Evolutionary Multimodal Reasoning via Hierarchical Semantic Representation for Intent Recognition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhou, Qianrui, Xu, Hua, Gu, Yunjin, Wang, Yifan, Li, Songze, Zhang, Hanlei
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908865505787904
author Zhou, Qianrui
Xu, Hua
Gu, Yunjin
Wang, Yifan
Li, Songze
Zhang, Hanlei
author_facet Zhou, Qianrui
Xu, Hua
Gu, Yunjin
Wang, Yifan
Li, Songze
Zhang, Hanlei
contents Multimodal intent recognition aims to infer human intents by jointly modeling various modalities, playing a pivotal role in real-world dialogue systems. However, current methods struggle to model hierarchical semantics underlying complex intents and lack the capacity for self-evolving reasoning over multimodal representations. To address these issues, we propose HIER, a novel method that integrates HIerarchical semantic representation with Evolutionary Reasoning based on Multimodal Large Language Model (MLLM). Inspired by human cognition, HIER introduces a structured reasoning paradigm that organizes multimodal semantics into three progressively abstracted levels. It starts with modality-specific tokens capturing localized semantic cues, which are then clustered via a label-guided strategy to form mid-level semantic concepts. To capture higher-order structure, inter-concept relations are selected using JS divergence scores to highlight salient dependencies across concepts. These hierarchical representations are then injected into MLLM via CoT-driven prompting, enabling step-wise reasoning. Besides, HIER utilizes a self-evolution mechanism that refines semantic representations through MLLM feedback, allowing dynamic adaptation during inference. Experiments on three challenging benchmarks show that HIER consistently outperforms state-of-the-art methods and MLLMs with 1-3% gains across all metrics. Code and more results are available at https://github.com/thuiar/HIER.
format Preprint
id arxiv_https___arxiv_org_abs_2603_03827
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Evolutionary Multimodal Reasoning via Hierarchical Semantic Representation for Intent Recognition
Zhou, Qianrui
Xu, Hua
Gu, Yunjin
Wang, Yifan
Li, Songze
Zhang, Hanlei
Multimedia
Multimodal intent recognition aims to infer human intents by jointly modeling various modalities, playing a pivotal role in real-world dialogue systems. However, current methods struggle to model hierarchical semantics underlying complex intents and lack the capacity for self-evolving reasoning over multimodal representations. To address these issues, we propose HIER, a novel method that integrates HIerarchical semantic representation with Evolutionary Reasoning based on Multimodal Large Language Model (MLLM). Inspired by human cognition, HIER introduces a structured reasoning paradigm that organizes multimodal semantics into three progressively abstracted levels. It starts with modality-specific tokens capturing localized semantic cues, which are then clustered via a label-guided strategy to form mid-level semantic concepts. To capture higher-order structure, inter-concept relations are selected using JS divergence scores to highlight salient dependencies across concepts. These hierarchical representations are then injected into MLLM via CoT-driven prompting, enabling step-wise reasoning. Besides, HIER utilizes a self-evolution mechanism that refines semantic representations through MLLM feedback, allowing dynamic adaptation during inference. Experiments on three challenging benchmarks show that HIER consistently outperforms state-of-the-art methods and MLLMs with 1-3% gains across all metrics. Code and more results are available at https://github.com/thuiar/HIER.
title Evolutionary Multimodal Reasoning via Hierarchical Semantic Representation for Intent Recognition
topic Multimedia
url https://arxiv.org/abs/2603.03827