MCMark James Carman

Mark James Carman

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Mark James Carman has a total of 126 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    L1RA: Dynamic Rank Assignment in LoRA Fine-Tuning

    by Raul Singh, Nicolo Brunello, Vincenzo Scotti, Mark James Carman

    The ability of Large Language Models (LLMs) to solve complex tasks has made them crucial in the development of AI-based applications. However, the high computational requirements to fine-tune these LLMs on downstream tasks pose significant challenges, particularly when resources are limited. In response to this challenge, we introduce L1RA, a novel technique aimed at dynamically distributing the rank of low-rank adapters during fine-tuning using LoRA. Given a rank budget (i.e., total sum of adapters rank), L1RA leverages L1 regularisation to prune redundant ranks and redistribute them across adapters, thereby optimising resource utilisation. Through a series of comprehensive experiments, we empirically demonstrate that L1RA maintains comparable or even reduced computational overhead compared to other LoRA variants, including the vanilla approach, while achieving same or better performances. Moreover, the post-training analysis of rank distribution unveiled insights into the specific model components requiring the most adaptation to align with the task objective: the feed-forward layers and the attention output projection. These results highlight the efficacy of L1RA in not only enhancing the efficiency of LLM fine-tuning, but also in providing valuable diagnostic information for model refinement and customisation. In conclusion, L1RA stands as a promising technique for advancing the performance and interpretability of LLM adaptation, particularly in scenarios where computational resources are constrained.

  • Open Access

    InTraVisTo: Inside Transformer Visualisation Tool

    by Nicolò Brunello, Davide Rigamonti, Andrea Sassella, Vincenzo Scotti, Mark James Carman

    The reasoning capabilities of Large Language Models (LLMs) have increased greatly over the last few years, as have their size and complexity. Nonetheless, the use of LLMs in production remains challenging due to their unpredictable nature and discrepancies that can exist between their desired behavior and their actual model output. In this paper, we introduce a new tool, InTraVisTo (Inside Transformer Visualisation Tool), designed to enable researchers to investigate and trace the computational process that generates each token in a Transformer-based LLM. InTraVisTo provides a visualization of both the internal state of the Transformer model (by decoding token embeddings at each layer of the model) and the information flow between the various components across the different layers of the model (using a Sankey diagram). With InTraVisTo, we aim to help researchers and practitioners better understand the computations being performed within the Transformer model and thus to shed some light on internal patterns and reasoning processes employed by LLMs.

  • Open Access

    A Citizen Science Approach for Analyzing Social Media With Crowdsourcing

    by Carlo Bono, Mehmet Oğuz Mülâyim, Cinzia Cappiello, Mark James Carman, Jesus Cerquides, Jose Luis Fernandez-Marquez, Maria Rosa Mondardini, Edoardo Ramalli, Barbara Pernici

    Social media have the potential to provide timely information about emergency situations and sudden events. However, finding relevant information among the millions of posts being added every day can be difficult, and in current approaches developing an automatic data analysis project requires time and technical skills. This work presents a new approach for the analysis of social media posts, based on configurable automatic classification combined with Citizen Science methodologies. The process is facilitated by a set of flexible, automatic and open-source data processing tools called the Citizen Science Solution Kit. The kit provides a comprehensive set of tools that can be used and personalized in different situations, particularly during natural emergencies, starting from images and text contained in the posts. The tools can be employed by citizen scientists for filtering, classifying, and geolocating the content with a human-in-the-loop approach to support the data analyst, including feedback and suggestions on how to configure the automated tools, and techniques to gather inputs from citizens. Using flooding scenario as a guiding example, this paper illustrates the structure and functioning of the different tools proposed to support citizens scientists in their projects, and a methodological approach to their use. The process is then validated by discussing three case studies based on the Albania earthquake of 2019, the Covid-19 pandemic, and the Thailand floods of 2021. The results suggest that a flexible approach to tools composition and configuration can support a timely setup of an analysis project by citizen scientists, especially in case of emergencies in unexpected locations.


  • Open Access

    GeMI: interactive interface for transformer-based Genomic Metadata Integration

    by Giuseppe Serna Garcia, Michele Leone, Anna Bernasconi, Mark J Carman

    The Gene Expression Omnibus (GEO) is a public archive containing >4 million digital samples from functional genomics experiments collected over almost two decades. The accompanying metadata describing the experiments suffer from redundancy, inconsistency and incompleteness due to the prevalence of free text and the lack of well-defined data formats and their validation. To remedy this situation, we created Genomic Metadata Integration (GeMI; http://gmql.eu/gemi/), a web application that learns to automatically extract structured metadata (in the form of key-value pairs) from the plain text descriptions of GEO experiments. The extracted information can then be indexed for structured search and used for various downstream data mining activities. GeMI works in continuous interaction with its users. The natural language processing transformer-based model at the core of our system is a fine-tuned version of the Generative Pre-trained Transformer 2 (GPT2) model that is able to learn continuously from the feedback of the users thanks to an active learning framework designed for the purpose. As a part of such a framework, a machine learning interpretation mechanism (that exploits saliency maps) allows the users to understand easily and quickly whether the predictions of the model are correct and improves the overall usability. GeMI’s ability to extract attributes not explicitly mentioned (such as sex, tissue type, cell type, ethnicity and disease) allows researchers to perform specific queries and classification of experiments, which was previously possible only after spending time and resources with tedious manual annotation. The usefulness of GeMI is demonstrated on practical research use cases. Database URL http://gmql.eu/gemi/


  • Open Access

    CDF Transform-and-Shift: An effective way to deal with datasets of inhomogeneous cluster densities

    by Ye Zhu, Kai Ming Ting, Mark Carman, Maia Angelova

    The problem of inhomogeneous cluster densities has been a long-standing issue for distance-based and density-based algorithms in clustering and anomaly detection. These algorithms implicitly assume that all clusters have approximately the same density. As a result, they often exhibit a bias towards dense clusters in the presence of sparse clusters. Many remedies have been suggested; yet, we show that they are partial solutions which do not address the issue satisfactorily. To match the implicit assumption, we propose to transform a given dataset such that the transformed clusters have approximately the same density while all regions of locally low density become globally low density -- homogenising cluster density while preserving the cluster structure of the dataset. We show that this can be achieved by using a new multi-dimensional Cumulative Distribution Function in a transform-and-shift method. The method can be applied to every dataset, before the dataset is used in many existing algorithms to match their implicit assumption without algorithmic modification. We show that the proposed method performs better than existing remedies.

  • Automated Integration of Genomic Metadata with Sequence-to-Sequence Models

    by Giuseppe Cannizzaro, Michele Leone, Anna Bernasconi, Arif Canakoglu, Mark J. Carman