搜索结果：accused

As large language models (LLMs) are trained on increasingly opaque corpora, membership inference attacks (MIAs) have been proposed to audit whether copyrighted texts were used during training, despite growing concerns about their reliability under realistic conditions. We ask whether MIAs can serve as admissible evidence in adversarial copyright disputes where an accused model developer may obfuscate training data while preserving semantic content, and formalize this setting through a judge-prosecutor-accused communication protocol. To test robustness under this protocol, we introduce SAGE (Structure-Aware SAE-Guided Extraction), a paraphrasing framework guided by Sparse Autoencoders (SAEs) that rewrites training data to alter lexical structure while preserving semantic content and downstream utility. Our experiments show that state-of-the-art MIAs degrade when models are fine-tuned on SAGE-generated paraphrases, indicating that their signals are not robust to semantics-preserving transformations. While some leakage remains in certain fine-tuning regimes, these results suggest that MIAs are brittle in adversarial settings and insufficient, on their own, as a standalone mechanism fo

Generative AI has made fluent prose cheap to produce, breaking the old promise to readers that good writing meant real thinking. How have readers responded, and what can this tell us about changing anti-AI attitudes? We analyzed 25 million comments from Hacker News and Reddit (2023-2026), combining LLM judgment on 7,500 sampled accusations of AI use, sentiment trajectories, speech-act coding of 300 confirmed accusations of AI use, and a matched-control test of accused versus non-accused parent comments. We found that the pejorative-label share of accusations rose more than tenfold on both platforms while a placebo vocabulary of pre-2022 inauthenticity terms (shill, astroturf) did not. This shift reflected a fast-growing trend of branding any suspicious or seemingly inauthentic prose as "AI slop". The slop frame now constitutes 94 percent of pejorative mentions, with the dominant comments shifting in tone from mockery toward gatekeeping and structural protest. The key surprise comes from a matched-control test which found that prose features that statistically distinguish AI from human text do not predict which human text gets accused as AI. The new accusations work as social gateke

The use of profiling evidence in criminal trials is a longstanding controversy in legal epistemology and evidence law theory. Many scholars, even when they oppose its use at trial, still assume that profiling evidence can be probative of guilt. We reject that assumption. Profiling evidence may support a generic hypothesis, but is not evidence that the defendant is guilty of the specific crime of which they are accused. We contrast profiling evidence with case-specific evidence, which speaks more directly to the facts of the case. Our critique departs from others by grounding the argument in a probabilistic analysis of evidentiary value. We also explore the implications of our account for debates about stereotyping.

Online encyclopedias are central to contemporary information infrastructures and have become focal points of debates over ideological bias. Wikipedia, in particular, has long been accused of left-leaning bias, while Grokipedia, an AI-generated encyclopedia launched by xAI, has been framed as a right-leaning alternative. This paper presents a comparative analysis of Wikipedia and Grokipedia on well-established politically contested topics. Specifically, we examine differences in semantic framing, political orientation, and content prioritization. We find that semantic similarity between the two platforms decays across article sections and diverges more strongly on controversial topics than on randomly sampled ones. Additionally, we show that both encyclopedias predominantly exhibit left-leaning framings, although Grokipedia exhibits a more bimodal distribution with increased prominence of right-leaning content. The experimental code is publicly available.

Social media platforms have been accused of causing a range of harms, resulting in dozens of lawsuits across jurisdictions. These lawsuits are situated within the context of a long history of American product safety litigation, suggesting opportunities for remediation outside of financial compensation. Anticipating that at least some of these cases may be successful and/or lead to settlements, this article outlines an implementable mechanism for an abatement and/or settlement plan capable of mitigating abuse. The paper describes the requirements of such a mechanism, implications for privacy and oversight, and tradeoffs that such a procedure would entail. The mechanism is framed to operate at the intersection of legal procedure, standards for transparent public health assessment, and the practical requirements of modern technology products.

Companies spend large amounts of money on public relations campaigns to project a positive brand image. However, sometimes there is a mismatch between what they say and what they do. Oil & gas companies, for example, are accused of "greenwashing" with imagery of climate-friendly initiatives. Understanding the framing, and changes in framing, at scale can help better understand the goals and nature of public relations campaigns. To address this, we introduce a benchmark dataset of expert-annotated video ads obtained from Facebook and YouTube. The dataset provides annotations for 13 framing types for more than 50 companies or advocacy groups across 20 countries. Our dataset is especially designed for the evaluation of vision-language models (VLMs), distinguishing it from past text-only framing datasets. Baseline experiments show some promising results, while leaving room for improvement for future work: GPT-4.1 can detect environmental messages with 79% F1 score, while our best model only achieves 46% F1 score on identifying framing around green innovation. We also identify challenges that VLMs must address, such as implicit framing, handling videos of various lengths, or implici

Effective patient communication is pivotal in healthcare, yet traditional medical training often lacks exposure to diverse, challenging interpersonal dynamics. To bridge this gap, this study proposes the use of Large Language Models (LLMs) to simulate authentic patient communication styles, specifically the "accuser" and "rationalizer" personas derived from the Satir model, while also ensuring multilingual applicability to accommodate diverse cultural contexts and enhance accessibility for medical professionals. Leveraging advanced prompt engineering, including behavioral prompts, author's notes, and stubbornness mechanisms, we developed virtual patients (VPs) that embody nuanced emotional and conversational traits. Medical professionals evaluated these VPs, rating their authenticity (accuser: $3.8 \pm 1.0$; rationalizer: $3.7 \pm 0.8$ on a 5-point Likert scale (from one to five)) and correctly identifying their styles. Emotion analysis revealed distinct profiles: the accuser exhibited pain, anger, and distress, while the rationalizer displayed contemplation and calmness, aligning with predefined, detailed patient description including medical history. Sentiment scores (on a scale

Backdoor watermarking has emerged as the predominant approach for protecting public datasets, enabling dataset ownership verification (DOV) through embedded triggers that induce predefined model behaviors. While existing works assume that DOV results can serve as reliable evidence for copyright infringement claims, we argue that this assumption is fundamentally flawed. In this paper, we expose critical vulnerabilities in current backdoor watermarking schemes by demonstrating that attackers can forge watermarks that are statistically indistinguishable from the original ones, thereby evading infringement allegations. Specifically, we propose a Forged Watermark Generator (FW-Gen), a lightweight variational autoencoder-based framework that generates forged watermarks preserving the statistical properties of original watermarks while exhibiting distinct visual patterns. Our attack operates under a realistic threat model where an accused attacker, upon receiving an infringement claim, extracts watermark information from the protected dataset and produces counterfeit evidence to refute the allegation. Extensive experiments across six backdoor watermarking methods, two benchmark datasets,

This study explores real-world human interactions with large language models (LLMs) in diverse, unconstrained settings in contrast to most prior research focusing on ethically trimmed models like ChatGPT for specific tasks. We aim to understand the originator of toxicity. Our findings show that although LLMs are rightfully accused of providing toxic content, it is mostly demanded or at least provoked by humans who actively seek such content. Our manual analysis of hundreds of conversations judged as toxic by APIs commercial vendors, also raises questions with respect to current practices of what user requests are refused to answer. Furthermore, we conjecture based on multiple empirical indicators that humans exhibit a change of their mental model, switching from the mindset of interacting with a machine more towards interacting with a human.

The rise of Large Language Models (LLMs) has triggered legal and ethical concerns, especially regarding the unauthorized use of copyrighted materials in their training datasets. This has led to lawsuits against tech companies accused of using protected content without permission. Membership Inference Attacks (MIAs) aim to detect whether specific documents were used in a given LLM pretraining, but their effectiveness is undermined by biases such as time-shifts and n-gram overlaps. This paper addresses the evaluation of MIAs on LLMs with partially inferable training sets, under the ex-post hypothesis, which acknowledges inherent distributional biases between members and non-members datasets. We propose and validate algorithms to create ``non-biased'' and ``non-classifiable'' datasets for fairer MIA assessment. Experiments using the Gutenberg dataset on OpenLamma and Pythia show that neutralizing known biases alone is insufficient. Our methods produce non-biased ex-post datasets with AUC-ROC scores comparable to those previously obtained on genuinely random datasets, validating our approach. Globally, MIAs yield results close to random, with only one being effective on both random and

Broadly accessible generative AI models like Dall-E have made it possible for anyone to create compelling visual art. In online communities, the introduction of AI-generated content (AIGC) may impact social dynamics, for example causing changes in who is posting content, or shifting the norms or the discussions around the posted content if posts are suspected of being generated by AI. We take steps towards examining the potential impact of AIGC on art-related communities on Reddit. We distinguish between communities that disallow AI content and those without such a direct policy. We look at image-based posts in these communities where the author transparently shares that the image was created by AI, and at comments in these communities that suspect or accuse authors of using generative AI. We find that AI posts (and accusations) have played a surprisingly small part in these communities through the end of 2023, accounting for fewer than 0.5% of the image-based posts. However, even as the absolute number of author-labeled AI posts dwindles over time, accusations of AI use remain more persistent. We show that AI content is more readily used by newcomers and may help increase particip

As large language models (LLMs) become increasingly commonplace, concern about distinguishing between human and AI text increases as well. The growing power of these models is of particular concern to teachers, who may worry that students will use LLMs to write school assignments. Facing a technology with which they are unfamiliar, teachers may turn to publicly-available AI text detectors. Yet the accuracy of many of these detectors has not been thoroughly verified, posing potential harm to students who are falsely accused of academic dishonesty. In this paper, we evaluate three different AI text detectors-Kirchenbauer et al. watermarks, ZeroGPT, and GPTZero-against human and AI-generated essays. We find that watermarking results in a high false positive rate, and that ZeroGPT has both high false positive and false negative rates. Further, we are able to significantly increase the false negative rate of all detectors by using ChatGPT 3.5 to paraphrase the original AI-generated texts, thereby effectively bypassing the detectors.

This paper describes a parsing/generation system for finite verbal forms in Akkadian, with the possible addition of suffixes, implemented in Prolog. The work described provides the framework and engine to interpret the D, N, and G stems along with accusative, dative and ventive endings.

In the age of information abundance, attention is a coveted resource. Social media platforms vigorously compete for users' engagement, influencing the evolution of their opinions on a variety of topics. With recommendation algorithms often accused of creating "filter bubbles", where like-minded individuals interact predominantly with one another, it's crucial to understand the consequences of this unregulated attention market. To address this, we present a model of opinion dynamics on a multiplex network. Each layer of the network represents a distinct social media platform, each with its unique characteristics. Users, as nodes in this network, share their opinions across platforms and decide how much time to allocate in each platform depending on its perceived quality. Our model reveals two key findings. i) When examining two platforms - one with a neutral recommendation algorithm and another with a homophily-based algorithm - we uncover that even if users spend the majority of their time on the neutral platform, opinion polarization can persist. ii) By allowing users to dynamically allocate their social energy across platforms in accordance to their homophilic preferences, a furt

The climate crisis is a salient issue in online discussions, and hypocrisy accusations are a central rhetorical element in these debates. However, for large-scale text analysis, hypocrisy accusation detection is an understudied tool, most often defined as a smaller subtask of fallacious argument detection. In this paper, we define hypocrisy accusation detection as an independent task in NLP, and identify different relevant subtypes of hypocrisy accusations. Our Climate Hypocrisy Accusation Corpus (CHAC) consists of 420 Reddit climate debate comments, expert-annotated into two different types of hypocrisy accusations: personal versus political hypocrisy. We evaluate few-shot in-context learning with 6 shots and 3 instruction-tuned Large Language Models (LLMs) for detecting hypocrisy accusations in this dataset. Results indicate that the GPT-4o and Llama-3 models in particular show promise in detecting hypocrisy accusations (F1 reaching 0.68, while previous work shows F1 of 0.44). However, context matters for a complex semantic concept such as hypocrisy accusations, and we find models struggle especially at identifying political hypocrisy accusations compared to personal moral hypocr

How many times you need to change your method description because you were "accused" of plagiarism from text you already published? I will use this preprint to add all the methods I currently used in running the simulations for my research works. Then, I will cite it as needed.

Robert A. Millikan (1868-1953) was the second American to win the Nobel Prize in physics. At the peak of his influence, no scientist save Einstein was more admired by the American public. Millikan, the head of the California Institute of Technology (Caltech) during its first 24 years, oversaw its rapid growth into one of the leading scientific institutions of the world. In response to demands for social justice, Caltech reached a decision to strip Millikan of honors (such as the library named after him), following accusations against him. This article analyzes a specific accusation against Millikan that was published in Nature: that he collaborated to deprive Japanese Americans of their rights during their forced relocation to internment camps during the Second World War. An examination of original historical sources will show that this accusation is false. On the contrary, Millikan actively campaigned during the war to promote the rights of Japanese Americans. The article also treats Caltech's central accusation against Millikan: he lent his name to a morally reprehensible eugenics movement that had been scientifically discredited in his time. In a reversal of Caltech's claims, th