skip to main content


Search for: All records

Creators/Authors contains: "Guo T"

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. Large Language Models (LLMs) have achieved remarkable success across a wide array of tasks. Due to the impressive planning and reasoning abilities of LLMs, they have been used as autonomous agents to do many tasks automatically. Recently, based on the development of using one LLM as a single planning or decision-making agent, LLM-based multi-agent systems have achieved considerable progress in complex problem-solving and world simulation. To provide the community with an overview of this dynamic field, we present this survey to offer an in-depth discussion on the essential aspects of multi-agent systems based on LLMs, as well as the challenges. Our goal is for readers to gain substantial insights on the following questions: What domains and environments do LLM-based multi-agents simulate? How are these agents profiled and how do they communicate? What mechanisms contribute to the growth of agents' capacities? For those interested in delving into this field of study, we also summarize the commonly used datasets or benchmarks for them to have convenient access. To keep researchers updated on the latest studies, we maintain an open-source GitHub repository, dedicated to outlining the research on LLM-based multi-agent systems. 
    more » « less
    Free, publicly-accessible full text available August 3, 2025
  2. Large Language Models (LLMs) with strong abilities in natural language processing tasks have emerged and have been applied in various kinds of areas such as science, finance and software engineering. However, the capability of LLMs to advance the field of chemistry remains unclear. In this paper, rather than pursuing state-of-the-art performance, we aim to evaluate capabilities of LLMs in a wide range of tasks across the chemistry domain. We identify three key chemistryrelated capabilities including understanding, reasoning and explaining to explore in LLMs and establish a benchmark containing eight chemistry tasks. Our analysis draws on widely recognized datasets facilitating a broad exploration of the capacities of LLMs within the context of practical chemistry. Five LLMs (GPT-4, GPT-3.5, Davinci-003, Llama and Galactica) are evaluated for each chemistry task in zero-shot and few-shot in-context learning settings with carefully selected demonstration examples and specially crafted prompts. Our investigation found that GPT-4 outperformed other models and LLMs exhibit different competitive levels in eight chemistry tasks. In addition to the key findings from the comprehensive benchmark analysis, our work provides insights into the limitation of current LLMs and the impact of in-context learning settings on LLMs’ performance across various chemistry tasks. The code and datasets used in this study are available at https://github.com/ChemFoundationModels/ChemLLMBench. 
    more » « less
    Free, publicly-accessible full text available December 16, 2024
  3. null (Ed.)
    Attention-based image classification has gained increasing popularity in recent years. State-of-the-art methods for attention-based classification typically require a large training set and operate under the assumption that the label of an image depends solely on a single object (i.e., region of interest) in the image. However, in many real-world applications (e.g., medical imaging), it is very expensive to collect a large training set. Moreover, the label of each image is usually determined jointly by multiple regions of interest (ROIs). Fortunately, for such applications, it is often possible to collect the locations of the ROIs in each training image. In this paper, we study the problem of guided multi-attention classification, the goal of which is to achieve high accuracy under the dual constraints of (1) small sample size, and (2) multiple ROIs for each image. We propose a model, called Guided Attention Recurrent Network (GARN), for multi-attention classification. Different from existing attention-based methods, GARN utilizes guidance information regarding multiple ROIs thus allowing it to work well even when sample size is small. Empirical studies on three different visual tasks show that our guided attention approach can effectively boost model performance for multi-attention image classification. 
    more » « less
  4. Free, publicly-accessible full text available April 1, 2025
  5. Free, publicly-accessible full text available May 1, 2025
  6. The evolution of underwater photogrammetry allows to realize 3D models of submerged object and structures throughout the use of rapid and efficient procedures either in terms of data acquisition and data processing. These procedures are based on solutions that are applied using natural control points, signalized markers and tie points; the most common algorithms are based on Structure from Motion (SfM) approach. The limit of these applications is sometimes due to the final accuracy, especially when the goal is a centimeter level of accuracy. This accuracy should be necessary when dealing with a survey devoted to deformation control purposes. An example is the underwater photogrammetry for the determination of coral growth; it is effectively a movement or a deformation detection issue where the geometric change is almost at centimeter or few centimeters accuracy level. When dealing with deformation control applications, a geodetic network is essential to realize a stable and unambiguous reference frame through the accurate and permanent installation of Ground Control Points (GCPs). Such a network, indeed, permits a robust reference frame for the georeferencing of images blocks in the different époques of data acquisition. Therefore, the comparison among subsequent photogrammetric restitutions is based on homogeneous 3D models that have been oriented in the same absolute reference system. The photogrammetric survey is based on a methodological approach especially adapted to underwater biometry (like coral growth determination) and to underwater archaeology. The approach is suitable both for modeling objects of relatively reduced dimensions and for structures with a length of ten meters or more, such as coral barriers, wrecks and long walls. The paper describes underwater photogrammetric surveys on sites at different extensions, the geodetic GCPs reference network installation and measurements (distance and elevation difference observations) as well as preliminary results of the network adjustment. A brief description of image acquisition at a different scales and the resulting 3D model of first campaign are also shown. 
    more » « less