-
Can formal argumentative reasoning enhance LLMs performances?
Authors:
Federico Castagna,
Isabel Sassoon,
Simon Parsons
Abstract:
Recent years witnessed significant performance advancements in deep-learning-driven natural language models, with a strong focus on the development and release of Large Language Models (LLMs). These improvements resulted in better quality AI-generated output but rely on resource-expensive training and upgrading of models. Although different studies have proposed a range of techniques to enhance LL…
▽ More
Recent years witnessed significant performance advancements in deep-learning-driven natural language models, with a strong focus on the development and release of Large Language Models (LLMs). These improvements resulted in better quality AI-generated output but rely on resource-expensive training and upgrading of models. Although different studies have proposed a range of techniques to enhance LLMs without retraining, none have considered computational argumentation as an option. This is a missed opportunity since computational argumentation is an intuitive mechanism that formally captures agents' interactions and the information conflict that may arise during such interplays, and so it seems well-suited for boosting the reasoning and conversational abilities of LLMs in a seamless manner. In this paper, we present a pipeline (MQArgEng) and preliminary study to evaluate the effect of introducing computational argumentation semantics on the performance of LLMs. Our experiment's goal was to provide a proof-of-concept and a feasibility analysis in order to foster (or deter) future research towards a fully-fledged argumentation engine plugin for LLMs. Exploratory results using the MT-Bench indicate that MQArgEng provides a moderate performance gain in most of the examined topical categories and, as such, show promise and warrant further research.
△ Less
Submitted 16 May, 2024;
originally announced May 2024.
-
Online Handbook of Argumentation for AI: Volume 4
Authors:
Lars Bengel,
Lydia Blümel,
Elfia Bezou-Vrakatseli,
Federico Castagna,
Giulia D'Agostino,
Isabelle Kuhlmann,
Jack Mumford,
Daphne Odekerken,
Fabrizio Russo,
Stefan Sarkadi,
Madeleine Waller,
Andreas Xydis
Abstract:
This volume contains revised versions of the papers selected for the fourth volume of the Online Handbook of Argumentation for AI (OHAAI). Previously, formal theories of argument and argument interaction have been proposed and studied, and this has led to the more recent study of computational models of argument. Argumentation, as a field within artificial intelligence (AI), is highly relevant for…
▽ More
This volume contains revised versions of the papers selected for the fourth volume of the Online Handbook of Argumentation for AI (OHAAI). Previously, formal theories of argument and argument interaction have been proposed and studied, and this has led to the more recent study of computational models of argument. Argumentation, as a field within artificial intelligence (AI), is highly relevant for researchers interested in symbolic representations of knowledge and defeasible reasoning. The purpose of this handbook is to provide an open access and curated anthology for the argumentation research community. OHAAI is designed to serve as a research hub to keep track of the latest and upcoming PhD-driven research on the theory and application of argumentation in all areas related to AI.
△ Less
Submitted 20 December, 2023;
originally announced January 2024.
-
Computational Argumentation-based Chatbots: a Survey
Authors:
Federico Castagna,
Nadin Kokciyan,
Isabel Sassoon,
Simon Parsons,
Elizabeth Sklar
Abstract:
Chatbots are conversational software applications designed to interact dialectically with users for a plethora of different purposes. Surprisingly, these colloquial agents have only recently been coupled with computational models of arguments (i.e. computational argumentation), whose aim is to formalise, in a machine-readable format, the ordinary exchange of information that characterises human co…
▽ More
Chatbots are conversational software applications designed to interact dialectically with users for a plethora of different purposes. Surprisingly, these colloquial agents have only recently been coupled with computational models of arguments (i.e. computational argumentation), whose aim is to formalise, in a machine-readable format, the ordinary exchange of information that characterises human communications. Chatbots may employ argumentation with different degrees and in a variety of manners. The present survey sifts through the literature to review papers concerning this kind of argumentation-based bot, drawing conclusions about the benefits and drawbacks that this approach entails in comparison with standard chatbots, while also envisaging possible future development and integration with the Transformer-based architecture and state-of-the-art Large Language models.
△ Less
Submitted 7 January, 2024;
originally announced January 2024.
-
A Serverless Architecture for Efficient and Scalable Monte Carlo Markov Chain Computation
Authors:
Fabio Castagna,
Alberto Trombetta,
Marco Landoni,
Stefano Andreon
Abstract:
Computer power is a constantly increasing demand in scientific data analyses, in particular when Markov Chain Monte Carlo (MCMC) methods are involved, for example for estimating integral functions or Bayesian posterior probabilities. In this paper, we describe the benefits of a parallel computation of MCMC using a cloud-based, serverless architecture: first, the computation time can be spread over…
▽ More
Computer power is a constantly increasing demand in scientific data analyses, in particular when Markov Chain Monte Carlo (MCMC) methods are involved, for example for estimating integral functions or Bayesian posterior probabilities. In this paper, we describe the benefits of a parallel computation of MCMC using a cloud-based, serverless architecture: first, the computation time can be spread over thousands of processes, hence greatly reducing the time the user should wait to have its computation completed. Second, the overhead time required for running in parallel several processes is minor and grows logarithmically with respect to the number of processes. Third, the serverless approach does not require time-consuming efforts for maintaining and updating the computing infrastructure when/if the number of walkers increases or for adapting the code to optimally use the infrastructure. The benefits are illustrated with the computation of the posterior probability distribution of a real astronomical analysis.
△ Less
Submitted 6 October, 2023;
originally announced October 2023.
-
Online Handbook of Argumentation for AI: Volume 3
Authors:
Lars Bengel,
Elfia Bezou-Vrakatseli,
Lydia Blümel,
Federico Castagna,
Giulia D'Agostino,
Daphne Odekerken,
Minal Suresh Patil,
Jordan Robinson,
Hao Wu,
Andreas Xydis
Abstract:
This volume contains revised versions of the papers selected for the third volume of the Online Handbook of Argumentation for AI (OHAAI). Previously, formal theories of argument and argument interaction have been proposed and studied, and this has led to the more recent study of computational models of argument. Argumentation, as a field within artificial intelligence (AI), is highly relevant for…
▽ More
This volume contains revised versions of the papers selected for the third volume of the Online Handbook of Argumentation for AI (OHAAI). Previously, formal theories of argument and argument interaction have been proposed and studied, and this has led to the more recent study of computational models of argument. Argumentation, as a field within artificial intelligence (AI), is highly relevant for researchers interested in symbolic representations of knowledge and defeasible reasoning. The purpose of this handbook is to provide an open access and curated anthology for the argumentation research community. OHAAI is designed to serve as a research hub to keep track of the latest and upcoming PhD-driven research on the theory and application of argumentation in all areas related to AI.
△ Less
Submitted 15 December, 2022;
originally announced December 2022.
-
The flat entropy profile at the outskirts of the Abell 2244 galaxy cluster
Authors:
S. Andreon,
A. Moretti,
H. Bohringer,
F. Castagna
Abstract:
Entropy is an advantageous diagnostics to study the thermodynamic history of the intracluster plasma of galaxy clusters. We present the entropy profile of the Abell 2244 galaxy cluster derived both exclusively using X-ray data from the low-background Swift XRT telescope and also using Planck y data. The entropy profile derivation using X-rays only is robust at least to the virial radius because th…
▽ More
Entropy is an advantageous diagnostics to study the thermodynamic history of the intracluster plasma of galaxy clusters. We present the entropy profile of the Abell 2244 galaxy cluster derived both exclusively using X-ray data from the low-background Swift XRT telescope and also using Planck y data. The entropy profile derivation using X-rays only is robust at least to the virial radius because the cluster brightness is large compared to the X-ray background at low energies, temperature is strongly bounded by the lack of cluster X-ray photons at energies kT>3 keV, and the XRT background is low, stable and understood. In the observed solid angle, about one quadrant, the entropy radial profile deviates from a power-law at the virial radius, mainly because of a sharp drop of the cluster temperature. This bending of the entropy profile is confirmed when X-ray spectral information is replaced by the Compton map. Clum** and non-thermal pressure support are insufficient to restore a power law entropy profile because they are bound to be small by: a) the agreement between mass estimates from different tracers (gas and galaxies), b) the agreement between entropy profile determinations based on combinations of observables with different sensitivities and systematics, and c) the low value of clum** as estimated using the azimuthal scatter and the gas fraction. Based on numerical simulations, ion-electron equilibration is also insufficient to restore a linear entropy profile. Therefore, the bending of the entropy profiles seems to be robustly derived and witnesses the teoretically-predicted decrease in the inflow through the virial boundary.
△ Less
Submitted 9 December, 2022;
originally announced December 2022.
-
Many-probes multi-object spatially-resolved analyses of galaxy clusters in the big data era
Authors:
Fabio Castagna,
Stefano Andreon,
Alberto Trombetta,
Marco Landoni
Abstract:
The thermal Sunyaev-Zeldovich (SZ) effect and the X-ray emission offer separate and highly complementary probes of the thermodynamics of the intracluster medium, particularly on their radial dependence. We already released \texttt{JoXSZ}, the first publicly available code designed to jointly fit SZ and X-ray data coming from various instruments to derive the thermodynamic radial profiles of galaxy…
▽ More
The thermal Sunyaev-Zeldovich (SZ) effect and the X-ray emission offer separate and highly complementary probes of the thermodynamics of the intracluster medium, particularly on their radial dependence. We already released \texttt{JoXSZ}, the first publicly available code designed to jointly fit SZ and X-ray data coming from various instruments to derive the thermodynamic radial profiles of galaxy clusters, including mass. \texttt{JoXSZ;} follows a fully Bayesian forward-modelling approach, adopts flexible parametrization for the thermodynamic profiles and includes many useful options that users can customize according to their needs. We are including shear measurement in our joint analysis, and moving from single-cluster to multi-cluster analyses, allowing to quantify the heterogeneity of thermodynamic properties within the cluster population. At the same time, we are creating a suitable framework that efficiently stores and optimally processes huge volumes of data being released by the current and new generation surveys.
△ Less
Submitted 2 November, 2021;
originally announced November 2021.
-
Thermodynamic evolution of the $z=1.75$ galaxy cluster IDCS J1426.5+3508
Authors:
S. Andreon,
C. Romero,
F. Castagna,
A. Ragagnin,
M. Devlin,
S. Dicker,
B. Mason,
T. Mroczkowski,
C. Sarazin,
J. Sievers,
S. Stanchfield
Abstract:
We present resolved thermodynamic profiles out to 500 kpc, about $r_{500}$, of the $z=1.75$ galaxy cluster IDCS J1426.5+3508 with 40 kpc resolution. Thanks to the combination of Sunyaev-Zel'dovich and X-ray datasets, IDCS J1426.5+3508 becomes the most distant cluster with resolved thermodynamic profiles. These are derived assuming a non-parametric pressure profile and a very flexible model for the…
▽ More
We present resolved thermodynamic profiles out to 500 kpc, about $r_{500}$, of the $z=1.75$ galaxy cluster IDCS J1426.5+3508 with 40 kpc resolution. Thanks to the combination of Sunyaev-Zel'dovich and X-ray datasets, IDCS J1426.5+3508 becomes the most distant cluster with resolved thermodynamic profiles. These are derived assuming a non-parametric pressure profile and a very flexible model for the electron density profile. The shape of the pressure profile is flatter than the universal pressure profile. The IDCS J1426.5+3508 temperature profile is increasing radially out to 500 kpc. To identify the possible future evolution of IDCS J1426.5+3508 , we compared it with its local descendants that numerical simulations show to be $0.65\pm0.12$ dex more massive. We found no evolution at 30 kpc, indicating a fine tuning between cooling and heating at small radii. At $30<r<300$ kpc, our observations show that entropy and heat must be deposited with little net gas transfer, while at 500 kpc the gas need to be replaced by a large amount of cold, lower entropy gas, consistent with theoretical expectation of a filamentary gas stream, which brings low entropy gas to 500 kpc and energy at even smaller radii. At $r \gtrsim 400$ kpc the polytropic index takes a low value, which indicates the presence of a large amount of non-thermal pressure. Our work also introduces a new definition of the evolutionary rate, which uses unscaled radii, unscaled thermodynamic quantities, and different masses at different redshifts to compare ancestors and descendants. It has the advantage of separating cluster evolution, dependence on mass, pseudo-evolution and returns a number with unique interpretation, unlike other definitions used in literature.
△ Less
Submitted 23 June, 2021; v1 submitted 21 June, 2021;
originally announced June 2021.
-
Online Handbook of Argumentation for AI: Volume 2
Authors:
OHAAI Collaboration,
Andreas Brannstrom,
Federico Castagna,
Theo Duchatelle,
Matt Foulis,
Timotheus Kampik,
Isabelle Kuhlmann,
Lars Malmqvist,
Mariela Morveli-Espinoza,
Jack Mumford,
Stipe Pandzic,
Robin Schaefer,
Luke Thorburn,
Andreas Xydis,
Antonio Yuste-Ginel,
Heng Zheng
Abstract:
This volume contains revised versions of the papers selected for the second volume of the Online Handbook of Argumentation for AI (OHAAI). Previously, formal theories of argument and argument interaction have been proposed and studied, and this has led to the more recent study of computational models of argument. Argumentation, as a field within artificial intelligence (AI), is highly relevant for…
▽ More
This volume contains revised versions of the papers selected for the second volume of the Online Handbook of Argumentation for AI (OHAAI). Previously, formal theories of argument and argument interaction have been proposed and studied, and this has led to the more recent study of computational models of argument. Argumentation, as a field within artificial intelligence (AI), is highly relevant for researchers interested in symbolic representations of knowledge and defeasible reasoning. The purpose of this handbook is to provide an open access and curated anthology for the argumentation research community. OHAAI is designed to serve as a research hub to keep track of the latest and upcoming PhD-driven research on the theory and application of argumentation in all areas related to AI.
△ Less
Submitted 23 June, 2021; v1 submitted 16 June, 2021;
originally announced June 2021.
-
Online Handbook of Argumentation for AI: Volume 1
Authors:
OHAAI Collaboration,
Federico Castagna,
Timotheus Kampik,
Atefeh Keshavarzi Zafarghandi,
Mickaël Lafages,
Jack Mumford,
Christos T. Rodosthenous,
Samy Sá,
Stefan Sarkadi,
Joseph Singleton,
Kenneth Skiba,
Andreas Xydis
Abstract:
This volume contains revised versions of the papers selected for the first volume of the Online Handbook of Argumentation for AI (OHAAI). Previously, formal theories of argument and argument interaction have been proposed and studied, and this has led to the more recent study of computational models of argument. Argumentation, as a field within artificial intelligence (AI), is highly relevant for…
▽ More
This volume contains revised versions of the papers selected for the first volume of the Online Handbook of Argumentation for AI (OHAAI). Previously, formal theories of argument and argument interaction have been proposed and studied, and this has led to the more recent study of computational models of argument. Argumentation, as a field within artificial intelligence (AI), is highly relevant for researchers interested in symbolic representations of knowledge and defeasible reasoning. The purpose of this handbook is to provide an open access and curated anthology for the argumentation research community. OHAAI is designed to serve as a research hub to keep track of the latest and upcoming PhD-driven research on the theory and application of argumentation in all areas related to AI.
△ Less
Submitted 22 June, 2020;
originally announced June 2020.
-
JoXSZ: Joint X-SZ fitter for galaxy clusters
Authors:
Fabio Castagna,
Stefano Andreon
Abstract:
High-resolution observations of the thermal Sunyaev-Zeldovich (SZ) effect and of the X-ray emission of galaxy clusters are becoming more and more widespread, offering us an unique asset to the study of the thermodynamic properties of the intracluster medium. We present JoXSZ, a Bayesian forward-modelling Python code designed to jointly fit the SZ data and the three dimensional X-ray data cube. JoX…
▽ More
High-resolution observations of the thermal Sunyaev-Zeldovich (SZ) effect and of the X-ray emission of galaxy clusters are becoming more and more widespread, offering us an unique asset to the study of the thermodynamic properties of the intracluster medium. We present JoXSZ, a Bayesian forward-modelling Python code designed to jointly fit the SZ data and the three dimensional X-ray data cube. JoXSZ is able to derive the thermodynamic profiles of galaxy clusters for the first time making full and consistent use of all the information contained in the observations. JoXSZ will be publicly available on GitHub in the near future.
△ Less
Submitted 30 May, 2020;
originally announced June 2020.
-
JoXSZ: Joint X-SZ fitting code for galaxy clusters
Authors:
Fabio Castagna,
Stefano Andreon
Abstract:
The thermal Sunyaev-Zeldovich (SZ) effect and the X-ray emission offer separate and highly complementary probes of the thermodynamics of the intracluster medium. We present JoXSZ, the first publicly available code designed to jointly fit SZ and X-ray data coming from various instruments to derive the thermodynamic profiles of galaxy clusters. JoXSZ follows a fully Bayesian forward-modelling approa…
▽ More
The thermal Sunyaev-Zeldovich (SZ) effect and the X-ray emission offer separate and highly complementary probes of the thermodynamics of the intracluster medium. We present JoXSZ, the first publicly available code designed to jointly fit SZ and X-ray data coming from various instruments to derive the thermodynamic profiles of galaxy clusters. JoXSZ follows a fully Bayesian forward-modelling approach, accounts for the SZ calibration uncertainty and X-ray background level systematic. It improves upon most state-of-the-art, and not publicly available, analyses because it adopts the correct Poisson-Gauss expression for the joint likelihood, makes full use of the information contained in the observations, even in the case of missing values within the datasets, has a more inclusive error budget, and adopts a consistent temperature across the various parts of the code, allowing for differences between X-ray and SZ gas mass weighted temperatures when required by the user. JoXSZ accounts for beam smearing and data analysis transfer function, accounts for the temperature and metallicity dependencies of the SZ and X-ray conversion factors, adopts flexible parametrization for the thermodynamic profiles, and on user request allows either adopting or relaxing the assumption of hydrostatic equilibrium (HE). When HE holds, JoXSZ uses a physical (positive) prior on the radial derivative of the enclosed mass and derives the mass profile and overdensity radii $r_Δ$. For these reasons, JoXSZ goes beyond simple SZ and electron density fits. We illustrate the use of JoXSZ by combining Chandra and NIKA data on the high-redshift cluster CL J1226.9+3332. The code is written in Python, it is fully documented and the users are free to customize their analysis in accordance with their needs and requirements. JoXSZ is publicly available on GitHub.
△ Less
Submitted 1 June, 2020; v1 submitted 25 May, 2020;
originally announced May 2020.
-
PreProFit -- Pressure Profile Fitter for galaxy clusters
Authors:
Fabio Castagna,
Stefano Andreon
Abstract:
Galaxy cluster analyses based on high-resolution observations of the Sunyaev-Zeldovich (SZ) effect have become common in the last decade. We present PreProFit, the first publicly available code designed to fit the pressure profile of galaxy clusters from SZ data. PreProFit is based on a Bayesian forward-modelling approach, allows the analysis of data coming from different sources, adopts a flexibl…
▽ More
Galaxy cluster analyses based on high-resolution observations of the Sunyaev-Zeldovich (SZ) effect have become common in the last decade. We present PreProFit, the first publicly available code designed to fit the pressure profile of galaxy clusters from SZ data. PreProFit is based on a Bayesian forward-modelling approach, allows the analysis of data coming from different sources, adopts a flexible parametrization for the pressure profile, and fits the model to the data accounting for Abel integral, beam smearing, and transfer function filtering. PreProFit is computationally efficient, is extensively documented, has been released as an open source Python project, and was developed to be part of a joint analysis of X-ray and SZ data on galaxy clusters. PreProFit returns $χ^2$, model parameters and uncertainties, marginal and joint probability contours, diagnostic plots, and surface brightness radial profiles. PreProFit also allows the use of analytic approximations for the beam and transfer functions useful for feasibility studies.
△ Less
Submitted 15 October, 2019;
originally announced October 2019.
