Can MLLMs Decode the Creative Leap? Introducing C4 for Cross-Concept Understanding

Hugging Face Daily Papers Papers
cross-concept mllm evaluation benchmark chengyu creativity multimodal

Summary

This paper introduces C4, a cognition-inspired evaluation framework for cross-concept understanding using Chinese idioms (Chengyu), and finds that current multimodal LLMs struggle with creatively encoded meaning.

Creative capabilities of MLLMs matter in design, communication, education, and human--AI collaboration, yet remain difficult to evaluate because explicit targets and reward signals are scarce compared with accuracy-oriented tasks. Cross-concept understanding is a core cognitive capacity underlying receptive creativity. It enables a perceiver to recover intended meaning from non-obvious but meaningful conceptual relations. We operationalize item construction as cross-concept encoding and model inference as cross-concept decoding. We introduce C4, a cognition-inspired evaluation framework for Chengyu (Chinese idiom)-based Cross-Concept Creativity. Its encoding component maps target slots to imageable substitute concepts along bridge paths in a manually annotated and third-party-reviewed cross-concept network, enabling batch generation with explicit structure, difficulty indexed by bridge count and depth, and exact answers. Using this framework, we instantiate the C4 Evaluation Set (C4-Eval), comprising 184 synthetic items and 37 human-created cross-concept chengyu figures collected from online sources. We manually construct and review cross-concept relations, bridge paths, and reasoning processes for the collected figures. Each C4-Eval item is instantiated in five task settings, yielding 884 primary answer-recovery cases. Across ten evaluated MLLMs, the strongest closed models reach 50.7% and 48.0% primary accuracy, while open-source models remain substantially lower. Candidate constraints improve accuracy sharply, but bridge hints and explanation requests provide only modest gains. These results expose a substantial gap in how current MLLMs decode creatively encoded meaning through cross-concept relations. The code is in the supplementary material.
Original Article
View Cached Full Text

Cached at: 08/10/26, 06:13 AM

Paper page - Can MLLMs Decode the Creative Leap? Introducing C4 for Cross-Concept Understanding

Source: https://huggingface.co/papers/2608.06501

Abstract

CreativecapabilitiesofMLLMsmatterindesign,communication,education,andhuman--AIcollaboration,yetremaindifficulttoevaluatebecauseexplicittargetsandrewardsignalsarescarcecomparedwithaccuracy-orientedtasks.Cross-conceptunderstandingisacorecognitivecapacityunderlyingreceptivecreativity.Itenablesaperceivertorecoverintendedmeaningfromnon-obviousbutmeaningfulconceptualrelations.Weoperationalizeitemconstructionascross-conceptencodingandmodelinferenceascross-conceptdecoding.WeintroduceC4,acognition-inspiredevaluationframeworkforChengyu(Chineseidiom)-basedCross-ConceptCreativity.Itsencodingcomponentmapstargetslotstoimageablesubstituteconceptsalongbridgepathsinamanuallyannotatedandthird-party-reviewedcross-conceptnetwork,enablingbatchgenerationwithexplicitstructure,difficultyindexedbybridgecountanddepth,andexactanswers.Usingthisframework,weinstantiatetheC4EvaluationSet(C4-Eval),comprising184syntheticitemsand37human-createdcross-conceptchengyufigurescollectedfromonlinesources.Wemanuallyconstructandreviewcross-conceptrelations,bridgepaths,andreasoningprocessesforthecollectedfigures.EachC4-Evalitemisinstantiatedinfivetasksettings,yielding884primaryanswer-recoverycases.AcrosstenevaluatedMLLMs,thestrongestclosedmodelsreach50.7%and48.0%primaryaccuracy,whileopen-sourcemodelsremainsubstantiallylower.Candidateconstraintsimproveaccuracysharply,butbridgehintsandexplanationrequestsprovideonlymodestgains.TheseresultsexposeasubstantialgapinhowcurrentMLLMsdecodecreativelyencodedmeaningthroughcross-conceptrelations.Thecodeisinthesupplementarymaterial.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2608\.06501

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/2608.06501 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.06501 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.06501 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

ArtECulture: Benchmarking Culture-Conditioned Visual Emotion Understanding in Multimodal Large Language Models

arXiv cs.CL

This paper introduces ArtECulture, a benchmark for culture-conditioned visual emotion understanding in multimodal large language models, covering English, Chinese, and Arabic cultures with balanced Western and non-Western artwork. Evaluations reveal the task remains challenging, and the authors propose a retrieval-augmented framework to inject cultural knowledge into MLLMs.

Causal Probing for Internal Visual Representations in Multimodal Large Language Models

arXiv cs.AI

This paper proposes a causal framework for probing internal visual representations in Multimodal Large Language Models, revealing differences in how entities and abstract concepts are encoded. The study highlights that increasing model depth is crucial for encoding abstract concepts and uncovers a disconnect between perception and reasoning in current MLLMs.