NExT-Chat: An LMM for Chat, Detection and Segmentation

TL;DR

NExT-Chat is a multimodal large language model that introduces a novel pix2emb method for object location modeling, enabling improved visual understanding and multi-task performance in visual grounding, captioning, and reasoning.

Contribution

The paper proposes the pix2emb paradigm for flexible object location modeling and trains NExT-Chat, a multimodal model capable of handling diverse visual tasks with enhanced accuracy.

Findings

NExT-Chat outperforms existing models on key visual tasks.

The pix2emb method enables flexible use of different location formats.

Comprehensive experiments validate the effectiveness of NExT-Chat.

Abstract

The development of large language models (LLMs) has greatly advanced the field of multimodal understanding, leading to the emergence of large multimodal models (LMMs). In order to enhance the level of visual comprehension, recent studies have equipped LMMs with region-level understanding capabilities by representing object bounding box coordinates as a series of text sequences (pix2seq). In this paper, we introduce a novel paradigm for object location modeling called pix2emb method, where we ask the LMM to output the location embeddings and then decode them with different decoders. This paradigm allows us to use different location formats (such as bounding boxes and masks) in multimodal conversations. Leveraging the proposed pix2emb method, we train an LMM named NExT-Chat and demonstrate its capability of handling multiple tasks like visual grounding, region captioning, and grounded reasoning. Comprehensive experiments show the effectiveness of our NExT-Chat on various tasks, e.g., NExT-Chat (87.7) vs. Shikra (86.9) on POPE-Random, NExT-Chat (68.9) vs. LISA (67.9) on referring expression segmentation task, and NExT-Chat (79.6) vs. Kosmos-2 (62.3) on region caption task. The code and model are released at https://github.com/NExT-ChatV/NExT-Chat.