Tong Bill Xu

Shukang Yin, Chaoyou Fu, Sirui Zhao et al. · tencent-ai

Recently, Multimodal Large Language Model (MLLM) represented by GPT-4V has been a new rising research hotspot, which uses powerful Large Language Models (LLMs) as a brain to perform multimodal tasks. The surprising emergent capabilities of MLLM, such as writing stories based on images and OCR-free math reasoning, are rare in traditional multimodal methods, suggesting a potential path to artificial general intelligence. To this end, both academia and industry have endeavored to develop MLLMs that can compete with or even better than GPT-4V, pushing the limit of research at a surprising speed. In this paper, we aim to trace and summarize the recent progress of MLLMs. First of all, we present the basic formulation of MLLM and delineate its related concepts, including architecture, training strategy and data, as well as evaluation. Then, we introduce research topics about how MLLMs can be extended to support more granularity, modalities, languages, and scenarios. We continue with multimodal hallucination and extended techniques, including Multimodal ICL (M-ICL), Multimodal CoT (M-CoT), and LLM-Aided Visual Reasoning (LAVR). To conclude the paper, we discuss existing challenges and point out promising research directions. In light of the fact that the era of MLLM has only just begun, we will keep updating this survey and hope it can inspire more research. An associated GitHub link collecting the latest papers is available at https://github.com/BradyFU/Awesome-Multimodal-Large-Language-Models.

Shukang Yin, Chaoyou Fu, Sirui Zhao et al.

Hallucination is a big shadow hanging over the rapidly evolving Multimodal Large Language Models (MLLMs), referring to the phenomenon that the generated text is inconsistent with the image content. In order to mitigate hallucinations, existing studies mainly resort to an instruction-tuning manner that requires retraining the models with specific data. In this paper, we pave a different way, introducing a training-free method named Woodpecker. Like a woodpecker heals trees, it picks out and corrects hallucinations from the generated text. Concretely, Woodpecker consists of five stages: key concept extraction, question formulation, visual knowledge validation, visual claim generation, and hallucination correction. Implemented in a post-remedy manner, Woodpecker can easily serve different MLLMs, while being interpretable by accessing intermediate outputs of the five stages. We evaluate Woodpecker both quantitatively and qualitatively and show the huge potential of this new paradigm. On the POPE benchmark, our method obtains a 30.66%/24.33% improvement in accuracy over the baseline MiniGPT-4/mPLUG-Owl. The source code is released at https://github.com/BradyFU/Woodpecker.

Derong Xu, Pengyue Jia, Xiaopeng Li et al.

Large language models (LLMs) have demonstrated remarkable capabilities, but still struggle with issues like hallucinations and outdated information. Retrieval-augmented generation (RAG) addresses these issues by grounding LLM outputs in external knowledge with an Information Retrieval (IR) system. Building on this foundation, graph-based RAG systems go a step further by retrieving subgraphs, which preserve the relationships between knowledge entities and provide more comprehensive context. However, graph RAG faces two challenges: (1) Retrieving relevant information introduces irrelevant nodes (especially in dense graph databases, where retrieval usually extends to adjacent nodes), and leads to overly lengthy inputs that hinder efficiency; (2) The representation gap between graph and language during generation with LLMs limits the ability to fully leverage graph structures for enhanced understanding. To address these limitations, we propose Align-GRAG, a novel reasoning-guided dual alignment framework in post-retrieval phrase. It first formulates a subgraph by retrieving nodes and edges. Then an Aligner is proposed to jointly optimize a graph encoder with an LLM-summarized reasoning chain. It achieves dual alignment of graph node and representation by leveraging KL divergence loss and contrastive loss, facilitating efficient pruning of irrelevant knowledge and establishing a unified semantic space. The Generator integrates the aligned graph data with LLM to produce coherent and accurate answers. Experiments on the GraphQA benchmark across three tasks (including common sense reasoning, scene graph understanding, and knowledge graph reasoning) validate the effectiveness of our method. The codes are available in this repository\footnote{https://anonymous.4open.science/r/Align-GRAG-F3D8/}.

Saleh Kalantari, Tong Bill Xu, Armin Mostafavi et al.

Engaging with natural environments and representations of nature has been shown to improve mood states and reduce cognitive decline in older adults. The current study evaluated the use of virtual reality (VR) for presenting immersive 360 degree nature videos and a digitally designed interactive garden for this purpose. Fifty participants (age 60 plus), with varied cognitive and physical abilities, were recruited. Data were collected through pre/post-intervention surveys, standardized observations during the interventions, and post-intervention semi structured interviews. The results indicated significant improvements in attitudes toward VR and in some aspects of mood and engagement. The responses to the environment did not significantly differ among participants with different cognitive abilities; however, those with physical disabilities expressed stronger positive reactions on some metrics compared to participants without disabilities. Almost no negative impacts (cybersickness, task frustration) were found. In the interviews some participants expressed resistance to the technology, in particular the digital garden, indicating that it felt cartoonish or unappealing and that it could not substitute for real nature. However, the majority felt that the VR experiences could be a beneficial activity in situations when real-world contact with nature was not immediately feasible.

Dan Li, Yang Yang, Hongyin Tang et al.

With the booming of pre-trained transformers, representation-based models based on Siamese transformer encoders have become mainstream techniques for efficient text matching. However, these models suffer from severe performance degradation due to the lack of interaction between the text pair, compared with interaction-based models. Prior arts attempt to address this through performing extra interaction for Siamese encoded representations, while the interaction during encoding is still ignored. To remedy this, we propose a \textit{Virtual} InteRacTion mechanism (VIRT) to transfer interactive knowledge from interaction-based models into Siamese encoders through attention map distillation. As a train-time-only component, VIRT could completely maintain the high efficiency of the Siamese structure and brings no extra computation cost during inference. To fully utilize the learned interactive knowledge, we further design a VIRT-adapted interaction strategy. Experimental results on multiple text matching datasets demonstrate that our method outperforms state-of-the-art representation-based models. What's more, VIRT can be easily integrated into existing representation-based methods to achieve further improvements.