Search

Auditing Large Language Models (LLMs) to discover their biases and preferences is an emerging challenge in creating Responsible Artificial Intelligence (AI). While various methods have been proposed to elicit the preferences of such models, countermeasures have been taken by LLM trainers, such that LLMs hide, obfuscate or point blank refuse to disclosure their positions on certain subjects. This paper presents PRISM, a flexible, inquiry-based methodology for auditing LLMs - that seeks to illicit such positions indirectly through task-based inquiry prompting rather than direct inquiry of said preferences. To demonstrate the utility of the methodology, we applied PRISM on the Political Compass Test, where we assessed the political leanings of twenty-one LLMs from seven providers. We show LLMs, by default, espouse positions that are economically left and socially liberal (consistent with prior work). We also show the space of positions that these models are willing to espouse - where some models are more constrained and less compliant than others - while others are more neutral and objective. In sum, PRISM can more reliably probe and audit LLMs to understand their preferences, biases and constraints.

Attributing answers to source documents is an approach used to enhance the verifiability of a model's output in retrieval augmented generation (RAG). Prior work has mainly focused on improving and evaluating the attribution quality of large language models (LLMs) in RAG, but this may come at the expense of inducing biases in the attribution of answers. We define and examine two aspects in the evaluation of LLMs in RAG pipelines, namely attribution sensitivity and bias with respect to authorship information. We explicitly inform an LLM about the authors of source documents, instruct it to attribute its answers, and analyze (i) how sensitive the LLM's output is to the author of source documents, and (ii) whether the LLM exhibits a bias towards human-written or AI-generated source documents. We design an experimental setup in which we use counterfactual evaluation to study three LLMs in terms of their attribution sensitivity and bias in RAG pipelines. Our results show that adding authorship information to source documents can significantly change the attribution quality of LLMs by 3% to 18%. Moreover, we show that LLMs can have an attribution bias towards explicit human authorship, which can serve as a competing hypothesis for findings of prior work that shows that LLM-generated content may be preferred over human-written contents. Our findings indicate that metadata of source documents can influence LLMs' trust, and how they attribute their answers. Furthermore, our research highlights attribution bias and sensitivity as a novel aspect of brittleness in LLMs.

One can find various temporal deontic logics in literature, most focusing on discrete time. The literature on real-time constraints and deontic norms is much sparser. Thus, many analysis techniques which have been developed for deontic logics have not been considered for continuous time. In this paper we focus on the notion of conflict analysis which has been extensively studied for discrete time deontic logics. We present a sound, but not complete algorithm for detecting conflicts in timed contract automata and prove the correctness of the algorithm, illustrating the analysis on a case study., Comment: To be published in JURIX 2024

The morphologies of astronomical sources are highly complex, making it essential not only to classify the identified sources into their predefined categories but also to determine the sources that are most similar to a given query source. Image-based retrieval is essential, as it allows an astronomer with a source under study to ask a computer to sift through the large archived database of sources to find the most similar ones. This is of particular interest if the source under study does not fall into a "known" category (anomalous). Our work uses the trainable COSFIRE (Combination of Shifted Filter Responses) approach for image retrieval. COSFIRE filters are automatically configured to extract the hyperlocal geometric arrangements that uniquely describe the morphological characteristics of patterns of interest in a given image; in this case astronomical sources. This is achieved by automatically examining the shape properties of a given prototype source in an image, which ultimately determines the selectivity of a COSFIRE filter. We further utilize hashing techniques, which are efficient in terms of required computation and storage, enabling scalability in handling large data sets in the image retrieval process. We evaluated the effectiveness of our approach by conducting experiments on a benchmark data set of radio galaxies, containing 1,180 training images and 404 test images. Notably, our approach achieved a mean average precision of 91% for image retrieval, surpassing the performance of the competing DenseNet-based method. Moreover, the COSFIRE filters are significantly more computationally efficient, requiring $\sim\!14\times$ fewer operations than the DenseNet-based method., Comment: 11 pages, 7 figures

We introduce a novel computational unit, termed PushPull-Conv, in the first layer of a ResNet architecture, inspired by the anti-phase inhibition phenomenon observed in the primary visual cortex. This unit redefines the traditional convolutional layer by implementing a pair of complementary filters: a trainable push kernel and its counterpart, the pull kernel. The push kernel (analogous to traditional convolution) learns to respond to specific stimuli, while the pull kernel reacts to the same stimuli but of opposite contrast. This configuration enhances stimulus selectivity and effectively inhibits response in regions lacking preferred stimuli. This effect is attributed to the push and pull kernels, which produce responses of comparable magnitude in such regions, thereby neutralizing each other. The incorporation of the PushPull-Conv into ResNets significantly increases their robustness to image corruption. Our experiments with benchmark corruption datasets show that the PushPull-Conv can be combined with other data augmentation techniques to further improve model robustness. We set a new robustness benchmark on ResNet50 achieving an $mCE$ of 49.95$\%$ on ImageNet-C when combining PRIME augmentation with PushPull inhibition., Comment: Accepted at ICPR 2024, code available at https://github.com/bgswaroop/pushpull-conv

Incomplete relevance judgments limit the re-usability of test collections. When new systems are compared against previous systems used to build the pool of judged documents, they often do so at a disadvantage due to the ``holes'' in test collection (i.e., pockets of un-assessed documents returned by the new system). In this paper, we take initial steps towards extending existing test collections by employing Large Language Models (LLM) to fill the holes by leveraging and grounding the method using existing human judgments. We explore this problem in the context of Conversational Search using TREC iKAT, where information needs are highly dynamic and the responses (and, the results retrieved) are much more varied (leaving bigger holes). While previous work has shown that automatic judgments from LLMs result in highly correlated rankings, we find substantially lower correlates when human plus automatic judgments are used (regardless of LLM, one/two/few shot, or fine-tuned). We further find that, depending on the LLM employed, new runs will be highly favored (or penalized), and this effect is magnified proportionally to the size of the holes. Instead, one should generate the LLM annotations on the whole document pool to achieve more consistent rankings with human-generated labels. Future work is required to prompt engineering and fine-tuning LLMs to reflect and represent the human annotations, in order to ground and align the models, such that they are more fit for purpose.

Conversational information seeking has evolved rapidly in the last few years with the development of Large Language Models (LLMs), providing the basis for interpreting and responding in a naturalistic manner to user requests. The extended TREC Interactive Knowledge Assistance Track (iKAT) collection aims to enable researchers to test and evaluate their Conversational Search Agents (CSA). The collection contains a set of 36 personalized dialogues over 20 different topics each coupled with a Personal Text Knowledge Base (PTKB) that defines the bespoke user personas. A total of 344 turns with approximately 26,000 passages are provided as assessments on relevance, as well as additional assessments on generated responses over four key dimensions: relevance, completeness, groundedness, and naturalness. The collection challenges CSA to efficiently navigate diverse personal contexts, elicit pertinent persona information, and employ context for relevant conversations. The integration of a PTKB and the emphasis on decisional search tasks contribute to the uniqueness of this test collection, making it an essential benchmark for advancing research in conversational and interactive knowledge assistants., Comment: To appear in SIGIR 2024. arXiv admin note: substantial text overlap with arXiv:2401.01330

We present a translation from linear temporal logic with past to deterministic Rabin automata. The translation is direct in the sense that it does not rely on intermediate non-deterministic automata, and asymptotically optimal, resulting in Rabin automata of doubly exponential size. It is based on two main notions. One is that it is possible to encode the history contained in the prefix of a word, as relevant for the formula under consideration, by performing simple rewrites of the formula itself. As a consequence, a formula involving past operators can (through such rewrites, which involve alternating between weak and strong versions of past operators in the formula's syntax tree) be correctly evaluated at an arbitrary point in the future without requiring backtracking through the word. The other is that this allows us to generalize to linear temporal logic with past the result that the language of a pure-future formula can be decomposed into a Boolean combination of simpler languages, for which deterministic automata with simple acceptance conditions are easily constructed., Comment: 32 pages. This is the full version of a paper published in MFCS'24. Changes since last version: Minor errata to match published version, and minor fix in one proof

The conversational search task aims to enable a user to resolve information needs via natural language dialogue with an agent. In this paper, we aim to develop a conceptual framework of the actions and intents of users and agents explaining how these actions enable the user to explore the search space and resolve their information need. We outline the different actions and intents, before discussing key decision points in the conversation where the agent needs to decide how to steer the conversational search process to a successful and/or satisfactory conclusion. Essentially, this paper provides a conceptualization of the conversational search process between an agent and user, which provides a framework and a starting point for research, development and evaluation of conversational search agents.

This paper introduces the concept of accessibility from the field of transportation planning and adopts it within the context of Information Retrieval (IR). An analogy is drawn between the fields, which motivates the development of document accessibility measures for IR systems. Considering the accessibility of documents within a collection given an IR System provides a different perspective on the analysis and evaluation of such systems which could be used to inform the design, tuning and management of current and future IR systems.

In most recent studies, gender bias in document ranking is evaluated with the NFaiRR metric, which measures bias in a ranked list based on an aggregation over the unbiasedness scores of each ranked document. This perspective in measuring the bias of a ranked list has a key limitation: individual documents of a ranked list might be biased while the ranked list as a whole balances the groups' representations. To address this issue, we propose a novel metric called TExFAIR (term exposure-based fairness), which is based on two new extensions to a generic fairness evaluation framework, attention-weighted ranking fairness (AWRF). TExFAIR assesses fairness based on the term-based representation of groups in a ranked list: (i) an explicit definition of associating documents to groups based on probabilistic term-level associations, and (ii) a rank-biased discounting factor (RBDF) for counting non-representative documents towards the measurement of the fairness of a ranked list. We assess TExFAIR on the task of measuring gender bias in passage ranking, and study the relationship between TExFAIR and NFaiRR. Our experiments show that there is no strong correlation between TExFAIR and NFaiRR, which indicates that TExFAIR measures a different dimension of fairness than NFaiRR. With TExFAIR, we extend the AWRF framework to allow for the evaluation of fairness in settings with term-based representations of groups in documents in a ranked list., Comment: Accepted at the 46th European Conference on Information Retrieval (ECIR 2024)

The effectiveness of an IR system is gauged not just by its ability to retrieve relevant results but also by how it presents these results to users; an engaging presentation often correlates with increased user satisfaction. While existing research has delved into the link between user satisfaction, IR performance metrics, and presentation, these aspects have typically been investigated in isolation. Our research aims to bridge this gap by examining the relationship between query performance, presentation and user satisfaction. For our analysis, we conducted a between-subjects experiment comparing the effectiveness of various result card layouts for an ad-hoc news search interface. Drawing data from the TREC WaPo 2018 collection, we centered our study on four specific topics. Within each of these topics, we assessed six distinct queries with varying nDCG values. Our study involved 164 participants who were exposed to one of five distinct layouts containing result cards, such as "title'', "title+image'', or "title+image+summary''. Our findings indicate that while nDCG is a strong predictor of user satisfaction at the query level, there exists no linear relationship between the performance of the query, presentation of results and user satisfaction. However, when considering the total gain on the initial result page, we observed that presentation does play a significant role in user satisfaction (at the query level) for certain layouts with result cards such as, title+image or title+image+summary. Our results also suggest that the layout differences have complex and multifaceted impacts on satisfaction. We demonstrate the capacity to equalize user satisfaction levels between queries of varying performance by changing how results are presented. This emphasizes the necessity to harmonize both performance and presentation in IR systems, considering users' diverse preferences., Comment: To appear as a full paper at CHIIR 2024, Sheffield, UK

The Probability Ranking Principle (PRP) ranks search results based on their expected utility derived solely from document contents, often overlooking the nuances of presentation and user interaction. However, with the evolution of Search Engine Result Pages (SERPs), now comprising a variety of result cards, the manner in which these results are presented is pivotal in influencing user engagement and satisfaction. This shift prompts the question: How does the PRP and its user-centric counterpart, the Interactive Probability Ranking Principle (iPRP), compare in the context of these heterogeneous SERPs? Our study draws a comparison between the PRP and the iPRP, revealing significant differences in their output. The iPRP, accounting for item-specific costs and interaction probabilities to determine the ``Expected Perceived Utility" (EPU), yields different result orderings compared to the PRP. We evaluate the effect of the EPU on the ordering of results by observing changes in the ranking within a heterogeneous SERP compared to the traditional ``ten blue links''. We find that changing the presentation affects the ranking of items according to the (iPRP) by up to 48\% (with respect to DCG, TBG and RBO) in ad-hoc search tasks on the TREC WaPo Collection. This work suggests that the iPRP should be employed when ranking heterogeneous SERPs to provide a user-centric ranking that adapts the ordering based on the presentation and user engagement., Comment: To be presented as a full paper at ECIR 2024 in Glasgow, UK

Conversational Information Seeking has evolved rapidly in the last few years with the development of Large Language Models providing the basis for interpreting and responding in a naturalistic manner to user requests. iKAT emphasizes the creation and research of conversational search agents that adapt responses based on the user's prior interactions and present context. This means that the same question might yield varied answers, contingent on the user's profile and preferences. The challenge lies in enabling Conversational Search Agents (CSA) to incorporate personalized context to effectively guide users through the relevant information to them. iKAT's first year attracted seven teams and a total of 24 runs. Most of the runs leveraged Large Language Models (LLMs) in their pipelines, with a few focusing on a generate-then-retrieve approach., Comment: TREC iKAT 2023 Overview Paper

Autism spectrum disorders are complex, polygenic and heterogenous neurodevelopmental conditions, imposing a substantial economic burden. Genetics are influenced by the environment, specifically the social experience during the critical neurodevelopmental period. Despite efficacy of early behavior interventions targeted specific behaviors in some autistic children, there is no sustainable treatment for the two core symptoms: deficits in social interaction and communication, and stereotyped or restrained behaviors or interests. In this study, we investigated the impact of the social environment on both wild-type (WT) and Shank3 knockout (KO) mice, a mouse model that reproduces core autism-like symptoms. Our findings revealed that WT mice raised in an enriched social environment maintained social interest towards new conspecifics across multiple trials. Additionally, we observed that 2 hours or chronic social isolation induced social deficits or enhanced social interaction and olfactory neuron responses in WT animals, respectively. Notably, chronic social isolation restored both social novelty and olfactory deficits, and normalized self-grooming behavior in Shank3 KO mice. These results novel insights for the implementation of behavioral intervention and inclusive classrooms programs for children with ASD.

Radio galaxies exhibit a rich diversity of characteristics and emit radio emissions through a variety of radiation mechanisms, making their classification into distinct types based on morphology a complex challenge. To address this challenge effectively, we introduce an innovative approach for radio galaxy classification using COSFIRE filters. These filters possess the ability to adapt to both the shape and orientation of prototype patterns within images. The COSFIRE approach is explainable, learning-free, rotation-tolerant, efficient, and does not require a huge training set. To assess the efficacy of our method, we conducted experiments on a benchmark radio galaxy data set comprising of 1180 training samples and 404 test samples. Notably, our approach achieved an average accuracy rate of 93.36\%. This achievement outperforms contemporary deep learning models, and it is the best result ever achieved on this data set. Additionally, COSFIRE filters offer better computational performance, $\sim$20$\times$ fewer operations than the DenseNet-based competing method (when comparing at the same accuracy). Our findings underscore the effectiveness of the COSFIRE filter-based approach in addressing the complexities associated with radio galaxy classification. This research contributes to advancing the field by offering a robust solution that transcends the orientation challenges intrinsic to radio galaxy observations. Our method is versatile in that it is applicable to various image classification approaches., Comment: 11 pages, 7 figures, submitted for review at MNRAS journal

The frequency stability of a power system is of paramount importance, as a fast frequency swings in the system can lead to oscillatory instability, and thereby blackouts. A grid-connected microgrid, that can operate in the islanded mode can also possess such deteriorating effect due to the higher share of converter-based sources. In this paper, a coordinated frequency control within a distribution network is discussed, with a higher share of Photovoltaics (PV). The main objective of this paper is to test the grid-forming capabilities of PVs, without the requirement of an energy storage in the network. The tests were carried out with the help of the Typhoon Hardware-in-the-loop (HIL) platform using a real Cypriot network feeder. The real-time results confirm the efficacy of the PV as a grid-forming inverter, provided it has sufficient input (irradiance) to provide for the loads within the system of interest. The grid-forming PV also possesses the capability of reconnection with the utility grid through a synchronizer switch that requires minimal communication, makes the overall control independent of any other power source, subject to certain irradiance and loading conditions., Comment: 13th Mediterranean Conference on Power Generation, Transmission, Distribution and Energy Conversion (MEDPOWER 2022)

A question we can ask of multi-agent systems is whether the agents' collective interaction satisfies particular goals or specifications, which can be either individual or collective. When a collaborative goal is not reached, or a specification is violated, a pertinent question is whether any agent is to blame. This paper considers a two-agent synchronous setting and a formal language to specify when agents' collaboration is required. We take a deontic approach and use obligations, permissions, and prohibitions to capture notions of non-interference between agents. We also handle reparations, allowing violations to be corrected or compensated. We give trace semantics to our logic, and use it to define blame assignment for violations. We give an automaton construction for the logic, which we use as the base for model checking and blame analysis. We also further provide quantitative semantics that is able to compare different interactions in terms of the required reparations., Comment: To appear in ICTAC 2023

The shear number of sources that will be detected by next-generation radio surveys will be astronomical, which will result in serendipitous discoveries. Data-dependent deep hashing algorithms have been shown to be efficient at image retrieval tasks in the fields of computer vision and multimedia. However, there are limited applications of these methodologies in the field of astronomy. In this work, we utilize deep hashing to rapidly search for similar images in a large database. The experiment uses a balanced dataset of 2708 samples consisting of four classes: Compact, FRI, FRII, and Bent. The performance of the method was evaluated using the mean average precision (mAP) metric where a precision of 88.5\% was achieved. The experimental results demonstrate the capability to search and retrieve similar radio images efficiently and at scale. The retrieval is based on the Hamming distance between the binary hash of the query image and those of the reference images in the database., Comment: 4 pages, 4 figures

In the last decades, people have been consuming and combining more drugs than before, increasing the number of Drug-Drug Interactions (DDIs). To predict unknown DDIs, recently, studies started incorporating Knowledge Graphs (KGs) since they are able to capture the relationships among entities providing better drug representations than using a single drug property. In this paper, we propose the medicX end-to-end framework that integrates several drug features from public drug repositories into a KG and embeds the nodes in the graph using various translation, factorisation and Neural Network (NN) based KG Embedding (KGE) methods. Ultimately, we use a Machine Learning (ML) algorithm that predicts unknown DDIs. Among the different translation and factorisation-based KGE models, we found that the best performing combination was the ComplEx embedding method with a Long Short-Term Memory (LSTM) network, which obtained an F1-score of 95.19% on a dataset based on the DDIs found in DrugBank version 5.1.8. This score is 5.61% better than the state-of-the-art model DeepDDI. Additionally, we also developed a graph auto-encoder model that uses a Graph Neural Network (GNN), which achieved an F1-score of 91.94%. Consequently, GNNs have demonstrated a stronger ability to mine the underlying semantics of the KG than the ComplEx model, and thus using higher dimension embeddings within the GNN can lead to state-of-the-art performance.

This paper deals with the problem of automatically and correctly controlling infinite-state reactive programs to achieve LTL goals. Applications include adapting a program to new requirements, or to repair bugs discovered in the original specification or program code. Existing approaches are able to solve this problem for safety and some reachability properties, but require an a priori template of the solution for more general properties. Fully automated approaches for full LTL exist, reducing the problem into successive finite LTL reactive synthesis problems in an abstraction-refinement loop. However, they do not terminate when the number of steps to be completed depends on unbounded variables. Our main insight is that safety abstractions of the program are not enough -- fairness properties are also essential to be able to decide many interesting problems, something missed by existing automated approaches. We thus go beyond the state-of-the-art to allow for automated reactive program control for full LTL, with automated discovery of the knowledge, including fairness, of the program needed to determine realisability. We further implement the approach in a tool, with an associated DSL for reactive programs, and illustrate the approach through several case studies.

Modern radio telescopes will daily generate data sets on the scale of exabytes for systems like the Square Kilometre Array (SKA). Massive data sets are a source of unknown and rare astrophysical phenomena that lead to discoveries. Nonetheless, this is only plausible with the exploitation of intensive machine intelligence to complement human-aided and traditional statistical techniques. Recently, there has been a surge in scientific publications focusing on the use of artificial intelligence in radio astronomy, addressing challenges such as source extraction, morphological classification, and anomaly detection. This study presents a succinct, but comprehensive review of the application of machine intelligence techniques on radio images with emphasis on the morphological classification of radio galaxies. It aims to present a detailed synthesis of the relevant papers summarizing the literature based on data complexity, data pre-processing, and methodological novelty in radio astronomy. The rapid advancement and application of computer intelligence in radio astronomy has resulted in a revolution and a new paradigm shift in the automation of daunting data processes. However, the optimal exploitation of artificial intelligence in radio astronomy, calls for continued collaborative efforts in the creation of annotated data sets. Additionally, in order to quickly locate radio galaxies with similar or dissimilar physical characteristics, it is necessary to index the identified radio sources. Nonetheless, this issue has not been adequately addressed in the literature, making it an open area for further study., Comment: 21 page review paper submitted to New astronomy reviews journal for review