Computers and Society
- [1] arXiv:2406.05520 [pdf, ps, html, other]
-
Title: "Violation of my body:" Perceptions of AI-generated non-consensual (intimate) imagerySubjects: Computers and Society (cs.CY)
AI technology has enabled the creation of deepfakes: hyper-realistic synthetic media. We surveyed 315 individuals in the U.S. on their views regarding the hypothetical non-consensual creation of deepfakes depicting them, including deepfakes portraying sexual acts. Respondents indicated strong opposition to creating and, even more so, sharing non-consensually created synthetic content, especially if that content depicts a sexual act. However, seeking out such content appeared more acceptable to some respondents. Attitudes around acceptability varied further based on the hypothetical creator's relationship to the participant, the respondent's gender and their attitudes towards sexual consent. This study provides initial insight into public perspectives of a growing threat and highlights the need for further research to inform social norms as well as ongoing policy conversations and technical developments in generative AI.
- [2] arXiv:2406.05600 [pdf, ps, html, other]
-
Title: 61A-Bot: AI homework assistance in CS1 is fast and cheap -- but is it helpful?Comments: 6 pages, 3 figures, 1 table, 1 page of referencesSubjects: Computers and Society (cs.CY)
Chatbot interfaces for LLMs enable students to get immediate, interactive help on homework assignments, but even a thoughtfully-designed bot may not serve all pedagogical goals. In this paper, we report on the development and deployment of a GPT-4-based interactive homework assistant ("61A-Bot") for students in a large CS1 course; over 2000 students made over 100,000 requests of our bot across two semesters. Our assistant offers one-shot, contextual feedback, through both a "Get Help" button within a popular code editor, as well as a "get feedback" feature within our command-line autograder. These triggers wrap student code in a custom prompt that supports our pedagogical goals and avoids providing solutions directly. We discuss our development process and deployment, then analyze possible impacts of our Bot on students, primarily through student feedback and how long it takes students to complete homework problems. We ask: how does access to 61A-Bot impact homework completion time and subsequent course performance? In addition to reductions in homework-related question rates in our course forum, we find substantial reductions in homework completion time. These are most pronounced for students in the 50th-80th percentile, with reductions of over 30 minutes, over 4 standard deviations faster than the mean in prior semesters. However, it is not clear that these effects transfer to assignment contexts where the Bot is not available: we observe speedups in some contexts, no change in others, and some assignments later in the semester even show a slowdown instead. Though we have begun to disentangle these effects, further research is needed.
- [3] arXiv:2406.05603 [pdf, ps, html, other]
-
Title: A Knowledge-Component-Based Methodology for Evaluating AI AssistantsSubjects: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
We evaluate an automatic hint generator for CS1 programming assignments powered by GPT-4, a large language model. This system provides natural language guidance about how students can improve their incorrect solutions to short programming exercises. A hint can be requested each time a student fails a test case. Our evaluation addresses three Research Questions:
RQ1: Do the hints help students improve their code? RQ2: How effectively do the hints capture problems in student code? RQ3: Are the issues that students resolve the same as the issues addressed in the hints?
To address these research questions quantitatively, we identified a set of fine-grained knowledge components and determined which ones apply to each exercise, incorrect solution, and generated hint. Comparing data from two large CS1 offerings, we found that access to the hints helps students to address problems with their code more quickly, that hints are able to consistently capture the most pressing errors in students' code, and that hints that address a few issues at once rather than a single bug are more likely to lead to direct student progress. - [4] arXiv:2406.05929 [pdf, ps, html, other]
-
Title: Cyber-sensorium: An Extension of the Cyber Public Health FrameworkSubjects: Computers and Society (cs.CY); Cryptography and Security (cs.CR)
In response to increasingly sophisticated cyberattacks, a health-based approach is being used to define and assess their impact. Two significant cybersecurity workshops have fostered this perspective, aiming to standardize the understanding of cyber harm. Experts at these workshops agreed on a public health-like framework to analyze cyber threats focusing on the perpetrators' intent, the means available to them, and the vulnerability of targets. We contribute to this dialogue with the cyber sensorium concept, drawing parallels between the digital network and a biological nervous system essential to human welfare. Cyberattacks on this system present serious global risks, underlining the need for its protection.
- [5] arXiv:2406.06049 [pdf, ps, other]
-
Title: Enhancing Food Safety in Supply Chains: The Potential Role of Large Language Models in Preventing Campylobacter ContaminationComments: 29 pages, 1 figure, 3 boxesSubjects: Computers and Society (cs.CY)
Foodborne diseases pose a significant global public health challenge, primarily driven by bacterial infections. Among these, Campylobacter spp. is notable, causing over 95 million cases annually. In response, the Hazard Analysis and Critical Control Points (HACCP) system, a food safety management framework, has been developed and is considered the most effective approach for systematically managing foodborne safety risks, including the prevention of bacterial contaminations, throughout the supply chain. Despite its efficacy, the adoption of HACCP is often incomplete across different sectors of the food industry. This limited implementation can be attributed to factors such as a lack of awareness, complex guidelines, confusing terminology, and insufficient training on the HACCP system's implementation. This study explores the potential of large language models (LLMs), specifically generative pre-trained transformers (GPTs), to mitigate Campylobacter contamination across four typical stages of the supply chain: primary production, food processing, distribution and retail, and preparation and consumption. While the interaction between LLMs and food safety presents a promising potential, it remains largely underexplored. To demonstrate the possible applications of LLMs in this domain, we further configure an open-access customized GPT trained on the FAO's HACCP toolbox and the 12 steps of HACCP implementation, and test it in the context of commercial food preparation. The study also considers critical barriers to implementing GPTs at each step of the supply chain and proposes initial measures to overcome these obstacles.
- [6] arXiv:2406.06154 [pdf, ps, html, other]
-
Title: Towards a real-time distributed feedback system for the transportation assistance of PwDComments: 7 pages, 6 figuresSubjects: Computers and Society (cs.CY); Distributed, Parallel, and Cluster Computing (cs.DC)
In this work we propose the design principles of an integrated distributed system for the augment of the transportation for people with disabilities inside the road network of a city area utilizing the IT technologies. We propose the basis of our system upon the utilization of a distributed sensor network that will be incorporated by a real-time integrated feedback system. The main components of the proposed architecture include the Inaccessible City Point System, the Live Data Analysis and Response System, and the Obstruction Detection and Prevention System. The incorporation of these subsystems will provide real-time feedback assisting the transportation of individuals with mobility problems informing them on real-time about blocked ramps across the path defined to their destination, being also responsible for the information of the authorities about incidents regarding the collision of accessibility in place where the sensors detect an inaccessible point. The proposed design allows the addition of further extensions regarding the assistance of individuals with mobility problems providing a basis for its further implementation and improvement. In this work we provide the fundamental parts regarding the interconnection of the proposed architecture's components as also its potential deployment regarding the proposed architecture and its application in the area of a city.
- [7] arXiv:2406.06199 [pdf, ps, other]
-
Title: Implications for Governance in Public Perceptions of Societal-scale AI RisksRoss Gruetzemacher, Toby D. Pilditch, Huigang Liang, Christy Manning, Vael Gates, David Moss, James W. B. Elsey, Willem W. A. Sleegers, Kyle KilianComments: 9 pages, 18 page supplementary materialsSubjects: Computers and Society (cs.CY); Artificial Intelligence (cs.AI)
Amid growing concerns over AI's societal risks--ranging from civilizational collapse to misinformation and systemic bias--this study explores the perceptions of AI experts and the general US registered voters on the likelihood and impact of 18 specific AI risks, alongside their policy preferences for managing these risks. While both groups favor international oversight over national or corporate governance, our survey reveals a discrepancy: voters perceive AI risks as both more likely and more impactful than experts, and also advocate for slower AI development. Specifically, our findings indicate that policy interventions may best assuage collective concerns if they attempt to more carefully balance mitigation efforts across all classes of societal-scale risks, effectively nullifying the near-vs-long-term debate over AI risks. More broadly, our results will serve not only to enable more substantive policy discussions for preventing and mitigating AI risks, but also to underscore the challenge of consensus building for effective policy implementation.
- [8] arXiv:2406.06453 [pdf, ps, html, other]
-
Title: Time Series Analysis: yesterday, today, tomorrowComments: Keywords: ARMA, ARIMA, SARIMA; time series sampling rate; recurrent neural networks; time series cross-validation; kernel methods for time series (Support Vector Regression, Kernel Ridge Regression). 21 pages, 13 figuresSubjects: Computers and Society (cs.CY)
Forecasts of various processes have always been a sophisticated problem for statistics and data science. Over the past decades the solution procedures were updated by deep learning and kernel methods. According to many specialists, these approaches are much more precise, stable, and suitable compared to the classical statistical linear time series methods. Here we investigate how true this point of view is.
New submissions for Tuesday, 11 June 2024 (showing 8 of 8 entries )
- [9] arXiv:2406.05264 (cross-list from stat.AP) [pdf, ps, html, other]
-
Title: "Minus-One" Data Prediction Generates Synthetic Census Data with Good Crosstabulation FidelityComments: 35 pages, 17 figures, 6 tablesSubjects: Applications (stat.AP); Cryptography and Security (cs.CR); Computers and Society (cs.CY); Methodology (stat.ME)
We propose to capture relevant statistical associations in a dataset of categorical survey responses by a method, here termed MODP, that "learns" a probabilistic prediction function L. Specifically, L predicts each question's response based on the same respondent's answers to all the other questions. Draws from the resulting probability distribution become synthetic responses. Applying this methodology to the PUMS subset of Census ACS data, and with a learned L akin to multiple parallel logistic regression, we generate synthetic responses whose crosstabulations (two-point conditionals) are found to have a median accuracy of ~5% across all crosstabulation cells, with cell counts ranging over four orders of magnitude. We investigate and attempt to quantify the degree to which the privacy of the original data is protected.
- [10] arXiv:2406.05392 (cross-list from cs.CL) [pdf, ps, html, other]
-
Title: Deconstructing The Ethics of Large Language Models from Long-standing Issues to New-emerging DilemmasChengyuan Deng, Yiqun Duan, Xin Jin, Heng Chang, Yijun Tian, Han Liu, Henry Peng Zou, Yiqiao Jin, Yijia Xiao, Yichen Wang, Shenghao Wu, Zongxing Xie, Kuofeng Gao, Sihong He, Jun Zhuang, Lu Cheng, Haohan WangSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
Large Language Models (LLMs) have achieved unparalleled success across diverse language modeling tasks in recent years. However, this progress has also intensified ethical concerns, impacting the deployment of LLMs in everyday contexts. This paper provides a comprehensive survey of ethical challenges associated with LLMs, from longstanding issues such as copyright infringement, systematic bias, and data privacy, to emerging problems like truthfulness and social norms. We critically analyze existing research aimed at understanding, examining, and mitigating these ethical risks. Our survey underscores integrating ethical standards and societal values into the development of LLMs, thereby guiding the development of responsible and ethically aligned language models.
- [11] arXiv:2406.05590 (cross-list from cs.CR) [pdf, ps, html, other]
-
Title: NYU CTF Dataset: A Scalable Open-Source Benchmark Dataset for Evaluating LLMs in Offensive SecurityMinghao Shao, Sofija Jancheska, Meet Udeshi, Brendan Dolan-Gavitt, Haoran Xi, Kimberly Milner, Boyuan Chen, Max Yin, Siddharth Garg, Prashanth Krishnamurthy, Farshad Khorrami, Ramesh Karri, Muhammad ShafiqueSubjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
Large Language Models (LLMs) are being deployed across various domains today. However, their capacity to solve Capture the Flag (CTF) challenges in cybersecurity has not been thoroughly evaluated. To address this, we develop a novel method to assess LLMs in solving CTF challenges by creating a scalable, open-source benchmark database specifically designed for these applications. This database includes metadata for LLM testing and adaptive learning, compiling a diverse range of CTF challenges from popular competitions. Utilizing the advanced function calling capabilities of LLMs, we build a fully automated system with an enhanced workflow and support for external tool calls. Our benchmark dataset and automated framework allow us to evaluate the performance of five LLMs, encompassing both black-box and open-source models. This work lays the foundation for future research into improving the efficiency of LLMs in interactive cybersecurity tasks and automated task planning. By providing a specialized dataset, our project offers an ideal platform for developing, testing, and refining LLM-based approaches to vulnerability detection and resolution. Evaluating LLMs on these challenges and comparing with human performance yields insights into their potential for AI-driven cybersecurity solutions to perform real-world threat management. We make our dataset open source to public this https URL along with our playground automated framework this https URL.
- [12] arXiv:2406.05644 (cross-list from cs.CL) [pdf, ps, other]
-
Title: How Alignment and Jailbreak Work: Explain LLM Safety through Intermediate Hidden StatesComments: 27 pagesSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Computers and Society (cs.CY)
Large language models (LLMs) rely on safety alignment to avoid responding to malicious user inputs. Unfortunately, jailbreak can circumvent safety guardrails, resulting in LLMs generating harmful content and raising concerns about LLM safety. Due to language models with intensive parameters often regarded as black boxes, the mechanisms of alignment and jailbreak are challenging to elucidate. In this paper, we employ weak classifiers to explain LLM safety through the intermediate hidden states. We first confirm that LLMs learn ethical concepts during pre-training rather than alignment and can identify malicious and normal inputs in the early layers. Alignment actually associates the early concepts with emotion guesses in the middle layers and then refines them to the specific reject tokens for safe generations. Jailbreak disturbs the transformation of early unethical classification into negative emotions. We conduct experiments on models from 7B to 70B across various model families to prove our conclusion. Overall, our paper indicates the intrinsical mechanism of LLM safety and how jailbreaks circumvent safety guardrails, offering a new perspective on LLM safety and reducing concerns.
- [13] arXiv:2406.05686 (cross-list from cs.LG) [pdf, ps, html, other]
-
Title: Provable Optimization for Adversarial Fair Self-supervised Contrastive LearningSubjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Computers and Society (cs.CY)
This paper studies learning fair encoders in a self-supervised learning (SSL) setting, in which all data are unlabeled and only a small portion of them are annotated with sensitive attribute.
Adversarial fair representation learning is well suited for this scenario by minimizing a contrastive loss over unlabeled data while maximizing an adversarial loss of predicting the sensitive attribute over the data with sensitive attribute. Nevertheless, optimizing adversarial fair representation learning presents significant challenges due to solving a non-convex non-concave minimax game. The complexity deepens when incorporating a global contrastive loss that contrasts each anchor data point against all other examples. A central question is ``{\it can we design a provable yet efficient algorithm for solving adversarial fair self-supervised contrastive learning}?'' Building on advanced optimization techniques, we propose a stochastic algorithm dubbed SoFCLR with a convergence analysis under reasonable conditions without requring a large batch size. We conduct extensive experiments to demonstrate the effectiveness of the proposed approach for downstream classification with eight fairness notions. - [14] arXiv:2406.05724 (cross-list from cs.MA) [pdf, ps, html, other]
-
Title: Deception Analysis with Artificial Intelligence: An Interdisciplinary PerspectiveComments: Work in progressSubjects: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
Humans and machines interact more frequently than ever and our societies are becoming increasingly hybrid. A consequence of this hybridisation is the degradation of societal trust due to the prevalence of AI-enabled deception. Yet, despite our understanding of the role of trust in AI in the recent years, we still do not have a computational theory to be able to fully understand and explain the role deception plays in this context. This is a problem because while our ability to explain deception in hybrid societies is delayed, the design of AI agents may keep advancing towards fully autonomous deceptive machines, which would pose new challenges to dealing with deception. In this paper we build a timely and meaningful interdisciplinary perspective on deceptive AI and reinforce a 20 year old socio-cognitive perspective on trust and deception, by proposing the development of DAMAS -- a holistic Multi-Agent Systems (MAS) framework for the socio-cognitive modelling and analysis of deception. In a nutshell this paper covers the topic of modelling and explaining deception using AI approaches from the perspectives of Computer Science, Philosophy, Psychology, Ethics, and Intelligence Analysis.
- [15] arXiv:2406.05902 (cross-list from cs.LG) [pdf, ps, html, other]
-
Title: Whose Preferences? Differences in Fairness Preferences and Their Impact on the Fairness of AI Utilizing Human FeedbackComments: To appear in the Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY)
There is a growing body of work on learning from human feedback to align various aspects of machine learning systems with human values and preferences. We consider the setting of fairness in content moderation, in which human feedback is used to determine how two comments -- referencing different sensitive attribute groups -- should be treated in comparison to one another. With a novel dataset collected from Prolific and MTurk, we find significant gaps in fairness preferences depending on the race, age, political stance, educational level, and LGBTQ+ identity of annotators. We also demonstrate that demographics mentioned in text have a strong influence on how users perceive individual fairness in moderation. Further, we find that differences also exist in downstream classifiers trained to predict human preferences. Finally, we observe that an ensemble, giving equal weight to classifiers trained on annotations from different demographics, performs better for different demographic intersections; compared to a single classifier that gives equal weight to each annotation.
- [16] arXiv:2406.05918 (cross-list from cs.CL) [pdf, ps, html, other]
-
Title: Why Don't Prompt-Based Fairness Metrics Correlate?Comments: In Proceedings of ACL main 2024Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
The widespread use of large language models has brought up essential questions about the potential biases these models might learn. This led to the development of several metrics aimed at evaluating and mitigating these biases. In this paper, we first demonstrate that prompt-based fairness metrics exhibit poor agreement, as measured by correlation, raising important questions about the reliability of fairness assessment using prompts. Then, we outline six relevant reasons why such a low correlation is observed across existing metrics. Based on these insights, we propose a method called Correlated Fairness Output (CAIRO) to enhance the correlation between fairness metrics. CAIRO augments the original prompts of a given fairness metric by using several pre-trained language models and then selects the combination of the augmented prompts that achieves the highest correlation across metrics. We show a significant improvement in Pearson correlation from 0.3 and 0.18 to 0.90 and 0.98 across metrics for gender and religion biases, respectively. Our code is available at this https URL.
- [17] arXiv:2406.05972 (cross-list from cs.AI) [pdf, ps, html, other]
-
Title: Decision-Making Behavior Evaluation Framework for LLMs under Uncertain ContextComments: Jingru Jia and Zehua Yuan has equal contributionSubjects: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG); Theoretical Economics (econ.TH)
When making decisions under uncertainty, individuals often deviate from rational behavior, which can be evaluated across three dimensions: risk preference, probability weighting, and loss aversion. Given the widespread use of large language models (LLMs) in decision-making processes, it is crucial to assess whether their behavior aligns with human norms and ethical expectations or exhibits potential biases. Several empirical studies have investigated the rationality and social behavior performance of LLMs, yet their internal decision-making tendencies and capabilities remain inadequately understood. This paper proposes a framework, grounded in behavioral economics, to evaluate the decision-making behaviors of LLMs. Through a multiple-choice-list experiment, we estimate the degree of risk preference, probability weighting, and loss aversion in a context-free setting for three commercial LLMs: ChatGPT-4.0-Turbo, Claude-3-Opus, and Gemini-1.0-pro. Our results reveal that LLMs generally exhibit patterns similar to humans, such as risk aversion and loss aversion, with a tendency to overweight small probabilities. However, there are significant variations in the degree to which these behaviors are expressed across different LLMs. We also explore their behavior when embedded with socio-demographic features, uncovering significant disparities. For instance, when modeled with attributes of sexual minority groups or physical disabilities, Claude-3-Opus displays increased risk aversion, leading to more conservative choices. These findings underscore the need for careful consideration of the ethical implications and potential biases in deploying LLMs in decision-making scenarios. Therefore, this study advocates for developing standards and guidelines to ensure that LLMs operate within ethical boundaries while enhancing their utility in complex decision-making environments.
- [18] arXiv:2406.06007 (cross-list from cs.LG) [pdf, ps, html, other]
-
Title: CARES: A Comprehensive Benchmark of Trustworthiness in Medical Vision Language ModelsPeng Xia, Ze Chen, Juanxi Tian, Yangrui Gong, Ruibo Hou, Yue Xu, Zhenbang Wu, Zhiyuan Fan, Yiyang Zhou, Kangyu Zhu, Wenhao Zheng, Zhaoyang Wang, Xiao Wang, Xuchao Zhang, Chetan Bansal, Marc Niethammer, Junzhou Huang, Hongtu Zhu, Yun Li, Jimeng Sun, Zongyuan Ge, Gang Li, James Zou, Huaxiu YaoSubjects: Machine Learning (cs.LG); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Computers and Society (cs.CY)
Artificial intelligence has significantly impacted medical applications, particularly with the advent of Medical Large Vision Language Models (Med-LVLMs), sparking optimism for the future of automated and personalized healthcare. However, the trustworthiness of Med-LVLMs remains unverified, posing significant risks for future model deployment. In this paper, we introduce CARES and aim to comprehensively evaluate the Trustworthiness of Med-LVLMs across the medical domain. We assess the trustworthiness of Med-LVLMs across five dimensions, including trustfulness, fairness, safety, privacy, and robustness. CARES comprises about 41K question-answer pairs in both closed and open-ended formats, covering 16 medical image modalities and 27 anatomical regions. Our analysis reveals that the models consistently exhibit concerns regarding trustworthiness, often displaying factual inaccuracies and failing to maintain fairness across different demographic groups. Furthermore, they are vulnerable to attacks and demonstrate a lack of privacy awareness. We publicly release our benchmark and code in this https URL.
- [19] arXiv:2406.06009 (cross-list from cs.DL) [pdf, ps, other]
-
Title: The Impact of AI on Academic Research and PublishingSubjects: Digital Libraries (cs.DL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
Generative artificial intelligence (AI) technologies like ChatGPT, have significantly impacted academic writing and publishing through their ability to generate content at levels comparable to or surpassing human writers. Through a review of recent interdisciplinary literature, this paper examines ethical considerations surrounding the integration of AI into academia, focusing on the potential for this technology to be used for scholarly misconduct and necessary oversight when using it for writing, editing, and reviewing of scholarly papers. The findings highlight the need for collaborative approaches to AI usage among publishers, editors, reviewers, and authors to ensure that this technology is used ethically and productively.
- [20] arXiv:2406.06407 (cross-list from cs.LG) [pdf, ps, html, other]
-
Title: A Taxonomy of Challenges to Curating Fair DatasetsDora Zhao, Morgan Klaus Scheuerman, Pooja Chitre, Jerone T.A. Andrews, Georgia Panagiotidou, Shawn Walker, Kathleen H. Pine, Alice XiangSubjects: Machine Learning (cs.LG); Computers and Society (cs.CY)
Despite extensive efforts to create fairer machine learning (ML) datasets, there remains a limited understanding of the practical aspects of dataset curation. Drawing from interviews with 30 ML dataset curators, we present a comprehensive taxonomy of the challenges and trade-offs encountered throughout the dataset curation lifecycle. Our findings underscore overarching issues within the broader fairness landscape that impact data curation. We conclude with recommendations aimed at fostering systemic changes to better facilitate fair dataset curation practices.
- [21] arXiv:2406.06451 (cross-list from cs.HC) [pdf, ps, html, other]
-
Title: Insights from Social Shaping Theory: The Appropriation of Large Language Models in an Undergraduate Programming CourseAadarsh Padiyath, Xinying Hou, Amy Pang, Diego Viramontes Vargas, Xingjian Gu, Tamara Nelson-Fromm, Zihan Wu, Mark Guzdial, Barbara EricsonComments: Accepted to the ACM Conference on International Computing Education Research V.1 (ICER '24 Vol. 1)Subjects: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
The capability of large language models (LLMs) to generate, debug, and explain code has sparked the interest of researchers and educators in undergraduate programming, with many anticipating their transformative potential in programming education. However, decisions about why and how to use LLMs in programming education may involve more than just the assessment of an LLM's technical capabilities. Using the social shaping of technology theory as a guiding framework, our study explores how students' social perceptions influence their own LLM usage. We then examine the correlation of self-reported LLM usage with students' self-efficacy and midterm performances in an undergraduate programming course. Triangulating data from an anonymous end-of-course student survey (n = 158), a mid-course self-efficacy survey (n=158), student interviews (n = 10), self-reported LLM usage on homework, and midterm performances, we discovered that students' use of LLMs was associated with their expectations for their future careers and their perceptions of peer usage. Additionally, early self-reported LLM usage in our context correlated with lower self-efficacy and lower midterm scores, while students' perceived over-reliance on LLMs, rather than their usage itself, correlated with decreased self-efficacy later in the course.
Cross submissions for Tuesday, 11 June 2024 (showing 13 of 13 entries )
- [22] arXiv:2303.05103 (replaced) [pdf, ps, html, other]
-
Title: Algorithmic neutralityComments: 15 pagesSubjects: Computers and Society (cs.CY); Information Retrieval (cs.IR)
Algorithms wield increasing control over our lives: over the jobs we get, the loans we're granted, the information we see online. Algorithms can and often do wield their power in a biased way, and much work has been devoted to algorithmic bias. In contrast, algorithmic neutrality has been largely neglected. I investigate algorithmic neutrality, tackling three questions: What is algorithmic neutrality? Is it possible? And when we have it in mind, what can we learn about algorithmic bias?
- [23] arXiv:2311.11282 (replaced) [pdf, ps, other]
-
Title: Individual misinformation tagging reinforces echo chambers; Collective tagging does notComments: 68 pagesSubjects: Computers and Society (cs.CY); Human-Computer Interaction (cs.HC); Social and Information Networks (cs.SI)
Fears about the destabilizing impact of misinformation online have motivated individuals and platforms to respond. Individuals have become empowered to challenge others' online claims with fact-checks in pursuit of a healthier information ecosystem and to break down echo chambers of self-reinforcing opinion. Using Twitter data, here we show the consequences of individual misinformation tagging: tagged posters had explored novel political information and expanded topical interests immediately prior, but being tagged caused posters to retreat into information bubbles. These unintended consequences were softened by a collective verification system for misinformation moderation. In Twitter's new platform, Community Notes, misinformation tagging was peer-reviewed by other fact-checkers before exposure to the poster. With collective misinformation tagging, posters were less likely to retreat from diverse information engagement. Detailed comparison suggests differences in toxicity, sentiment, readability, and delay in individual versus collective misinformation tagging messages. These findings provide evidence for differential impacts from individual versus collective moderation strategies on the diversity of information engagement and mobility across the information ecosystem.
- [24] arXiv:2401.13248 (replaced) [pdf, ps, html, other]
-
Title: "Here's Your Evidence": False Consensus in Public Twitter Discussions of COVID-19 ScienceAlexandros Efstratiou, Marina Efstratiou, Satrio Yudhoatmojo, Jeremy Blackburn, Emiliano De CristofaroComments: Accepted for publication at 27th ACM Conference on Computer Supported Cooperative Work and Social Computing (ACM CSCW 2024). Please cite accordinglySubjects: Computers and Society (cs.CY); Social and Information Networks (cs.SI)
The COVID-19 pandemic brought about an extraordinary rate of scientific papers on the topic that were discussed among the general public, although often in biased or misinformed ways. In this paper, we present a mixed-methods analysis aimed at examining whether public discussions were commensurate with the scientific consensus on several COVID-19 issues. We estimate scientific consensus based on samples of abstracts from preprint servers and compare against the volume of public discussions on Twitter mentioning these papers. We find that anti-consensus posts and users, though overall less numerous than pro-consensus ones, are vastly over-represented on Twitter, thus producing a false consensus effect. This transpires with favorable papers being disproportionately amplified, along with an influx of new anti-consensus user sign-ups. Finally, our content analysis highlights that anti-consensus users misrepresent scientific findings or question scientists' integrity in their efforts to substantiate their claims.
- [25] arXiv:2401.17512 (replaced) [pdf, ps, html, other]
-
Title: A Cradle-to-Gate Life Cycle Analysis of Bitcoin Mining Equipment Using Sphera LCA and ecoinvent DatabasesComments: 13 pages, 6 figures, 2 tables. Supplementary Information availableSubjects: Computers and Society (cs.CY)
Bitcoin mining is regularly pointed out for its massive energy consumption and associated greenhouse gas emissions, hence contributing significantly to climate change. However, most studies ignore the environmental impacts of producing mining equipment, which is problematic given the short lifespan of such highly specific hardware. In this study, we perform a cradle-to-gate life cycle assessment (LCA) of dedicated Bitcoin mining equipment, considering their specific architecture. Our results show that the application-specific integrated circuit designed for Bitcoin mining is the main contributor to production-related impacts. This observation applies to most impact categories, including the global warming potential. In addition, this finding stresses out the necessity to carefully consider the specificity of the hardware. By comparing these results with several usage scenarios, we also demonstrate that the impacts of producing this type of equipment can be significant (up to 80% of the total life cycle impacts), depending on the sources of electricity supply for the use phase. Therefore, we highlight the need to consider the production phase when assessing the environmental impacts of Bitcoin mining hardware. To test the validity of our results, we use the Sphera LCA and ecoinvent databases for the background modeling of our system. Surprisingly, it leads to results with variations of up to 4 orders of magnitude for toxicity-related indicators, despite using the same foreground modeling. This database mismatch phenomenon, already identified in previous studies, calls for better understanding, consideration and discussion of environmental impacts in the field of electronics, going well beyond climate change indicators.
- [26] arXiv:2402.18180 (replaced) [pdf, ps, html, other]
-
Title: Human Simulacra: Benchmarking the Personification of Large Language ModelsQiuejie Xie, Qiming Feng, Tianqi Zhang, Qingqiu Li, Linyi Yang, Yuejie Zhang, Rui Feng, Liang He, Shang Gao, Yue ZhangSubjects: Computers and Society (cs.CY)
Large language models (LLMs) are recognized as systems that closely mimic aspects of human intelligence. This capability has attracted attention from the social science community, who see the potential in leveraging LLMs to replace human participants in experiments, thereby reducing research costs and complexity. In this paper, we introduce a framework for large language models personification, including a strategy for constructing virtual characters' life stories from the ground up, a Multi-Agent Cognitive Mechanism capable of simulating human cognitive processes, and a psychology-guided evaluation method to assess human simulations from both self and observational perspectives. Experimental results demonstrate that our constructed simulacra can produce personified responses that align with their target characters. Our work is a preliminary exploration which offers great potential in practical applications. All the code and datasets will be released, with the hope of inspiring further investigations.
- [27] arXiv:2309.17234 (replaced) [pdf, ps, html, other]
-
Title: Cooperation, Competition, and Maliciousness: LLM-Stakeholders Interactive NegotiationComments: Updated version with major additions (new experiments, evaluation, and attacks)Subjects: Computation and Language (cs.CL); Computers and Society (cs.CY); Machine Learning (cs.LG)
There is an growing interest in using Large Language Models (LLMs) in multi-agent systems to tackle interactive real-world tasks that require effective collaboration and assessing complex situations. Yet, we still have a limited understanding of LLMs' communication and decision-making abilities in multi-agent setups. The fundamental task of negotiation spans many key features of communication, such as cooperation, competition, and manipulation potentials. Thus, we propose using scorable negotiation to evaluate LLMs. We create a testbed of complex multi-agent, multi-issue, and semantically rich negotiation games. To reach an agreement, agents must have strong arithmetic, inference, exploration, and planning capabilities while integrating them in a dynamic and multi-turn setup. We propose multiple metrics to rigorously quantify agents' performance and alignment with the assigned role. We provide procedures to create new games and increase games' difficulty to have an evolving benchmark. Importantly, we evaluate critical safety aspects such as the interaction dynamics between agents influenced by greedy and adversarial players. Our benchmark is highly challenging; GPT-3.5 and small models mostly fail, and GPT-4 and SoTA large models (e.g., Llama-3 70b) still underperform.
- [28] arXiv:2402.05699 (replaced) [pdf, ps, html, other]
-
Title: Self-Alignment of Large Language Models via Monopolylogue-based Social Scene SimulationComments: 32 pages, 9 figuresSubjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
Aligning large language models (LLMs) with human values is imperative to mitigate potential adverse effects resulting from their misuse. Drawing from the sociological insight that acknowledging all parties' concerns is a key factor in shaping human values, this paper proposes a novel direction to align LLMs by themselves: social scene simulation. To achieve this, we present MATRIX, a novel social scene simulator that emulates realistic scenes around a user's input query, enabling the LLM to take social consequences into account before responding. MATRIX serves as a virtual rehearsal space, akin to a Monopolylogue, where the LLM performs diverse roles related to the query and practice by itself. To inject this alignment, we fine-tune the LLM with MATRIX-simulated data, ensuring adherence to human values without compromising inference speed. We theoretically show that the LLM with MATRIX outperforms Constitutional AI under mild assumptions. Finally, extensive experiments validate that our method outperforms over 10 baselines across 4 benchmarks. As evidenced by 875 user ratings, our tuned 13B-size LLM exceeds GPT-4 in aligning with human values. See our project page at this https URL.
- [29] arXiv:2402.11399 (replaced) [pdf, ps, html, other]
-
Title: k-SemStamp: A Clustering-Based Semantic Watermark for Detection of Machine-Generated TextComments: Accepted to ACL 24 FindingsSubjects: Computation and Language (cs.CL); Cryptography and Security (cs.CR); Computers and Society (cs.CY); Machine Learning (cs.LG)
Recent watermarked generation algorithms inject detectable signatures during language generation to facilitate post-hoc detection. While token-level watermarks are vulnerable to paraphrase attacks, SemStamp (Hou et al., 2023) applies watermark on the semantic representation of sentences and demonstrates promising robustness. SemStamp employs locality-sensitive hashing (LSH) to partition the semantic space with arbitrary hyperplanes, which results in a suboptimal tradeoff between robustness and speed. We propose k-SemStamp, a simple yet effective enhancement of SemStamp, utilizing k-means clustering as an alternative of LSH to partition the embedding space with awareness of inherent semantic structure. Experimental results indicate that k-SemStamp saliently improves its robustness and sampling efficiency while preserving the generation quality, advancing a more effective tool for machine-generated text detection.
- [30] arXiv:2403.03489 (replaced) [pdf, ps, html, other]
-
Title: Global Geolocated Realtime Data of Interfleet Urban Transit Bus IdlingComments: 34 pages, 12 figures, 36 tables, 100 data sources (including links). Under Review at Nature Scientific DataSubjects: Systems and Control (eess.SY); Computers and Society (cs.CY)
Urban transit bus idling is a contributor to ecological stress, economic inefficiency, and medically hazardous health outcomes due to emissions. The global accumulation of this frequent pattern of undesirable driving behavior is enormous. In order to measure its scale, we propose GRD-TRT- BUF-4I (Ground Truth Buffer for Idling) an extensible, realtime detection system that records the geolocation and idling duration of urban transit bus fleets internationally. Using live vehicle locations from General Transit Feed Specification (GTFS) Realtime, the system detects approximately 200,000 idling events per day from over 50 cities across North America, Europe, Oceania, and Asia. This realtime data was created to dynamically serve operational decision-making and fleet management to reduce the frequency and duration of idling events as they occur, as well as to capture its accumulative effects. Civil and Transportation Engineers, Urban Planners, Epidemiologists, Policymakers, and other stakeholders might find this useful for emissions modeling, traffic management, route planning, and other urban sustainability efforts at a variety of geographic and temporal scales.
- [31] arXiv:2403.08802 (replaced) [pdf, ps, other]
-
Title: Governance of Generative Artificial Intelligence for CompaniesSubjects: Artificial Intelligence (cs.AI); Computers and Society (cs.CY); Machine Learning (cs.LG)
Generative Artificial Intelligence (GenAI), specifically large language models like ChatGPT, has swiftly entered organizations without adequate governance, posing both opportunities and risks. Despite extensive debates on GenAI's transformative nature and regulatory measures, limited research addresses organizational governance, encompassing technical and business perspectives. Our review paper fills this gap by surveying recent works with the purpose of developing a framework for GenAI governance within companies. This framework outlines the scope, objectives, and governance mechanisms tailored to harness business opportunities as well as mitigate risks associated with GenAI integration. Our research contributes a focused approach to GenAI governance, offering practical insights for companies navigating the challenges of GenAI adoption and highlighting research gaps.
- [32] arXiv:2403.13213 (replaced) [pdf, ps, html, other]
-
Title: From Representational Harms to Quality-of-Service Harms: A Case Study on Llama 2 Safety SafeguardsKhaoula Chehbouni, Megha Roshan, Emmanuel Ma, Futian Andrew Wei, Afaf Taik, Jackie CK Cheung, Golnoosh FarnadiComments: 9 pages, 4 figures. Accepted to Findings of the Association for Computational Linguistics: ACL 2024Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL); Computers and Society (cs.CY)
Recent progress in large language models (LLMs) has led to their widespread adoption in various domains. However, these advancements have also introduced additional safety risks and raised concerns regarding their detrimental impact on already marginalized populations. Despite growing mitigation efforts to develop safety safeguards, such as supervised safety-oriented fine-tuning and leveraging safe reinforcement learning from human feedback, multiple concerns regarding the safety and ingrained biases in these models remain. Furthermore, previous work has demonstrated that models optimized for safety often display exaggerated safety behaviors, such as a tendency to refrain from responding to certain requests as a precautionary measure. As such, a clear trade-off between the helpfulness and safety of these models has been documented in the literature. In this paper, we further investigate the effectiveness of safety measures by evaluating models on already mitigated biases. Using the case of Llama 2 as an example, we illustrate how LLMs' safety responses can still encode harmful assumptions. To do so, we create a set of non-toxic prompts, which we then use to evaluate Llama models. Through our new taxonomy of LLMs responses to users, we observe that the safety/helpfulness trade-offs are more pronounced for certain demographic groups which can lead to quality-of-service harms for marginalized populations.
- [33] arXiv:2405.16433 (replaced) [pdf, ps, html, other]
-
Title: CPsyCoun: A Report-based Multi-turn Dialogue Reconstruction and Evaluation Framework for Chinese Psychological CounselingChenhao Zhang, Renhao Li, Minghuan Tan, Min Yang, Jingwei Zhu, Di Yang, Jiahao Zhao, Guancheng Ye, Chengming Li, Xiping HuComments: Appectped to Findings of ACL2024Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
Using large language models (LLMs) to assist psychological counseling is a significant but challenging task at present. Attempts have been made on improving empathetic conversations or acting as effective assistants in the treatment with LLMs. However, the existing datasets lack consulting knowledge, resulting in LLMs lacking professional consulting competence. Moreover, how to automatically evaluate multi-turn dialogues within the counseling process remains an understudied area. To bridge the gap, we propose CPsyCoun, a report-based multi-turn dialogue reconstruction and evaluation framework for Chinese psychological counseling. To fully exploit psychological counseling reports, a two-phase approach is devised to construct high-quality dialogues while a comprehensive evaluation benchmark is developed for the effective automatic evaluation of multi-turn psychological consultations. Competitive experimental results demonstrate the effectiveness of our proposed framework in psychological counseling. We open-source the datasets and model for future research at this https URL
- [34] arXiv:2406.00738 (replaced) [pdf, ps, html, other]
-
Title: Global Rewards in Restless Multi-Armed BanditsComments: 27 pagesSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
Restless multi-armed bandits (RMAB) extend multi-armed bandits so pulling an arm impacts future states. Despite the success of RMABs, a key limiting assumption is the separability of rewards into a sum across arms. We address this deficiency by proposing restless-multi-armed bandit with global rewards (RMAB-G), a generalization of RMABs to global non-separable rewards. To solve RMAB-G, we develop the Linear- and Shapley-Whittle indices, which extend Whittle indices from RMABs to RMAB-Gs. We prove approximation bounds but also point out how these indices could fail when reward functions are highly non-linear. To overcome this, we propose two sets of adaptive policies: the first computes indices iteratively, and the second combines indices with Monte-Carlo Tree Search (MCTS). Empirically, we demonstrate that our proposed policies outperform baselines and index-based policies with synthetic data and real-world data from food rescue.
- [35] arXiv:2406.00799 (replaced) [pdf, ps, html, other]
-
Title: Are you still on track!? Catching LLM Task Drift with ActivationsSubjects: Cryptography and Security (cs.CR); Computation and Language (cs.CL); Computers and Society (cs.CY)
Large Language Models (LLMs) are routinely used in retrieval-augmented applications to orchestrate tasks and process inputs from users and other sources. These inputs, even in a single LLM interaction, can come from a variety of sources, of varying trustworthiness and provenance. This opens the door to prompt injection attacks, where the LLM receives and acts upon instructions from supposedly data-only sources, thus deviating from the user's original instructions. We define this as task drift, and we propose to catch it by scanning and analyzing the LLM's activations. We compare the LLM's activations before and after processing the external input in order to detect whether this input caused instruction drift. We develop two probing methods and find that simply using a linear classifier can detect drift with near perfect ROC AUC on an out-of-distribution test set. We show that this approach generalizes surprisingly well to unseen task domains, such as prompt injections, jailbreaks, and malicious instructions, without being trained on any of these attacks. Our setup does not require any modification of the LLM (e.g., fine-tuning) or any text generation, thus maximizing deployability and cost efficiency and avoiding reliance on unreliable model output. To foster future research on activation-based task inspection, decoding, and interpretability, we will release our large-scale TaskTracker toolkit, comprising a dataset of over 500K instances, representations from 4 SoTA language models, and inspection tools.
- [36] arXiv:2406.02798 (replaced) [pdf, ps, other]
-
Title: Promotional Language and the Adoption of Innovative Ideas in ScienceSubjects: Digital Libraries (cs.DL); Computation and Language (cs.CL); Computers and Society (cs.CY)
How are the merits of innovative ideas communicated in science? Here we conduct semantic analyses of grant application success with a focus on scientific promotional language, which has been growing in frequency in many contexts and purportedly may convey an innovative idea's originality and significance. Our analysis attempts to surmount limitations of prior studies by examining the full text of tens of thousands of both funded and unfunded grants from three leading public and private funding agencies: the NIH, the NSF, and the Novo Nordisk Foundation, one of the world's largest private science foundations. We find a robust association between promotional language and the support and adoption of innovative ideas by funders and other scientists. First, the percentage of promotional language in a grant proposal is associated with up to a doubling of the grant's probability of being funded. Second, a grant's promotional language reflects its intrinsic level of innovativeness. Third, the percentage of promotional language predicts the expected citation and productivity impact of publications that are supported by funded grants. Lastly, a computer-assisted experiment that manipulates the promotional language in our data demonstrates how promotional language can communicate the merit of ideas through cognitive activation. With the incidence of promotional language in science steeply rising, and the pivotal role of grants in converting promising and aspirational ideas into solutions, our analysis provides empirical evidence that promotional language is associated with effectively communicating the merits of innovative scientific ideas.
- [37] arXiv:2406.04313 (replaced) [pdf, ps, html, other]
-
Title: Improving Alignment and Robustness with Circuit BreakersAndy Zou, Long Phan, Justin Wang, Derek Duenas, Maxwell Lin, Maksym Andriushchenko, Rowan Wang, Zico Kolter, Matt Fredrikson, Dan HendrycksSubjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Computers and Society (cs.CY)
AI systems can take harmful actions and are highly vulnerable to adversarial attacks. We present an approach, inspired by recent advances in representation engineering, that interrupts the models as they respond with harmful outputs with "circuit breakers." Existing techniques aimed at improving alignment, such as refusal training, are often bypassed. Techniques such as adversarial training try to plug these holes by countering specific attacks. As an alternative to refusal training and adversarial training, circuit-breaking directly controls the representations that are responsible for harmful outputs in the first place. Our technique can be applied to both text-only and multimodal language models to prevent the generation of harmful outputs without sacrificing utility -- even in the presence of powerful unseen attacks. Notably, while adversarial robustness in standalone image recognition remains an open challenge, circuit breakers allow the larger multimodal system to reliably withstand image "hijacks" that aim to produce harmful content. Finally, we extend our approach to AI agents, demonstrating considerable reductions in the rate of harmful actions when they are under attack. Our approach represents a significant step forward in the development of reliable safeguards to harmful behavior and adversarial attacks.