arXiv subjects

Stephen Cranefield

Publications and source records attributed to Stephen Cranefield.

At least 19 recordsLinked to original sources

arXiv2026

Designing for Accountable Agents: a Viewpoint

AI systems are becoming increasingly complex, ubiquitous and autonomous, leading to increasing concerns about their impacts on individuals and society. In response, researchers have begun investigating how to ensure that the methods underlying AI decision-making are transparent and their decisions are explainable to people and conformant to human values and ethical principles. As part of this research thrust, the need for accountability within AI systems has been noted, but this notion has proven elusive to define; we aim to address this issue in the current paper. Unlike much recent work, we do not address accountability within the human organisational processes of developing and deploying AI; rather we consider what it would it mean for the agents within a multi-agent system (MAS), potentially including human agents, to be accountable to other agents or to have others accountable to them. In this work, we make the following contributions: we provide an in-depth survey of existing work on accountability in multiple disciplines, seeking to identify a coherent definition of the concept; we give a realistic example of a multi-agent system application domain that illustrates the benefits of enabling agents to follow accountability processes, and we identify a set of research challenges for the MAS community in building accountable agents, sketching out some initial solutions to these, thereby laying out a road-map for future research. Our focus is on laying the groundwork to enable autonomous elements within open socio-technical systems to take part in accountability processes.

cs.MA↗

arXiv2026

Evaluating LLM Alignment With Human Trust Models

Trust plays a pivotal role in enabling effective cooperation, reducing uncertainty, and guiding decision-making in both human interactions and multi-agent systems. Although it is significant, there is limited understanding of how large language models (LLMs) internally conceptualize and reason about trust. This work presents a white-box analysis of trust representation in EleutherAI/gpt-j-6B, using contrastive prompting to generate embedding vectors within the activation space of the LLM for diadic trust and related interpersonal relationship attributes. We first identified trust-related concepts from five established human trust models. We then determined a threshold for significant conceptual alignment by computing pairwise cosine similarities across 60 general emotional concepts. Then we measured the cosine similarities between the LLM's internal representation of trust and the derived trust-related concepts. Our results show that the internal trust representation of EleutherAI/gpt-j-6B aligns most closely with the Castelfranchi socio-cognitive model, followed by the Marsh Model. These findings indicate that LLMs encode socio-cognitive constructs in their activation space in ways that support meaningful comparative analyses, inform theories of social cognition, and support the design of human-AI collaborative systems.

cs.MA↗

arXiv2025

Can LLMs Reason About Trust?: A Pilot Study

In human society, trust is an essential component of social attitude that helps build and maintain long-term, healthy relationships which creates a strong foundation for cooperation, enabling individuals to work together effectively and achieve shared goals. As many human interactions occur through electronic means such as using mobile apps, the potential arises for AI systems to assist users in understanding the social state of their relationships. In this paper we investigate the ability of Large Language Models (LLMs) to reason about trust between two individuals in an environment which requires fostering trust relationships. We also assess whether LLMs are capable of inducing trust by role-playing one party in a trust based interaction and planning actions which can instil trust.

cs.HC↗

arXiv2025

Evolution of Cooperation in LLM-Agent Societies: A Preliminary Study Using Different Punishment Strategies

The evolution of cooperation has been extensively studied using abstract mathematical models and simulations. Recent advances in Large Language Models (LLMs) and the rise of LLM agents have demonstrated their ability to perform social reasoning, thus providing an opportunity to test the emergence of norms in more realistic agent-based simulations with human-like reasoning using natural language. In this research, we investigate whether the cooperation dynamics presented in Boyd and Richerson's model persist in a more realistic simulation of the Diner's Dilemma using LLM agents compared to the abstract mathematical nature in the work of Boyd and Richerson. Our findings indicate that agents follow the strategies defined in the Boyd and Richerson model, and explicit punishment mechanisms drive norm emergence, reinforcing cooperative behaviour even when the agent strategy configuration varies. Our results suggest that LLM-based Multi-Agent System simulations, in fact, can replicate the evolution of cooperation predicted by the traditional mathematical models. Moreover, our simulations extend beyond the mathematical models by integrating natural language-driven reasoning and a pairwise imitation method for strategy adoption, making them a more realistic testbed for cooperative behaviour in MASs.

cs.MA↗

arXiv2024

Norm Violation Detection in Multi-Agent Systems using Large Language Models: A Pilot Study

Norms are an important component of the social fabric of society by prescribing expected behaviour. In Multi-Agent Systems (MAS), agents interacting within a society are equipped to possess social capabilities such as reasoning about norms and trust. Norms have long been of interest within the Normative Multi-Agent Systems community with researchers studying topics such as norm emergence, norm violation detection and sanctioning. However, these studies have some limitations: they are often limited to simple domains, norms have been represented using a variety of representations with no standard approach emerging, and the symbolic reasoning mechanisms generally used may suffer from a lack of extensibility and robustness. In contrast, Large Language Models (LLMs) offer opportunities to discover and reason about norms across a large range of social situations. This paper evaluates the capability of LLMs to detecting norm violations. Based on simulated data from 80 stories in a household context, with varying complexities, we investigated whether 10 norms are violated. For our evaluations we first obtained the ground truth from three human evaluators for each story. Then, the majority result was compared against the results from three well-known LLM models (Llama 2 7B, Mixtral 7B and ChatGPT-4). Our results show the promise of ChatGPT-4 for detecting norm violations, with Mixtral some distance behind. Also, we identify areas where these models perform poorly and discuss implications for future work.

cs.MA↗

arXiv2024

Harnessing the power of LLMs for normative reasoning in MASs

Software agents, both human and computational, do not exist in isolation and often need to collaborate or coordinate with others to achieve their goals. In human society, social mechanisms such as norms ensure efficient functioning, and these techniques have been adopted by researchers in multi-agent systems (MAS) to create socially aware agents. However, traditional techniques have limitations, such as operating in limited environments often using brittle symbolic reasoning. The advent of Large Language Models (LLMs) offers a promising solution, providing a rich and expressive vocabulary for norms and enabling norm-capable agents that can perform a range of tasks such as norm discovery, normative reasoning and decision-making. This paper examines the potential of LLM-based agents to acquire normative capabilities, drawing on recent Natural Language Processing (NLP) and LLM research. We present our vision for creating normative LLM agents. In particular, we discuss how the recently proposed "LLM agent" approaches can be extended to implement such normative LLM agents. We also highlight challenges in this emerging field. This paper thus aims to foster collaboration between MAS, NLP and LLM researchers in order to advance the field of normative agents.

cs.AI↗

arXiv2024

Ultra Low-Cost Two-Stage Multimodal System for Non-Normative Behavior Detection

The online community has increasingly been inundated by a toxic wave of harmful comments. In response to this growing challenge, we introduce a two-stage ultra-low-cost multimodal harmful behavior detection method designed to identify harmful comments and images with high precision and recall rates. We first utilize the CLIP-ViT model to transform tweets and images into embeddings, effectively capturing the intricate interplay of semantic meaning and subtle contextual clues within texts and images. Then in the second stage, the system feeds these embeddings into a conventional machine learning classifier like SVM or logistic regression, enabling the system to be trained rapidly and to perform inference at an ultra-low cost. By converting tweets into rich multimodal embeddings through the CLIP-ViT model and utilizing them to train conventional machine learning classifiers, our system is not only capable of detecting harmful textual information with near-perfect performance, achieving precision and recall rates above 99\% but also demonstrates the ability to zero-shot harmful images without additional training, thanks to its multimodal embedding input. This capability empowers our system to identify unseen harmful images without requiring extensive and costly image datasets. Additionally, our system quickly adapts to new harmful content; if a new harmful content pattern is identified, we can fine-tune the classifier with the corresponding tweets' embeddings to promptly update the system. This makes it well suited to addressing the ever-evolving nature of online harmfulness, providing online communities with a robust, generalizable, and cost-effective tool to safeguard their communities.

cs.MA↗

arXiv2022

Variational Transfer Learning using Cross-Domain Latent Modulation

To successfully apply trained neural network models to new domains, powerful transfer learning solutions are essential. We propose to introduce a novel cross-domain latent modulation mechanism to a variational autoencoder framework so as to achieve effective transfer learning. Our key idea is to procure deep representations from one data domain and use it to influence the reparameterization of the latent variable of another domain. Specifically, deep representations of the source and target domains are first extracted by a unified inference model and aligned by employing gradient reversal. The learned deep representations are then cross-modulated to the latent encoding of the alternative domain, where consistency constraints are also applied. In the empirical validation that includes a number of transfer learning benchmark tasks for unsupervised domain adaptation and image-to-image translation, our model demonstrates competitive performance, which is also supported by evidence obtained from visualization.

cs.LG↗

arXiv2021

Towards offensive language detection and reduction in four Software Engineering communities

Software Engineering (SE) communities such as Stack Overflow have become unwelcoming, particularly through members' use of offensive language. Research has shown that offensive language drives users away from active engagement within these platforms. This work aims to explore this issue more broadly by investigating the nature of offensive language in comments posted by users in four prominent SE platforms - GitHub, Gitter, Slack and Stack Overflow (SO). It proposes an approach to detect and classify offensive language in SE communities by adopting natural language processing and deep learning techniques. Further, a Conflict Reduction System (CRS), which identifies offence and then suggests what changes could be made to minimize offence has been proposed. Beyond showing the prevalence of offensive language in over 1 million comments from four different communities which ranges from 0.07% to 0.43%, our results show promise in successful detection and classification of such language. The CRS system has the potential to drastically reduce manual moderation efforts to detect and reduce offence in SE communities.

cs.SE↗

arXiv2021

Solving social dilemmas by reasoning about expectations

It has been argued that one role of social constructs, such as institutions, trust and norms, is to coordinate the expectations of autonomous entities in order to resolve collective action situations (such as collective risk dilemmas) through the coordination of behaviour. While much work has addressed the formal representation of these social constructs, in this paper we focus specifically on the formal representation of, and associated reasoning with, the expectations themselves. In particular, we investigate how explicit reasoning about expectations can be used to encode both traditional game theory solution concepts and social mechanisms for the social dilemma situation. We use the Collective Action Simulation Platform (CASP) to model a collective risk dilemma based on a flood plain scenario and show how using expectations in the reasoning mechanisms of the agents making decisions supports the choice of cooperative behaviour.

cs.MA↗

arXiv2021

A Bayesian model of information cascades

An information cascade is a circumstance where agents make decisions in a sequential fashion by following other agents. Bikhchandani et al., predict that once a cascade starts it continues, even if it is wrong, until agents receive an external input such as public information. In an information cascade, even if an agent has its own personal choice, it is always overridden by observation of previous agents' actions. This could mean agents end up in a situation where they may act without valuing their own information. As information cascades can have serious social consequences, it is important to have a good understanding of what causes them. We present a detailed Bayesian model of the information gained by agents when observing the choices of other agents and their own private information. Compared to prior work, we remove the high impact of the first observed agent's action by incorporating a prior probability distribution over the information of unobserved agents and investigate an alternative model of choice to that considered in prior work: weighted random choice. Our results show that, in contrast to Bikhchandani's results, cascades will not necessarily occur and adding prior agents' information will delay the effects of cascades.

cs.MA↗

arXiv2020

Cross-Domain Latent Modulation for Variational Transfer Learning

We propose a cross-domain latent modulation mechanism within a variational autoencoders (VAE) framework to enable improved transfer learning. Our key idea is to procure deep representations from one data domain and use it as perturbation to the reparameterization of the latent variable in another domain. Specifically, deep representations of the source and target domains are first extracted by a unified inference model and aligned by employing gradient reversal. Second, the learned deep representations are cross-modulated to the latent encoding of the alternate domain. The consistency between the reconstruction from the modulated latent encoding and the generation using deep representation samples is then enforced in order to produce inter-class alignment in the latent space. We apply the proposed model to a number of transfer learning tasks including unsupervised domain adaptation and image-toimage translation. Experimental results show that our model gives competitive performance.

cs.LG↗

arXiv2020

Deep Adversarial Transition Learning using Cross-Grafted Generative Stacks

Current deep domain adaptation methods used in computer vision have mainly focused on learning discriminative and domain-invariant features across different domains. In this paper, we present a novel "deep adversarial transition learning" (DATL) framework that bridges the domain gap by projecting the source and target domains into intermediate, transitional spaces through the employment of adjustable, cross-grafted generative network stacks and effective adversarial learning between transitions. Specifically, we construct variational auto-encoders (VAE) for the two domains, and form bidirectional transitions by cross-grafting the VAEs' decoder stacks. Furthermore, generative adversarial networks (GAN) are employed for domain adaptation, mapping the target domain data to the known label space of the source domain. The overall adaptation process hence consists of three phases: feature representation learning by VAEs, transitions generation, and transitions alignment by GANs. Experimental results demonstrate that our method outperforms the state-of-the art on a number of unsupervised domain adaptation benchmarks.

cs.CV↗

arXiv2020

Integrating Industrial Artifacts and Agents Through Apache Camel

There are many challenges for building up the smart factory, among them to deal with distributed data, high volume of information, and wide diversity of devices and applications. In this sense, Cyber-Physical System (CPS) concept emerges to virtualize and integrate factory resources. Based on studies that use Multi-Agent System as the core of a CPS, in this paper, we show that many resources of the factories can be modelled following the well-known Agents and Artifacts method of integrating agents and their environment. To enhance the interoperability of this system, we use Apache Camel framework, a middleware to define routes allowing the integration with a wide range of endpoints using different protocols. Finally, we present a Camel component for artifacts, designed in this research, illustrating its use.

cs.MA↗

arXiv2020

Norm violation in online communities -- A study of Stack Overflow comments

Norms are behavioral expectations in communities. Online communities are also expected to abide by the rules and regulations that are expressed in the code of conduct of a system. Even though community authorities continuously prompt their users to follow the regulations, it is observed that hate speech and abusive language usage are on the rise. In this paper, we quantify and analyze the patterns of violations of normative behaviour among the users of Stack Overflow (SO) - a well-known technical question-answer site for professionals and enthusiast programmers, while posting a comment. Even though the site has been dedicated to technical problem solving and debugging, hate speech as well as posting offensive comments make the community "toxic". By identifying and minimising various patterns of norm violations in different SO communities, the community would become less toxic and thereby the community can engage more effectively in its goal of knowledge sharing. Moreover, through automatic detection of such comments, the authors can be warned by the moderators, so that it is less likely to be repeated, thereby the reputation of the site and community can be improved. Based on the comments extracted from two different data sources on SO, this work first presents a taxonomy of norms that are violated. Second, it demonstrates the sanctions for certain norm violations. Third, it proposes a recommendation system that can be used to warn users that they are about to violate a norm. This can help achieve norm adherence in online communities.

cs.SI↗

arXiv2020

Mining International Political Norms from the GDELT Database

Researchers have long been interested in the role that norms can play in governing agent actions in multi-agent systems. Much work has been done on formalising normative concepts from human society and adapting them for the government of open software systems, and on the simulation of normative processes in human and artificial societies. However, there has been comparatively little work on applying normative MAS mechanisms to understanding the norms in human society. This work investigates this issue in the context of international politics. Using the GDELT dataset, containing machine-encoded records of international events extracted from news reports, we extracted bilateral sequences of inter-country events and applied a Bayesian norm mining mechanism to identify norms that best explained the observed behaviour. A statistical evaluation showed that the normative model fitted the data significantly better than a probabilistic discrete event model.

cs.AI↗

arXiv2019

Incorporating social practices in BDI agent systems

When agents interact with humans, either through embodied agents or because they are embedded in a robot, it would be easy if they could use fixed interaction protocols as they do with other agents. However, people do not keep fixed protocols in their day-to-day interactions and the environments are often dynamic, making it impossible to use fixed protocols. Deliberating about interactions from fundamentals is not very scalable either, because in that case all possible reactions of a user have to be considered in the plans. In this paper we argue that social practices can be used as an inspiration for designing flexible and scalable interaction mechanisms that are also robust. However, using social practices requires extending the traditional BDI deliberation cycle to monitor landmark states and perform expected actions by leveraging existing plans. We define and implement this mechanism in Jason using a periodically run meta-deliberation plan, supported by a metainterpreter, and illustrate its use in a realistic scenario.

cs.MA↗

arXiv2019

Unsupervised Domain Adaptation using Deep Networks with Cross-Grafted Stacks

Current deep domain adaptation methods used in computer vision have mainly focused on learning discriminative and domain-invariant features across different domains. In this paper, we present a novel approach that bridges the domain gap by projecting the source and target domains into a common association space through an unsupervised ``cross-grafted representation stacking'' (CGRS) mechanism. Specifically, we construct variational auto-encoders (VAE) for the two domains, and form bidirectional associations by cross-grafting the VAEs' decoder stacks. Furthermore, generative adversarial networks (GAN) are employed for label alignment (LA), mapping the target domain data to the known label space of the source domain. The overall adaptation process hence consists of three phases: feature representation learning by VAEs, association generation, and association label alignment by GANs. Experimental results demonstrate that our CGRS-LA approach outperforms the state-of-the-art on a number of unsupervised domain adaptation benchmarks.

cs.CV↗