The rapid deployment of generative language models (LMs) has raised concerns about social biases affecting the well-being of diverse consumers. The extant literature on generative LMs has primarily examined bias via explicit identity prompting. However, prior research on bias in earlier language-based technology platforms, including search engines, has shown that discrimination can occur even when identity terms are not specified explicitly. Studies of bias in LM responses to open-ended prompts (where identity classifications are left unspecified) are lacking and have not yet been grounded in end-consumer harms. Here, we advance studies of generative LM bias by considering a broader set of natural use cases via open-ended prompting. In this "laissez-faire" setting, we find that synthetically generated texts from five of the most pervasive LMs (ChatGPT3.5, ChatGPT4, Claude2.0, Llama2, and PaLM2) perpetuate harms of omission, subordination, and stereotyping for minoritized individuals with intersectional race, gender, and/or sexual orientation identities (AI/AN, Asian, Black, Latine, MENA, NH/PI, Female, Non-binary, Queer). We find widespread evidence of bias to an extent that such individuals are hundreds to thousands of times more likely to encounter LM-generated outputs that portray their identities in a subordinated manner compared to representative or empowering portrayals. We also document a prevalence of stereotypes (e.g. perpetual foreigner) in LM-generated outputs that are known to trigger psychological harms that disproportionately affect minoritized individuals. These include stereotype threat, which leads to impaired cognitive performance and increased negative self-perception. Our findings highlight the urgent need to protect consumers from discriminatory harms caused by language models and invest in critical AI education programs tailored towards empowering diverse consumers.

Methods that can generate synthetic speech which is perceptually indistinguishable from speech recorded by a human speaker, are easily available. Several incidents report misuse of synthetic speech generated from these methods to commit fraud. To counter such misuse, many methods have been proposed to detect synthetic speech. Some of these detectors are more interpretable, can generalize to detect synthetic speech in the wild and are robust to noise. However, limited work has been done on understanding bias in these detectors. In this work, we examine bias in existing synthetic speech detectors to determine if they will unfairly target a particular gender, age and accent group. We also inspect whether these detectors will have a higher misclassification rate for bona fide speech from speech-impaired speakers w.r.t fluent speakers. Extensive experiments on 6 existing synthetic speech detectors using more than 0.9 million speech signals demonstrate that most detectors are gender, age and accent biased, and future work is needed to ensure fairness. To support future research, we release our evaluation dataset, models used in our study and source code at https://gitlab.com/viper-purdue/fairssd.

While the progression of Large Language Models (LLMs) has notably propelled financial analysis, their application has largely been confined to singular language realms, leaving untapped the potential of bilingual Chinese-English capacity. To bridge this chasm, we introduce ICE-PIXIU, seamlessly amalgamating the ICE-INTENT model and ICE-FLARE benchmark for bilingual financial analysis. ICE-PIXIU uniquely integrates a spectrum of Chinese tasks, alongside translated and original English datasets, enriching the breadth and depth of bilingual financial modeling. It provides unrestricted access to diverse model variants, a substantial compilation of diverse cross-lingual and multi-modal instruction data, and an evaluation benchmark with expert annotations, comprising 10 NLP tasks, 20 bilingual specific tasks, totaling 95k datasets. Our thorough evaluation emphasizes the advantages of incorporating these bilingual datasets, especially in translation tasks and utilizing original English data, enhancing both linguistic flexibility and analytical acuity in financial contexts. Notably, ICE-INTENT distinguishes itself by showcasing significant enhancements over conventional LLMs and existing financial LLMs in bilingual milieus, underscoring the profound impact of robust bilingual data on the accuracy and efficacy of financial NLP.

Dense passage retrieval (DPR) is the first step in the retrieval augmented generation (RAG) paradigm for improving the performance of large language models (LLM). DPR fine-tunes pre-trained networks to enhance the alignment of the embeddings between queries and relevant textual data. A deeper understanding of DPR fine-tuning will be required to fundamentally unlock the full potential of this approach. In this work, we explore DPR-trained models mechanistically by using a combination of probing, layer activation analysis, and model editing. Our experiments show that DPR training decentralizes how knowledge is stored in the network, creating multiple access pathways to the same information. We also uncover a limitation in this training style: the internal knowledge of the pre-trained model bounds what the retrieval model can retrieve. These findings suggest a few possible directions for dense retrieval: (1) expose the DPR training process to more knowledge so more can be decentralized, (2) inject facts as decentralized representations, (3) model and incorporate knowledge uncertainty in the retrieval process, and (4) directly map internal model knowledge to a knowledge base.

Recognizing if LLM output can be grounded in evidence is central to many tasks in NLP: retrieval-augmented generation, summarization, document-grounded dialogue, and more. Current approaches to this kind of "fact-checking" are based on verifying each piece of a model generation against potential evidence using an LLM. However, this process can be very computationally expensive, requiring many calls to LLMs to check a single response. In this work, we show how to build small models that have GPT-4-level performance but for 400x lower cost. We do this by constructing synthetic training data with GPT-4, which involves creating realistic yet challenging instances of factual errors via a structured generation procedure. Training on this data teaches models to check each fact in the claim and recognize synthesis of information across sentences. For evaluation, we unify pre-existing datasets into a benchmark LLM-AggreFact, collected from recent work on fact-checking and grounding LLM generations. Our best system MiniCheck-FT5 (770M parameters) outperforms all systems of comparable size and reaches GPT-4 accuracy. We release LLM-AggreFact, code for data synthesis, and models.

Creating a sexually explicit "deepfake" image is to be made an offence under a new law, the Ministry of Justice has announced. Under the legislation, anyone who creates such an image without consent will face a criminal record and an unlimited fine. They could also face jail if the image is shared more widely. The creation of a deepfake image will be an offence regardless of whether the creator intended to share it, the department said. The Online Safety Act, introduced last year, has already criminalised the sharing of deepfake intimate images, whose creation is being facilitated by advances in artificial intelligence.

For people in eastern Ukraine, where nightly barrages of drones from Russia outpace the military's overwhelmed air defenses, the response by Western allies to Iran's aerial assault against Israel this weekend produced uncomfortable comparisons. The militaries of the United States, Britain, France and others stepped in to help Israel defend against the fusillade of more than 300 Iranian drones and missiles, nearly all of which were intercepted. A similar number of aerial weapons are fired at Ukraine on a weekly basis, its officials say, with many of the drones in those attacks designed by Iran and now produced by Russia. Since the start of this year, Russia has fired 1,000 missiles, 2,800 drones and 7,000 guided aerial bombs at Ukraine, according to Ukraine's permanent representative to the United Nations, Sergiy Kyslytsya. While Washington and other allies have provided Kyiv with some powerful air defense weapons, they have not directly confronted Russian forces, and Ukrainian officials have long argued that the supplied weapons are insufficient to counter the threat from Moscow.

An intricate network of Israel's missile defense tech faced a serious test of its mettle Saturday night, downing '99 percent' of an aerial assault launched from Iran. Approximately 170 Iranian drones, 120 ballistic missiles and over 30 cruise missiles had been launched from the Iranian territory in the attack, soaring over 1,100 miles. Iran's airborne phalanx was repelled by ground-based anti-air missiles with names like the'Iron Dome,' 'David's Sling' and'Arrow-3,' the latest hardware in Israel's frequently updated national defense arsenal. Below, an overview of the equipment Israel has developed, sometimes with the help of American military contractors, and how it keeps bombardments in check. First operational in 2011, Israel's Iron Dome faced its first test over a decade ago, when militants in Gaza fired an estimated 1,500 rockets at Israel over eight days in Nov. 2014 - at least 10 Iron Dome missile batteries are known to exist, total (like this one pictured above) First operational in 2011, Israel's Iron Dome faced its first test over a decade ago, when militants in Gaza fired an estimated 1,500 rockets at Israel over eight days in November of 2014.

Former Secretary of State Hillary Clinton took a swipe at voters "upset" by the forthcoming rematch between President Biden and former President Trump during her appearance on "The Tonight Show." Hillary Clinton reacted to a recent ruling in Arizona, which bans abortion in nearly all circumstances, calling it "cruelty" during an interview with Kelly Clarkson and encouraging Americans to vote in a way that would "make life better" for the largest number of people. "I feared it would happen but I hoped it wouldn't happen. Now here we are in the middle of this very difficult period for women in about half the states in our country, who cannot get the care that they need. And the old law in Arizona is without exceptions and the danger to women's lives as well as to our right to make our own decisions about our bodies and ourselves is so profound," Clinton said during the interview with Clarkson on "The Kelly Clarkson Show."

Video captures the moment and aftermath of what appears to be a drone, allegedly of Ukrainian origin, striking Russian drone production facility. Russian officials claimed that only a worker's dormitory was hit. The success of Israel and its allies in largely thwarting a massive Iranian missile and drone attack shows what Ukraine could achieve against Russian aerial barrages if it had more support from its partners, Ukrainian Foreign Minister Dmytro Kuleba said Monday. A recent Russian aerial campaign targeting Ukraine's energy infrastructure and other targets has wrought extensive damage, and Ukrainian officials have pleaded with the country's Western allies to provide more air defense systems as the war stretches into its third year. Israel's defense system, with assistance from the U.S. and Britain -- countries that are also supporting Ukraine's war effort -- is credited with preventing serious damage or casualties in Sunday's attack by Iran using more than 300 drones and missiles. Kuleba, speaking to reporters in Kyiv, urged Ukraine's allies to "give us what we need and we will do the rest of the job."