AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

M3UCD: A Multi-task Multimodal Metaphor Understanding Challenge Dataset for LLMs

Tianlong Zheng, Yating Yang, Rui Dong, Bo Ma, Lei Wang, Xi Zhou, Siru Miao, Turghun Osman

AAAI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Unified any-modality language models · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Introduces M3UCD, a multi-task multimodal metaphor understanding challenge dataset for evaluating LLMs, addressing superficial annotations, limited LLM evaluation, and fragmented task frameworks.

Problem

Multimodal metaphor understanding is important for machine cognition, but existing research has superficial dataset annotations, insufficient systematic evaluation of LLMs, and fragmented task frameworks.

Approach

Presents a large, fine-grained multi-task multimodal metaphor dataset with a unified task framework to systematically evaluate large language models on metaphor understanding.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testEmotion Analysis macro-F1: +5.41Improved
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testIntention Recognition macro-F1: +4.93Improved
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testOffensiveness Detection macro-F1: +2.55Improved
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testSarcasm Detection macro-F1: +0.25No clear effect
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testMetaphor Detection macro-F1: +0.65No clear effect
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testSource Domain Detection macro-F1: +7.85Improved
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testTarget Domain Detection macro-F1: +6.92Improved
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testMetaphor Classification macro-F1: +8.72Improved
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testTenor Extraction F1: +2.47Improved
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testVehicle Extraction F1: +2.22Improved
train with MCLF instead of single-task LoRAMLLMs (8 models avg) · M3UCD testDual Extraction F1: +2.05Improved
add LoRA fine-tuning to zero-shot modelQwen2-VL-7B-Instruct · M3UCD testDual Extraction F1: 63.57 -> 68.62Improved

Experimental setup

Benchmarks: M3UCD

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy