionable,weintroduceENTRAP-VL(ENTRainmentAssessmentProbeforVisionandLanguage),amanuallycurateddatasetof1,500itemsacrosseightcategories,organizedbyataxonomythatspanstwoaxes,i.e.,theassociationofcontextwiththeitemanditsrelationshiptotruth,andsplitintoatextual-entrainmentstream(eightcontextconditions)andavisual-entrainmentstream(threecontextconditions).Wedonotclaimtomeasureentrainmentinanyparticularmodel;weprovidetheinstrument,thetaxonomythatmotivatesit,andtheevaluationprotocolsitenables,sothatthecommunitycaninvestigatethephenomenonrigorously.Wewillreleasethedatasetanditsdocumentationpublicly.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2607\.20092

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.20092 in a model README.md to link it from this page.

Datasets citing this paper1

#### goyalkaraniit/ENTRAP-VL Viewer• Updatedabout 18 hours ago • 2.9k • 10

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.20092 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Sentence-Level Contextual Entrainment in Large Language Models

arXiv cs.CL

This paper extends contextual entrainment from token-level to sentence-level, showing that even counterfactual sentences in prompts increase their probability during inference. The effect decreases with model size and is driven by 2-4% of attention heads, which can be ablated without performance loss.

Stateful Visual Encoders for Vision-Language Models

Hugging Face Daily Papers

This paper introduces a stateful visual encoder for vision-language models that conditions visual representations on prior features, enabling better visual comparison in multi-image and agentic settings. The method shows consistent improvements across tasks such as cross-image spatial aggregation and longitudinal radiology.

DataComp-VLM: Improved Open Datasets for Vision-Language Models

Hugging Face Daily Papers

This paper introduces DataComp-VLM (DCVLM), a comprehensive benchmark for evaluating data curation strategies for vision-language models. The authors find that data mixing, rather than filtering, significantly improves performance, and their resulting DCVLM-Baseline dataset achieves state-of-the-art results on 33 downstream tasks.