arXiv cs.CL· Yuanchen Shi, Fang Kong·· 11 小时前AI 评分18
S2TDR：面向贴纸对话理解的结构化多模态重建框架
S2TDR: Structured Multimodal Reconstruction for Sticker-Aware Dialogue Understanding
AI 导读
S2TDR 提出一种结构化多模态重建框架，将对话中的贴纸轮次转换为符合上下文的文本话语，使现有纯文本对话模型可直接消费。该框架把贴纸解读拆解为视觉线索提取、情感与意图推断、上下文感知对话重建三步，OCR 文本仅作辅助线索。在情感分析、意图识别和对话摘要任务上，S2TDR 相比直接融合贴纸的方法普遍提升了多个下游模型的表现。
正文 · 原文
Abstract:Stickers are widely used in social media dialogues to express affect, stance, and intent beyond literal text. Their meanings are often non-literal and context-dependent, making them difficult to handle with generic visual fusion or standalone captioning. We propose Sticker-to-Text Dialogue Reconstruction (S2TDR), a structured multimodal reconstruction framework that converts sticker turns into contextually appropriate textual utterances. S2TDR decomposes sticker interpretation into visual cue extraction, affective-intent inference, and context-aware dialogue reconstruction, using OCR text only as an auxiliary cue when available. The reconstructed dialogue can then be consumed by existing text-based dialogue models. Experiments on sentiment analysis, intent recognition, and dialogue summarization show that S2TDR generally improves multiple downstream models over direct sticker integration methods. Further analyses on component ablations, OCR-empty stickers, and reconstruction quality indicate that structured reconstruction provides an interpretable and effective way to integrate sticker semantics into dialogue understanding.
| Comments: | 12 pages, 4 figures, 5 tables |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.41176 [cs.CL] |
| (or arXiv:2610.41176v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.41176 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Shi Yuanchen [view email]
[v1]
Fri, 30 Oct 2026 02:57:24 UTC (1,677 KB)
来源：arXiv cs.CL · arxiv.org