Recent breakthroughs in natural language processing (NLP) have permitted the synthesis and comprehension of coherent text in an open-ended way, therefore translating the theoretical algorithms into practical applications. The large language models (LLMs) have significantly impacted businesses such as report summarization software and copywriters. Observations indicate, however, that LLMs may exhibit social prejudice and toxicity, posing ethical and societal dangers of consequences resulting from irresponsibility. Large-scale benchmarks for accountable LLMs should consequently be developed. Although several empirical investigations reveal the existence of a few ethical difficulties in advanced LLMs, there is little systematic examination and user study of the risks and harmful behaviors of current LLM usage. To further educate future efforts on constructing ethical LLMs responsibly, we perform a qualitative research method called ``red teaming'' on OpenAI's ChatGPT\footnote{In this paper, ChatGPT refers to the version released on Dec 15th.} to better understand the practical features of ethical dangers in recent LLMs. We analyze ChatGPT comprehensively from four perspectives: 1) \textit{Bias} 2) \textit{Reliability} 3) \textit{Robustness} 4) \textit{Toxicity}. In accordance with our stated viewpoints, we empirically benchmark ChatGPT on multiple sample datasets. We find that a significant number of ethical risks cannot be addressed by existing benchmarks, and hence illustrate them via additional case studies. In addition, we examine the implications of our findings on AI ethics and harmal behaviors of ChatGPT, as well as future problems and practical design considerations for responsible LLMs. We believe that our findings may give light on future efforts to determine and mitigate the ethical hazards posed by machines in LLM applications.

Scaling up language models has led to unprecedented performance gains, but little is understood about how the training dynamics change as models get larger. How do language models of different sizes learn during pre-training? Why do larger language models demonstrate more desirable behaviors? In this paper, we analyze the intermediate training checkpoints of differently sized OPT models (Zhang et al.,2022)--from 125M to 175B parameters--on next-token prediction, sequence-level generation, and downstream tasks. We find that 1) at a given perplexity and independent of model sizes, a similar subset of training tokens see the most significant reduction in loss, with the rest stagnating or showing double-descent behavior; 2) early in training, all models learn to reduce the perplexity of grammatical sequences that contain hallucinations, with small models halting at this suboptimal distribution and larger ones eventually learning to assign these sequences lower probabilities; 3) perplexity is a strong predictor of in-context learning performance on 74 multiple-choice tasks from BIG-Bench, and this holds independent of the model size. Together, these results show that perplexity is more predictive of model behaviors than model size or training computation.

Machine learning models have demonstrated promising performance in many areas. However, the concerns that they can be biased against specific demographic groups hinder their adoption in high-stake applications. Thus, it is essential to ensure fairness in machine learning models. Most previous efforts require direct access to sensitive attributes for mitigating bias. Nonetheless, it is often infeasible to obtain large-scale users' sensitive attributes considering users' concerns about privacy in the data collection process. Privacy mechanisms such as local differential privacy (LDP) are widely enforced on sensitive information in the data collection stage due to legal compliance and people's increasing awareness of privacy. Therefore, a critical problem is how to make fair predictions under privacy. We study a novel and practical problem of fair classification in a semi-private setting, where most of the sensitive attributes are private and only a small amount of clean ones are available. To this end, we propose a novel framework FairSP that can achieve Fair prediction under the Semi-Private setting. First, FairSP learns to correct the noise-protected sensitive attributes by exploiting the limited clean sensitive attributes. Then, it jointly models the corrected and clean data in an adversarial way for debiasing and prediction. Theoretical analysis shows that the proposed model can ensure fairness under mild assumptions in the semi-private setting. Extensive experimental results on real-world datasets demonstrate the effectiveness of our method for making fair predictions under privacy and maintaining high accuracy.

This paper presents a new learning algorithm, termed Deep Bi-directional Predictive Coding (DBPC) that allows developing networks to simultaneously perform classification and reconstruction tasks using the same weights. Predictive Coding (PC) has emerged as a prominent theory underlying information processing in the brain. The general concept for learning in PC is that each layer learns to predict the activities of neurons in the previous layer which enables local computation of error and in-parallel learning across layers. In this paper, we extend existing PC approaches by developing a network which supports both feedforward and feedback propagation of information. Each layer in the networks trained using DBPC learn to predict the activities of neurons in the previous and next layer which allows the network to simultaneously perform classification and reconstruction tasks using feedforward and feedback propagation, respectively. DBPC also relies on locally available information for learning, thus enabling in-parallel learning across all layers in the network. The proposed approach has been developed for training both, fully connected networks and convolutional neural networks. The performance of DBPC has been evaluated on both, classification and reconstruction tasks using the MNIST and FashionMNIST datasets. The classification and the reconstruction performance of networks trained using DBPC is similar to other approaches used for comparison but DBPC uses a significantly smaller network. Further, the significant benefit of DBPC is its ability to achieve this performance using locally available information and in-parallel learning mechanisms which results in an efficient training protocol. This results clearly indicate that DBPC is a much more efficient approach for developing networks that can simultaneously perform both classification and reconstruction.

The Brazilian judiciary has a large workload, resulting in a long time to finish legal proceedings. Brazilian National Council of Justice has established in Resolution 469/2022 formal guidance for document and process digitalization opening up the possibility of using automatic techniques to help with everyday tasks in the legal field, particularly in a large number of texts yielded on the routine of law procedures. Notably, Artificial Intelligence (AI) techniques allow for processing and extracting useful information from textual data, potentially speeding up the process. However, datasets from the legal domain required by several AI techniques are scarce and difficult to obtain as they need labels from experts. To address this challenge, this article contributes with four datasets from the legal domain, two with documents and metadata but unlabeled, and another two labeled with a heuristic aiming at its use in textual semantic similarity tasks. Also, to evaluate the effectiveness of the proposed heuristic label process, this article presents a small ground truth dataset generated from domain expert annotations. The analysis of ground truth labels highlights that semantic analysis of domain text can be challenging even for domain experts. Also, the comparison between ground truth and heuristic labels shows that heuristic labels are useful.

In modern distributed computing applications, such as federated learning and AIoT systems, protecting privacy is crucial to prevent adversarial parties from colluding to steal others' private information. However, guaranteeing the utility of computation outcomes while protecting all parties' data privacy can be challenging, particularly when the parties' privacy requirements are highly heterogeneous. In this paper, we propose a novel privacy framework for multi-party computation called Threshold Personalized Multi-party Differential Privacy (TPMDP), which addresses a limited number of semi-honest colluding adversaries. Our framework enables each party to have a personalized privacy budget. We design a multi-party Gaussian mechanism that is easy to implement and satisfies TPMDP, wherein each party perturbs the computation outcome in a secure multi-party computation protocol using Gaussian noise. To optimize the utility of the mechanism, we cast the utility loss minimization problem into a linear programming (LP) problem. We exploit the specific structure of this LP problem to compute the optimal solution after O(n) computations, where n is the number of parties, while a generic solver may require exponentially many computations. Extensive experiments demonstrate the benefits of our approach in terms of low utility loss and high efficiency compared to existing private mechanisms that do not consider personalized privacy requirements or collusion thresholds.

To recognize and mitigate harms from large language models (LLMs), we need to understand the prevalence and nuances of stereotypes in LLM outputs. Toward this end, we present Marked Personas, a prompt-based method to measure stereotypes in LLMs for intersectional demographic groups without any lexicon or data labeling. Grounded in the sociolinguistic concept of markedness (which characterizes explicitly linguistically marked categories versus unmarked defaults), our proposed method is twofold: 1) prompting an LLM to generate personas, i.e., natural language descriptions, of the target demographic group alongside personas of unmarked, default groups; 2) identifying the words that significantly distinguish personas of the target group from corresponding unmarked ones. We find that the portrayals generated by GPT-3.5 and GPT-4 contain higher rates of racial stereotypes than human-written portrayals using the same prompts. The words distinguishing personas of marked (non-white, non-male) groups reflect patterns of othering and exoticizing these demographics. An intersectional lens further reveals tropes that dominate portrayals of marginalized groups, such as tropicalism and the hypersexualization of minoritized women. These representational harms have concerning implications for downstream applications like story generation.

With the hype around AI reaching a fever pitch in recent months, many people fear programs like ChatGPT will one day put them out of a job. For one New York lawyer, that nightmare could become a reality sooner than expected, but not for the reasons you might think. As reported by The New York Times, attorney Steven Schwartz of the law firm Levidow, Levidow and Oberman recently turned to OpenAI's chatbot for assistance with writing a legal brief, with predictably disastrous results. A lawyer used ChatGPT to do "legal research" and cited a number of nonexistent cases in a filing, and is now in a lot of trouble with the judge pic.twitter.com/AJSE7Ts7W7 Schwartz's firm has been suing the Columbian airline Avianca on behalf of Roberto Mata, who claims he was injured on a flight to John F. Kennedy International Airport in New York City.

The waters in Venice's main canal have turned fluorescent green in the area near Italy's renowned Rialto Bridge, as authorities seek to determine the cause. Italy's fire department posted a video on Sunday as one of its boats sailed on phosphorescent waters. "The Grand Canal coloured in green is what the fire department found this morning as we intervened together with ARPAV to collect samples and analyse this abnormal colour," it said. ARPAV, Veneto's regional environmental protection agency, said it received samples of the altered waters and was working to identify the substance that changed their colour. The Venice prefect has called an emergency meeting of police forces to understand what happened and study possible countermeasures, the ANSA news agency reported.

Recently, there has been a growing attention in image generation models. However, concerns have emerged regarding potential misuse and intellectual property (IP) infringement associated with these models. Therefore, it is necessary to analyze the origin of images by inferring if a specific image was generated by a particular model, i.e., origin attribution. Existing methods are limited in their applicability to specific types of generative models and require additional steps during training or generation. This restricts their use with pre-trained models that lack these specific operations and may compromise the quality of image generation. To overcome this problem, we first develop an alteration-free and model-agnostic origin attribution method via input reverse-engineering on image generation models, i.e., inverting the input of a particular model for a specific image. Given a particular model, we first analyze the differences in the hardness of reverse-engineering tasks for the generated images of the given model and other images. Based on our analysis, we propose a method that utilizes the reconstruction loss of reverse-engineering to infer the origin. Our proposed method effectively distinguishes between generated images from a specific generative model and other images, including those generated by different models and real images.