Large language models (LLMs) have emerged as an integral part of modern societies, powering user-facing applications such as personal assistants and enterprise applications like recruitment tools. Despite their utility, research indicates that LLMs perpetuate systemic biases. Yet, prior works on LLM harms predominantly focus on Western concepts like race and gender, often overlooking cultural concepts from other parts of the world. Additionally, these studies typically investigate "harm" as a singular dimension, ignoring the various and subtle forms in which harms manifest. To address this gap, we introduce the Covert Harms and Social Threats (CHAST), a set of seven metrics grounded in social science literature. We utilize evaluation models aligned with human assessments to examine the presence of covert harms in LLM-generated conversations, particularly in the context of recruitment. Our experiments reveal that seven out of the eight LLMs included in this study generated conversations riddled with CHAST, characterized by malign views expressed in seemingly neutral language unlikely to be detected by existing methods. Notably, these LLMs manifested more extreme views and opinions when dealing with non-Western concepts like caste, compared to Western ones such as race.

Composing poetry or lyrics involves several creative factors, but a challenging aspect of generation is the adherence to a more or less strict metric and rhyming pattern. To address this challenge specifically, previous work on the task has mainly focused on reverse language modeling, which brings the critical selection of each rhyming word to the forefront of each verse. On the other hand, reversing the word order requires that models be trained from scratch with this task-specific goal and cannot take advantage of transfer learning from a Pretrained Language Model (PLM). We propose a novel fine-tuning approach that prepends the rhyming word at the start of each lyric, which allows the critical rhyming decision to be made before the model commits to the content of the lyric (as during reverse language modeling), but maintains compatibility with the word order of regular PLMs as the lyric itself is still generated in left-to-right order. We conducted extensive experiments to compare this fine-tuning against the current state-of-the-art strategies for rhyming, finding that our approach generates more readable text and better rhyming capabilities. Furthermore, we furnish a high-quality dataset in English and 12 other languages, analyse the approach's feasibility in a multilingual context, provide extensive experimental results shedding light on good and bad practices for lyrics generation, and propose metrics to compare methods in the future.

This thesis is a corpus-based, quantitative, and typological analysis of the functions of Early Slavic participle constructions and their finite competitors ($jegda$-'when'-clauses). The first part leverages detailed linguistic annotation on Early Slavic corpora at the morphosyntactic, dependency, information-structural, and lexical levels to obtain indirect evidence for different potential functions of participle clauses and their main finite competitor and understand the roles of compositionality and default discourse reasoning as explanations for the distribution of participle constructions and $jegda$-clauses in the corpus. The second part uses massively parallel data to analyze typological variation in how languages express the semantic space of English $when$, whose scope encompasses that of Early Slavic participle constructions and $jegda$-clauses. Probabilistic semantic maps are generated and statistical methods (including Kriging, Gaussian Mixture Modelling, precision and recall analysis) are used to induce cross-linguistically salient dimensions from the parallel corpus and to study conceptual variation within the semantic space of the hypothetical concept WHEN.

The rapid development of large language models has led to the widespread adoption of Retrieval-Augmented Generation (RAG), which integrates external knowledge to alleviate knowledge bottlenecks and mitigate hallucinations. However, the existing RAG paradigm inevitably suffers from the impact of flawed information introduced during the retrieval phrase, thereby diminishing the reliability and correctness of the generated outcomes. In this paper, we propose Credibility-aware Generation (CAG), a universally applicable framework designed to mitigate the impact of flawed information in RAG. At its core, CAG aims to equip models with the ability to discern and process information based on its credibility. To this end, we propose an innovative data transformation framework that generates data based on credibility, thereby effectively endowing models with the capability of CAG. Furthermore, to accurately evaluate the models' capabilities of CAG, we construct a comprehensive benchmark covering three critical real-world scenarios. Experimental results demonstrate that our model can effectively understand and utilize credibility for generation, significantly outperform other models with retrieval augmentation, and exhibit resilience against the disruption caused by noisy documents, thereby maintaining robust performance. Moreover, our model supports customized credibility, offering a wide range of potential applications.

Automatic citation generation for sentences in a document or report is paramount for intelligence analysts, cybersecurity, news agencies, and education personnel. In this research, we investigate whether large language models (LLMs) are capable of generating references based on two forms of sentence queries: (a) Direct Queries, LLMs are asked to provide author names of the given research article, and (b) Indirect Queries, LLMs are asked to provide the title of a mentioned article when given a sentence from a different article. To demonstrate where LLM stands in this task, we introduce a large dataset called REASONS comprising abstracts of the 12 most popular domains of scientific research on arXiv. From around 20K research articles, we make the following deductions on public and proprietary LLMs: (a) State-of-the-art, often called anthropomorphic GPT-4 and GPT-3.5, suffers from high pass percentage (PP) to minimize the hallucination rate (HR). When tested with Perplexity.ai (7B), they unexpectedly made more errors; (b) Augmenting relevant metadata lowered the PP and gave the lowest HR; (c) Advance retrieval-augmented generation (RAG) using Mistral demonstrates consistent and robust citation support on indirect queries and matched performance to GPT-3.5 and GPT-4. The HR across all domains and models decreased by an average of 41.93%, and the PP was reduced to 0% in most cases. In terms of generation quality, the average F1 Score and BLEU were 68.09% and 57.51%, respectively; (d) Testing with adversarial samples showed that LLMs, including the Advance RAG Mistral, struggle to understand context, but the extent of this issue was small in Mistral and GPT-4-Preview. Our study contributes valuable insights into the reliability of RAG for automated citation generation tasks.

Based on Pre-trained Language Models (PLMs), event coreference resolution (ECR) systems have demonstrated outstanding performance in clustering coreferential events across documents. However, the state-of-the-art system exhibits an excessive reliance on the'triggers lexical matching' spurious pattern in the input mention pair text. We formalize the decision-making process of the baseline ECR system using a Structural Causal Model (SCM), aiming to identify spurious and causal associations (i.e., rationales) within the ECR task. Leveraging the debiasing capability of counterfactual data augmentation, we develop a rationale-centric counterfactual data augmentation method with LLM-in-the-loop. This method is specialized for pairwise input in the Figure 1: The distribution of'triggers lexical matching' ECR system, where we conduct direct interventions in mention pairs from ECB+ training set, along with a on triggers and context to mitigate the false negative example from Held et al.'s system which spurious association while emphasizing the causation.

OpenAI is partnering with another publisher as it moves towards a licensed approach to training materials. Dotdash Meredith, the owner of brands like People and Better Homes & Gardens, will license its content for OpenAI to train ChatGPT while the publisher will use the AI company's models to boost its in-house ad-targeting tool. As part of the arrangement, ChatGPT will display content and links attributed to Dotdash Meredith's publications. It also provides OpenAI with fully licensed training material from trusted publications. That's a welcome change after the company got in hot water for allegedly using content for training purposes without permission.

Monday night's'Garden of Time'-themed Met Gala was the place to be seen for the biggest names in music, film, and fashion -- even if some stars weren't really there. 'I intend to continue creating content that inspires and enchants fans, without disrespecting the artist's image,' Brazilian fan'Sali' told DailyMail.com of his work. Sali -- who also crafted an uncanny AI fake of Billie Eilish at last night's gala -- knows his work is'sparking debates about ethical AI use,' particularly in an election year that has already seen AI-faked Biden robocalls and AI images of Trump supporters. The finished product was so vivid that the fans even fooled Katy Perry's own mother Brazilian Katy Perry fan'Sali' told DailyMail.com he knows his work is'sparking debates about ethical AI use,' particularly in an election year. But this sincere and dedicated Perry fan hopes there will still be freedom for creating'positive and admirable content' with advanced AI But this sincere and dedicated Katy Perry fan hopes there will still be freedom for the creation of'positive and admirable content' with the shockingly advanced tech, as he explained to DailyMail.com the craftsman-like process he used to make his fakes.

This story originally appeared on WIRED Italia, and has been translated from Italian. The Met Gala is undoubtedly one of the most anticipated events of the year, but this time the music and entertainment celebrities who graced its red carpet had some competition for the public's attention: generative AI deepfakes. In a post published on X this morning--and now counting nearly 15 million views--Katy Perry is pictured wearing a stunning dress decorated with three-dimensional floral appliqués, which descends to the ground transforming into incredibly realistic-looking moss. But the image is far from real, as a Community Note attached to the post makes clear. Not surprisingly, a few minutes after this first photo of the singer at the Met Gala was shared, another one immediately arrived showing her wearing a bronze-colored corset and a gorgeous floral skirt, in perfect Xena style.

A YouTube based in Florida's iconic Haulover Inlet, set between Bal Harbour and Sunny Isles in Miami-Dade County, posted this video during a boozing weekend. The family of one of two teen boys facing felonies for dumping drums of trash into the Atlantic Ocean at Florida's annual Boca Bash issued an apology after their son turned himself in to the Florida Fish and Wildlife Conservation Commission (FWC). Now-viral drone footage shows the teens hefting two trash bins filled with bottles and other plastics over the railing of their fishing vessel as they speed away from the boozy water gathering on April 28. As the boat of partiers zoomed away into the choppy waters of the Boca Raton inlet, the video pans out to the spread of debris left floating in their wake. Footage from the front of the boat shows the teens waving and laughing.