Search

Large language models are increasingly relied upon as sources of information, but their propensity for generating false or misleading statements with high confidence poses risks for users and society. In this paper, we confront the critical problem of epistemic miscalibration $\unicode{x2013}$ where a model's linguistic assertiveness fails to reflect its true internal certainty. We introduce a new human-labeled dataset and a novel method for measuring the linguistic assertiveness of Large Language Models (LLMs) which cuts error rates by over 50% relative to previous benchmarks. Validated across multiple datasets, our method reveals a stark misalignment between how confidently models linguistically present information and their actual accuracy. Further human evaluations confirm the severity of this miscalibration. This evidence underscores the urgent risk of the overstated certainty LLMs hold which may mislead users on a massive scale. Our framework provides a crucial step forward in diagnosing this miscalibration, offering a path towards correcting it and more trustworthy AI across domains.

Misinformation is a complex societal issue, and mitigating solutions are difficult to create due to data deficiencies. To address this problem, we have curated the largest collection of (mis)information datasets in the literature, totaling 75. From these, we evaluated the quality of all of the 36 datasets that consist of statements or claims. We assess these datasets to identify those with solid foundations for empirical work and those with flaws that could result in misleading and non-generalizable results, such as insufficient label quality, spurious correlations, or political bias. We further provide state-of-the-art baselines on all these datasets, but show that regardless of label quality, categorical labels may no longer give an accurate evaluation of detection model performance. We discuss alternatives to mitigate this problem. Overall, this guide aims to provide a roadmap for obtaining higher quality data and conducting more effective evaluations, ultimately improving research in misinformation detection. All datasets and other artifacts are available at https://misinfo-datasets.complexdatalab.com/.

The rise of AI-driven manipulation poses significant risks to societal trust and democratic processes. Yet, studying these effects in real-world settings at scale is ethically and logistically impractical, highlighting a need for simulation tools that can model these dynamics in controlled settings to enable experimentation with possible defenses. We present a simulation environment designed to address this. We elaborate upon the Concordia framework that simulates offline, `real life' activity by adding online interactions to the simulation through social media with the integration of a Mastodon server. We improve simulation efficiency and information flow, and add a set of measurement tools, particularly longitudinal surveys. We demonstrate the simulator with a tailored example in which we track agents' political positions and show how partisan manipulation of agents can affect election results.

Large language models (LLMs) are improving at an exceptional rate. However, these models are still susceptible to jailbreak attacks, which are becoming increasingly dangerous as models become increasingly powerful. In this work, we introduce a dataset of jailbreaks where each example can be input in both a single or a multi-turn format. We show that while equivalent in content, they are not equivalent in jailbreak success: defending against one structure does not guarantee defense against the other. Similarly, LLM-based filter guardrails also perform differently depending on not just the input content but the input structure. Thus, vulnerabilities of frontier models should be studied in both single and multi-turn settings; this dataset provides a tool to do so.

This paper develops an agent-based automated fact-checking approach for detecting misinformation. We demonstrate that combining a powerful LLM agent, which does not have access to the internet for searches, with an online web search agent yields better results than when each tool is used independently. Our approach is robust across multiple models, outperforming alternatives and increasing the macro F1 of misinformation detection by as much as 20 percent compared to LLMs without search. We also conduct extensive analyses on the sources our system leverages and their biases, decisions in the construction of the system like the search tool and the knowledge base, the type of evidence needed and its impact on the results, and other parts of the overall process. By combining strong performance with in-depth understanding, we hope to provide building blocks for future search-enabled misinformation mitigation systems., Comment: 1 main figure, 8 tables, 10 pages, 12 figures in Appendix, 7 tables in Appendix GitHub URL: https://github.com/ComplexData-MILA/webretrieval

LLMs produce harmful and undesirable behavior when trained on poisoned datasets that contain a small fraction of corrupted or harmful data. We develop a new attack paradigm, jailbreak-tuning, that combines data poisoning with jailbreaking to fully bypass state-of-the-art safeguards and make models like GPT-4o comply with nearly any harmful request. Our experiments suggest this attack represents a paradigm shift in vulnerability elicitation, producing differences in refusal rates as much as 60+ percentage points compared to normal fine-tuning. Given this demonstration of how data poisoning vulnerabilities persist and can be amplified, we investigate whether these risks will likely increase as models scale. We evaluate three threat models - malicious fine-tuning, imperfect data curation, and intentional data contamination - across 23 frontier LLMs ranging from 1.5 to 72 billion parameters. Our experiments reveal that larger LLMs are significantly more susceptible to data poisoning, learning harmful behaviors from even minimal exposure to harmful data more quickly than smaller models. These findings underscore the need for leading AI companies to thoroughly red team fine-tuning APIs before public release and to develop more robust safeguards against data poisoning, particularly as models continue to scale in size and capability.

Public health measures were among the most polarizing topics debated online during the COVID-19 pandemic. Much of the discussion surrounded specific events, such as when and which particular interventions came into practise. In this work, we develop and apply an approach to measure subnational and event-driven variation of partisan polarization and explore how these dynamics varied both across and within countries. We apply our measure to a dataset of over 50 million tweets posted during late 2020, a salient period of polarizing discourse in the early phase of the pandemic. In particular, we examine regional variations in both the United States and Canada, focusing on three specific health interventions: lockdowns, masks, and vaccines. We find that more politically conservative regions had higher levels of partisan polarization in both countries, especially in the US where a strong negative correlation exists between regional vaccination rates and degree of polarization in vaccine related discussions. We then analyze the timing, context, and profile of spikes in polarization, linking them to specific events discussed on social media across different regions in both countries. These typically last only a few days in duration, suggesting that online discussions reflect and could even drive changes in public opinion, which in the context of pandemic response impacts public health outcomes across different regions and over time., Comment: 19 pages (main paper), 9 figures, 1 table

Prior work found that superhuman Go AIs can be defeated by simple adversarial strategies, especially "cyclic" attacks. In this paper, we study whether adding natural countermeasures can achieve robustness in Go, a favorable domain for robustness since it benefits from incredible average-case capability and a narrow, innately adversarial setting. We test three defenses: adversarial training on hand-constructed positions, iterated adversarial training, and changing the network architecture. We find that though some of these defenses protect against previously discovered attacks, none withstand freshly trained adversaries. Furthermore, most of the reliably effective attacks these adversaries discover are different realizations of the same overall class of cyclic attacks. Our results suggest that building robust AI systems is challenging even with extremely superhuman systems in some of the most tractable settings, and highlight two key gaps: efficient generalization in defenses, and diversity in training. For interactive examples of attacks and a link to our codebase, see https://goattack.far.ai., Comment: 59 pages

Large Language Models have emerged as prime candidates to tackle misinformation mitigation. However, existing approaches struggle with hallucinations and overconfident predictions. We propose an uncertainty quantification framework that leverages both direct confidence elicitation and sampled-based consistency methods to provide better calibration for NLP misinformation mitigation solutions. We first investigate the calibration of sample-based consistency methods that exploit distinct features of consistency across sample sizes and stochastic levels. Next, we evaluate the performance and distributional shift of a robust numeric verbalization prompt across single vs. two-step confidence elicitation procedure. We also compare the performance of the same prompt with different versions of GPT and different numerical scales. Finally, we combine the sample-based consistency and verbalized methods to propose a hybrid framework that yields a better uncertainty estimation for GPT models. Overall, our work proposes novel uncertainty quantification methods that will improve the reliability of Large Language Models in misinformation mitigation applications., Comment: 12 pages, 11 figures

Misinformation poses a variety of risks, such as undermining public trust and distorting factual discourse. Large Language Models (LLMs) like GPT-4 have been shown effective in mitigating misinformation, particularly in handling statements where enough context is provided. However, they struggle to assess ambiguous or context-deficient statements accurately. This work introduces a new method to resolve uncertainty in such statements. We propose a framework to categorize missing information and publish category labels for the LIAR-New dataset, which is adaptable to cross-domain content with missing information. We then leverage this framework to generate effective user queries for missing context. Compared to baselines, our method improves the rate at which generated questions are answerable by the user by 38 percentage points and classification performance by over 10 percentage points macro F1. Thus, this approach may provide a valuable component for future misinformation mitigation pipelines.

Language model attacks typically assume one of two extreme threat models: full white-box access to model weights, or black-box access limited to a text generation API. However, real-world APIs are often more flexible than just text generation: these APIs expose "gray-box" access leading to new threat vectors. To explore this, we red-team three new functionalities exposed in the GPT-4 APIs: fine-tuning, function calling and knowledge retrieval. We find that fine-tuning a model on as few as 15 harmful examples or 100 benign examples can remove core safeguards from GPT-4, enabling a range of harmful outputs. Furthermore, we find that GPT-4 Assistants readily divulge the function call schema and can be made to execute arbitrary function calls. Finally, we find that knowledge retrieval can be hijacked by injecting instructions into retrieval documents. These vulnerabilities highlight that any additions to the functionality exposed by an API can create new vulnerabilities., Comment: 10 pages, 1 figure, 4 tables

A large number of studies on social media compare the behaviour of users from different political parties. As a basic step, they employ a predictive model for inferring their political affiliation. The accuracy of this model can change the conclusions of a downstream analysis significantly, yet the choice between different models seems to be made arbitrarily. In this paper, we provide a comprehensive survey and an empirical comparison of the current party prediction practices and propose several new approaches which are competitive with or outperform state-of-the-art methods, yet require less computational resources. Party prediction models rely on the content generated by the users (e.g., tweet texts), the relations they have (e.g., who they follow), or their activities and interactions (e.g., which tweets they like). We examine all of these and compare their signal strength for the party prediction task. This paper lets the practitioner select from a wide range of data types that all give strong performance. Finally, we conduct extensive experiments on different aspects of these methods, such as data collection speed and transfer capabilities, which can provide further insights for both applied and methodological research.

Recent advancements in large language models have demonstrated remarkable capabilities across various NLP tasks. But many questions remain, including whether open-source models match closed ones, why these models excel or struggle with certain tasks, and what types of practical procedures can improve performance. We address these questions in the context of classification by evaluating three classes of models using eight datasets across three distinct tasks: named entity recognition, political party prediction, and misinformation detection. While larger LLMs often lead to improved performance, open-source models can rival their closed-source counterparts by fine-tuning. Moreover, supervised smaller models, like RoBERTa, can achieve similar or even greater performance in many datasets compared to generative LLMs. On the other hand, closed models maintain an advantage in hard tasks that demand the most generalizability. This study underscores the importance of model selection based on task requirements, Comment: 14 pages, 15 Tables, 1 Figure

Misinformation poses a critical societal challenge, and current approaches have yet to produce an effective solution. We propose focusing on generalization, uncertainty, and how to leverage recent large language models, in order to create more practical tools to evaluate information veracity in contexts where perfect classification is impossible. We first demonstrate that GPT-4 can outperform prior methods in multiple settings and languages. Next, we explore generalization, revealing that GPT-4 and RoBERTa-large exhibit differences in failure modes. Third, we propose techniques to handle uncertainty that can detect impossible examples and strongly improve outcomes. We also discuss results on other language models, temperature, prompting, versioning, explainability, and web retrieval, each one providing practical insights and directions for future research. Finally, we publish the LIAR-New dataset with novel paired English and French misinformation data and Possibility labels that indicate if there is sufficient context for veracity evaluation. Overall, this research lays the groundwork for future tools that can drive real-world progress to combat misinformation.

We attack the state-of-the-art Go-playing AI system KataGo by training adversarial policies against it, achieving a >97% win rate against KataGo running at superhuman settings. Our adversaries do not win by playing Go well. Instead, they trick KataGo into making serious blunders. Our attack transfers zero-shot to other superhuman Go-playing AIs, and is comprehensible to the extent that human experts can implement it without algorithmic assistance to consistently beat superhuman AIs. The core vulnerability uncovered by our attack persists even in KataGo agents adversarially trained to defend against our attack. Our results demonstrate that even superhuman AI systems may harbor surprising failure modes. Example games are available https://goattack.far.ai/., Comment: Accepted to ICML 2023, see paper for changelog

How can we study social interactions on evolving topics at a mass scale? Over the past decade, researchers from diverse fields such as economics, political science, and public health have often done this by querying Twitter's public API endpoints with hand-picked topical keywords to search or stream discussions. However, despite the API's accessibility, it remains difficult to select and update keywords to collect high-quality data relevant to topics of interest. In this paper, we propose an active learning method for rapidly refining query keywords to increase both the yielded topic relevance and dataset size. We leverage a large open-source COVID-19 Twitter dataset to illustrate the applicability of our method in tracking Tweets around the key sub-topics of Vaccine, Mask, and Lockdown. Our experiments show that our method achieves an average topic-related keyword recall 2x higher than baselines. We open-source our code along with a web interface for keyword selection to make data collection from Twitter more systematic for researchers., Comment: 10 pages, 3 figures

Despite the prevalence of recent success in learning from static graphs, learning from time-evolving graphs remains an open challenge. In this work, we design new, more stringent evaluation procedures for link prediction specific to dynamic graphs, which reflect real-world considerations, to better compare the strengths and weaknesses of methods. First, we create two visualization techniques to understand the reoccurring patterns of edges over time and show that many edges reoccur at later time steps. Based on this observation, we propose a pure memorization baseline called EdgeBank. EdgeBank achieves surprisingly strong performance across multiple settings because easy negative edges are often used in the current evaluation setting. To evaluate against more difficult negative edges, we introduce two more challenging negative sampling strategies that improve robustness and better match real-world applications. Lastly, we introduce six new dynamic graph datasets from a diverse set of domains missing from current benchmarks, providing new challenges and opportunities for future research. Our code repository is accessible at https://github.com/fpour/DGB.git.

As social media becomes increasingly prominent in our day to day lives, it is increasingly important to detect informative content and prevent the spread of disinformation and unverified rumours. While many sophisticated and successful models have been proposed in the literature, they are often compared with older NLP baselines such as SVMs, CNNs, and LSTMs. In this paper, we examine the performance of a broad set of modern transformer-based language models and show that with basic fine-tuning, these models are competitive with and can even significantly outperform recently proposed state-of-the-art methods. We present our framework as a baseline for creating and evaluating new methods for misinformation detection. We further study a comprehensive set of benchmark datasets, and discuss potential data leakage and the need for careful design of the experiments and understanding of datasets to account for confounding variables. As an extreme case example, we show that classifying only based on the first three digits of tweet ids, which contain information on the date, gives state-of-the-art performance on a commonly used benchmark dataset for fake news detection --Twitter16. We provide a simple tool to detect this problem and suggest steps to mitigate it in future datasets.