EMNLP Findings2024
Stark: Social Long-Term Multi-Modal Conversation with Persona Commonsense Knowledge
Young-Jun Lee, Dokyong Lee, Junyoung Youn, Kyeongjin Oh, Byungsoo Ko, Jonghwan Hyeon, Ho-Jin Choi
摘要
Humans share a wide variety of images related to their personal experiences within conversations via instant messaging tools. However, existing works focus on (1) image-sharing behavior in singular sessions, leading to limited long-term social interaction, and (2) a lack of personalized image-sharing behavior. In this work, we introduce STARK , a large-scale long-term multi-modal conversation dataset that covers a wide range of social personas in a multi-modality format, time intervals, and images. To construct STARK automatically, we propose a novel multi-modal contextualization framework, MCU, that generates longterm multi-modal dialogue distilled from Chat-GPT and our proposed Plan-and-Execute image aligner. Using our STARK, we train a multimodal conversation model, ULTRON 7B, which demonstrates impressive visual imagination ability. Furthermore, we demonstrate the effectiveness of our dataset in human evaluation. We make our source code and dataset publicly available 1 .
问 Lune
几分钟读透这篇 Paper,不用再花一下午。
Lune 能读完全文,也能读引用它的顶会 Paper,讲清它的结论、证据和局限,每一点都注明出自哪篇 Paper。不用硬啃,也不靠猜。
已有 3 篇顶会 Paper 引用了它。别等审稿人问起,先弄清它们发现了什么。
脉络
- Enabling Chatbots with Eyes and Ears: An Immersive Multimodal Conversation System for Dynamic Interactions 引用了 Visual Instruction Tuning
- Enabling Chatbots with Eyes and Ears: An Immersive Multimodal Conversation System for Dynamic Interactions 引用了 Learning Transferable Visual Models From Natural Language Supervision
- Visual Instruction Tuning 引用了 High-Resolution Image Synthesis with Latent Diffusion Models
- Visual Instruction Tuning 引用了 Learning Transferable Visual Models From Natural Language Supervision
- MultiVerse: A Multi-Turn Conversation Benchmark for Evaluating Large Vision and Language Models 引用了 Visual Instruction Tuning
- MIRROR: Multimodal Cognitive Reframing Therapy for Rolling with Resistance 引用了 Learning Transferable Visual Models From Natural Language Supervision
引用它的顶会 Paper3
- MIRROR: Multimodal Cognitive Reframing Therapy for Rolling with ResistanceSubin Kim, Hoonrae Kim, Jihyun Lee, Yejin Jeon 等EMNLP 2025 · 6 次引用
- MultiVerse: A Multi-Turn Conversation Benchmark for Evaluating Large Vision and Language ModelsYoung-Jun Lee, Byung-Kwan Lee, Jianshu Zhang, Yechan Hwang 等ICCV 2025 · 2 次引用
- Enabling Chatbots with Eyes and Ears: An Immersive Multimodal Conversation System for Dynamic InteractionsJihyoung Jang, Minwook Bae, Minji Kim, Dilek Hakkani-Tür 等ACL 2025
它引用的 Paper35
- Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh 等ICML 2021 · 47,906 次引用
- High-Resolution Image Synthesis with Latent Diffusion ModelsRobin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser 等CVPR 2022 · 13,123 次引用
- Visual Instruction TuningHaotian Liu, Chunyuan Li, Qingyang Wu, Yong Jae LeeNeurIPS 2023 · 11,349 次引用
- QLoRA: Efficient Finetuning of Quantized LLMsTim Dettmers, Artidoro Pagnoni, Ari Holtzman, Luke ZettlemoyerNeurIPS 2023 · 5,863 次引用
- HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging FaceYongliang Shen, Kaitao Song, Xu Tan, Dongsheng Li 等NeurIPS 2023 · 1,778 次引用
相关 Paper
- LongMP-Bench: A Benchmark for Multimodal Persona Understanding in Long-Term DialoguesZhuoqun Li, Zhaopei Huang, Wenxuan Wang, Qin JinACL Findings 2026
- Let's Go Real Talk: Spoken Dialogue Model for Face-to-Face ConversationSe Jin Park, Chae Won Kim, Hyeongseop Rha, Minsu Kim 等ACL 2024
- Integrating Stickers into Multimodal Dialogue Summarization: A Novel Dataset and Approach for Enhancing Social Media InteractionYuanchen Shi, Fang KongACM MM 2024 · 7 次引用
- TikTalk: A Video-Based Dialogue Dataset for Multi-Modal Chitchat in Real WorldHongpeng Lin, Ludan Ruan, Wenke Xia, Peiyu Liu 等ACM MM 2023 · 10 次引用
- Multimodal Multi-turn Conversation Stance Detection: A Challenge Dataset and Effective ModelFuqiang Niu, Zebang Cheng, Xianghua Fu, Xiaojiang Peng 等ACM MM 2024 · 13 次引用
