Artificial intelligence had its break out year in 2023. We chose those two language models because they use live information from the internet to make their predictions, unlike ChatGPT and Microsoft's Bing which rely on older data. AGI is a theoretical intelligent agent able to complete any intellectual task a human can - and the arrival of AGI is forecast to cause huge changes to human society. Backed by Google and Amazon, Claude's parent company Anthropic was founded by former members of OpenAI, makers of ChatGPT. 'In recent years we've seen AI algorithms match or exceed human performance in specialized tasks like object recognition, game playing, and language processing.

Large language models like ChatGPT write impressively well--so well, in fact, that they've become a problem. Students have begun using these models to ghostwrite assignments, leading some schools to ban ChatGPT. In addition, these models are also prone to producing text with factual errors, so wary readers may want to know if generative AI tools have been used to ghostwrite news articles or other sources before trusting them. What can teachers and consumers do? Existing tools to detect AI-generated text sometimes do poorly on data that differs from what they were trained on.

Harnessing the power of human-annotated data through Supervised Fine-Tuning (SFT) is pivotal for advancing Large Language Models (LLMs). In this paper, we delve into the prospect of growing a strong LLM out of a weak one without the need for acquiring additional human-annotated data. We propose a new fine-tuning method called Self-Play fIne-tuNing (SPIN), which starts from a supervised fine-tuned model. At the heart of SPIN lies a self-play mechanism, where the LLM refines its capability by playing against instances of itself. More specifically, the LLM generates its own training data from its previous iterations, refining its policy by discerning these self-generated responses from those obtained from human-annotated data. Our method progressively elevates the LLM from a nascent model to a formidable one, unlocking the full potential of human-annotated demonstration data for SFT. Theoretically, we prove that the global optimum to the training objective function of our method is achieved only when the LLM policy aligns with the target data distribution. Empirically, we evaluate our method on several benchmark datasets including the HuggingFace Open LLM Leaderboard, MT-Bench, and datasets from Big-Bench. Our results show that SPIN can significantly improve the LLM's performance across a variety of benchmarks and even outperform models trained through direct preference optimization (DPO) supplemented with extra GPT-4 preference data. This sheds light on the promise of self-play, enabling the achievement of human-level performance in LLMs without the need for expert opponents.

As large language models (LLM) evolve in their capabilities, various recent studies have tried to quantify their behavior using psychological tools created to study human behavior. One such example is the measurement of "personality" of LLMs using self-assessment personality tests developed to measure human personality. Yet almost none of these works verify the applicability of these tests on LLMs. In this paper, we analyze the reliability of LLM personality scores obtained from self-assessment personality tests using two simple experiments. We first introduce the property of prompt sensitivity, where three semantically equivalent prompts representing three intuitive ways of administering self-assessment tests on LLMs are used to measure the personality of the same LLM. We find that all three prompts lead to very different personality scores, a difference that is statistically significant for all traits in a large majority of scenarios. We then introduce the property of option-order symmetry for personality measurement of LLMs. Since most of the self-assessment tests exist in the form of multiple choice question (MCQ) questions, we argue that the scores should also be robust to not just the prompt template but also the order in which the options are presented. This test unsurprisingly reveals that the self-assessment test scores are not robust to the order of the options. These simple tests, done on ChatGPT and three Llama2 models of different sizes, show that self-assessment personality tests created for humans are unreliable measures of personality in LLMs.

Organizations face the challenge of ensuring compliance with an increasing amount of requirements from various regulatory documents. Which requirements are relevant depends on aspects such as the geographic location of the organization, its domain, size, and business processes. Considering these contextual factors, as a first step, relevant documents (e.g., laws, regulations, directives, policies) are identified, followed by a more detailed analysis of which parts of the identified documents are relevant for which step of a given business process. Nowadays the identification of regulatory requirements relevant to business processes is mostly done manually by domain and legal experts, posing a tremendous effort on them, especially for a large number of regulatory documents which might frequently change. Hence, this work examines how legal and domain experts can be assisted in the assessment of relevant requirements. For this, we compare an embedding-based NLP ranking method, a generative AI method using GPT-4, and a crowdsourced method with the purely manual method of creating relevancy labels by experts. The proposed methods are evaluated based on two case studies: an Australian insurance case created with domain experts and a global banking use case, adapted from SAP Signavio's workflow example of an international guideline. A gold standard is created for both BPMN2.0 processes and matched to real-world textual requirements from multiple regulatory documents. The evaluation and discussion provide insights into strengths and weaknesses of each method regarding applicability, automation, transparency, and reproducibility and provide guidelines on which method combinations will maximize benefits for given characteristics such as process usage, impact, and dynamics of an application scenario.

Existing object recognition models have been shown to lack robustness in diverse geographical scenarios due to significant domain shifts in design and context. Class representations need to be adapted to more accurately reflect an object concept under these shifts. In the absence of training data from target geographies, we hypothesize that geography-specific descriptive knowledge of object categories can be leveraged to enhance robustness. For this purpose, we explore the feasibility of probing a large-language model for geography-specific object knowledge, and we investigate integrating knowledge in zero-shot and learnable soft prompting with the CLIP vision-language model. In particular, we propose a geography knowledge regularization method to ensure that soft prompts trained on a source set of geographies generalize to an unseen target set of geographies. Our gains on DollarStreet when generalizing from a model trained only on data from Europe are as large as +2.8 on countries from Africa, and +4.6 on the hardest classes. We further show competitive performance vs. few-shot target training, and provide insights into how descriptive knowledge captures geographical differences.

Summarization of electronic health records (EHRs) can substantially minimize 'screen time' for both patients as well as medical personnel. In recent years summarization of EHRs have employed machine learning pipelines using state of the art neural models. However, these models have produced less than adequate results that are attributed to the difficulty of obtaining sufficient annotated data for training. Moreover, the requirement to consider the entire content of an EHR in summarization has resulted in poor performance due to the fact that attention mechanisms in modern large language models (LLMs) adds a quadratic complexity in terms of the size of the input. We propose here a method that mitigates these shortcomings by combining semantic search, retrieval augmented generation (RAG) and question-answering using the latest LLMs. In our approach summarization is the extraction of answers to specific questions that are deemed important by subject-matter experts (SMEs). Our approach is quite efficient; requires minimal to no training; does not suffer from the 'hallucination' problem of LLMs; and it ensures diversity, since the summary will not have repeated content but diverse answers to specific questions.

Visual attribution in medical imaging seeks to make evident the diagnostically-relevant components of a medical image, in contrast to the more common detection of diseased tissue deployed in standard machine vision pipelines (which are less straightforwardly interpretable/explainable to clinicians). We here present a novel generative visual attribution technique, one that leverages latent diffusion models in combination with domain-specific large language models, in order to generate normal counterparts of abnormal images. The discrepancy between the two hence gives rise to a mapping indicating the diagnostically-relevant image components. To achieve this, we deploy image priors in conjunction with appropriate conditioning mechanisms in order to control the image generative process, including natural language text prompts acquired from medical science and applied radiology. We perform experiments and quantitatively evaluate our results on the COVID-19 Radiography Database containing labelled chest X-rays with differing pathologies via the Frechet Inception Distance (FID), Structural Similarity (SSIM) and Multi Scale Structural Similarity Metric (MS-SSIM) metrics obtained between real and generated images. The resulting system also exhibits a range of latent capabilities including zero-shot localized disease induction, which are evaluated with real examples from the cheXpert dataset.

Does Donald Trump speak differently from other presidents? If so, in what ways? Are these differences confined to any single medium of communication? To investigate these questions, this paper introduces a novel metric of uniqueness based on large language models, develops a new lexicon for divisive speech, and presents a framework for comparing the lexical features of political opponents. Applying these tools to a variety of corpora of presidential speeches, we find considerable evidence that Trump's speech patterns diverge from those of all major party nominees for the presidency in recent history. Some notable findings include Trump's employment of particularly divisive and antagonistic language targeting of his political opponents and his patterns of repetition for emphasis. Furthermore, Trump is significantly more distinctive than his fellow Republicans, whose uniqueness values are comparably closer to those of the Democrats. These differences hold across a variety of measurement strategies, arise on both the campaign trail and in official presidential addresses, and do not appear to be an artifact of secular time trends.

Generative Large Language Models (LLMs) based on the Transformer architecture have recently emerged as a dominant foundation model for a wide range of Natural Language Processing tasks. Nevertheless, their application in real-time scenarios has been highly restricted due to the significant inference latency associated with these models. This is particularly pronounced due to the autoregressive nature of generative LLM inference, where tokens are generated sequentially since each token depends on all previous output tokens. It is therefore challenging to achieve any token-level parallelism, making inference extremely memory-bound. In this work, we propose SPEED, which improves inference efficiency by speculatively executing multiple future tokens in parallel with the current token using predicted values based on early-layer hidden states. For Transformer decoders which employ parameter sharing, the memory operations for the tokens executing in parallel can be amortized, which allows us to accelerate generative LLM inference. We demonstrate the efficiency of our method in terms of latency reduction relative to model accuracy and demonstrate how speculation allows for training deeper decoders with parameter sharing with minimal runtime overhead.