MultiModal-GPT: A Vision and Language Model for Dialogue with Humans
By T. Gong, Chengqi Lyu, Shilong Zhang, Yudong Wang, Miao Zheng, Qianmengke Zhao, Kuikun Liu, Wenwei Zhang, Ping Luo, Kai Chen. 2023
328 citations recorded by the source corpus.
We present a vision and language model named MultiModal-GPT to conduct multi-round dialogue with humans. MultiModal-GPT can follow various instructions from humans, such as generating a detailed caption, counting the number of interested objects, and answering general questions from users. MultiModal-GPT is parameter-efficiently fine-tuned from OpenFlamingo, with Low-rank Adapter (LoRA) added both in the cross-attention part and the self-attention part of the language model. We first construct instruction templates with vision and language data for multi-modality instruction tuning to make the model understand and follow human instructions. We find the quality of training data is vital for the dialogue performance, where few data containing short answers can lead the model to respond shortly to any instructions. To further enhance the ability to chat with humans of the MultiModal-GPT, we utilize language-only instruction-following data to train the MultiModal-GPT jointly. The joint training of language-only and visual-language instructions with the \emph{same} instruction template effectively improves dialogue performance. Various demos show the ability of continuous dialogue of MultiModal-GPT with humans. Code, dataset, and demo are at https://github.com/open-mmlab/Multimodal-GPT
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Large Language Models, Multimodal Learning, Dialogue & Conversational AI
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Yu Qiao
- Xiaogang Wang
- Dacheng Tao
- Trevor Darrell
- Ming-Hsuan Yang
- Jitendra Malik
- Yi Yang
In Large Language Models
In Multimodal Learning
Citations in the corpus
- Improved Baselines with Visual Instruction Tuning (2023, background): “Recent works further demonstrate improved performance by scaling up the pretraining data [3, 13, 53], instruction-following data [13, 17, 28, 57], visual encoders [3], or language models [38], respectively.”
- MovieChat: From Dense Token to Sparse Memory for Long Video Understanding (2023, background): “Existing vision-centric MLLMs follow the paradigm that utilizing pre-trained LLMs and visual encoder with additional learnable modules (Q-former [21, 34, 36, 82] or simple projection layer [22, 39, 43, 61]).”
- MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models (2023, neutral): “The evaluated MLLMs include BLIP-2 [30], LLaVA [33], MiniGPT-4 [56], mPLUG-Owl [50], LLaMA-Adapter-v2 [20], Otter [29], Multimodal-GPT [22], InstructBLIP [18], VisualGLM-6B [12], PandaGPT [42], ImageBind-LLM [2], and LaVIN [36].”
- RSGPT: A Remote Sensing Vision Language Model and Benchmark (2023, neutral): “MM-GPT [14] and Otter [22] are fine-tuned on meticulously constructed instruction data to enhance user interaction.”
- MMICL: Empowering Vision-language Model with Multi-Modal In-Context Learning (2023, neutral): “, 2023), MultiModal-GPT (Gong et al., 2023), VisualGPM6B (Du et al.”
- Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding (2023, background): “These models are characterized by their effectiveness in capturing and translating complex visual patterns into coherent linguistic representations [5, 12, 18, 33, 45, 49, 71, 73, 78].”
- Kosmos-2: Grounding Multimodal Large Language Models to the World (2023, neutral): “On SEED-Bench, we compare K OSMOS -2 with popular MLLMs including MiniGPT4 (Zhu et al., 2023), LLaVA (Liu et al., 2023a), BLIP-2 (Li et al., 2023b), InstructBLIP (Dai et al., 2023), MultiModal-GPT (Gong et al., 2023), mPLUG-Owl (Ye et al., 2023), VideoChat (Li et al., 2023c) and Video-ChatGPT (Maaz ”
- MM-LLMs: Recent Advances in MultiModal Large Language Models (2024, neutral): “I+T → T: BLIP-2 (Li et al., 2023e), Kosmos-1 (Huang et al., 2023c), PaLM-E (Driess et al., 2023), ViperGPT (Surís et al., 2023), LLaVA (Liu et al., 2023e), MiniGPT-4 (Zhu et al., 2023a), mPLUG-Owl (Ye et al., 2023b), Otter (Li et al., 2023b), MultiModal-GPT (Gong et al., 2023), PandaGPT (Su et al., ”
- 3D-LLM: Injecting the 3D World into Large Language Models (2023, background): “Our work is also related to vision-language pre-trained models that connect images and natural language [30, 31, 18, 36, 25].”
- Shikra: Unleashing Multimodal LLM's Referential Dialogue Magic (2023, background): “Subsequently, MM-GPT (Gong et al., 2023), and Otter (Li et al.”
- Analyzing and Mitigating Object Hallucination in Large Vision-Language Models (2023, background): “Large Vision-Language Models (LVLMs) have made significant progress in understanding real-world images, showing potential towards achieving general artificial intelligence (Liu et al., 2023d; Zhu et al., 2023; Ye et al., 2023; Li et al., 2023a; Maaz et al., 2023; Gong et al., 2023).”
- InternVideo2: Scaling Foundation Models for Multimodal Video Understanding (2024, background): “It is a long-form movie question-answering dataset[Zhang et al., 2023].”
- SVIT: Scaling up Visual Instruction Tuning (2023, background): “However, these cap-tions and questions/answers are usually short and focus on visual perception and simple questions, which may lead to ineffective model training (Gong et al., 2023).”
- Instruction Tuning for Large Language Models: A Survey (2023, neutral): “Furthermore, the growing trend of using LLMs as judges for evaluating other models introduces significant risks of bias, inconsistency, and overconfidence, as the judge model’s own values and limitations can skew the results.”
- VisIT-Bench: A Benchmark for Vision-Language Instruction Following Inspired by Real-World Use (2023, neutral): “The other approach [13, 16, 65, 46, 66, 17, 11] focuses on building a single pretrained model that can follow instructions by supervised finetuning on multimodal vision-language data.”
- An Embodied Generalist Agent in 3D World (2023, neutral): “The two streams merged into instruction-tuned LVLMs (Liu et al., 2023b; Zhu et al., 2023b; Ye et al., 2023; Gao et al., 2023; Li et al., 2023b; Gong et al., 2023c; Dai et al., 2023).”