Large Language Models (LLMs) have achieved remarkable success across various industries due to their exceptional generative capabilities. However, for safe and effective real-world deployments, ensuring honesty and helpfulness is critical. This paper addresses the question: Can we prioritize the helpfulness of LLMs while preserving their honesty? To begin with, we establish exhaustive principles aimed at guaranteeing the honesty of LLM. Additionally, we introduce a novel dataset, referred to as HoneSet, comprising 930 queries spanning six categories meticulously crafted to assess an LLM's capacity for maintaining honesty. Subsequently, we present two approaches to augmenting honesty and helpfulness in LLMs: a training-free enhancement and a fine-tuning-based improvement. The training-free approach, which is based on curiosity-driven prompting, empowers LLMs to articulate internal confusion and uncertainty regarding queries, thereby optimizing their responses. Conversely, the fine-tuning-based method employs a two-stage process inspired by curriculum learning: initially instructing LLMs to discern between honest and dishonest responses, then refining their training to enhance helpfulness. Experiments conducted on nine prominent LLMs demonstrate a significant improvement in alignment with honesty across all models through the implementation of our proposed enhancements. Particularly noteworthy is the 65.3% enhancement observed in Llama3-8b and the remarkable 124.7% improvement in Mistral-7b, as measured by the H$^{2}$ (honest and helpful) assessment. We believe that our work can pave the way for developing more trustworthy LLMs for real-world applications.

Deep models have recently emerged as a promising tool to solve partial differential equations (PDEs), known as neural PDE solvers. While neural solvers trained from either simulation data or physics-informed loss can solve the PDEs reasonably well, they are mainly restricted to a specific set of PDEs, e.g. a certain equation or a finite set of coefficients. This bottleneck limits the generalizability of neural solvers, which is widely recognized as its major advantage over numerical solvers. In this paper, we present the Universal PDE solver (Unisolver) capable of solving a wide scope of PDEs by leveraging a Transformer pre-trained on diverse data and conditioned on diverse PDEs. Instead of simply scaling up data and parameters, Unisolver stems from the theoretical analysis of the PDE-solving process. Our key finding is that a PDE solution is fundamentally under the control of a series of PDE components, e.g. equation symbols, coefficients, and initial and boundary conditions. Inspired by the mathematical structure of PDEs, we define a complete set of PDE components and correspondingly embed them as domain-wise (e.g. equation symbols) and point-wise (e.g. boundaries) conditions for Transformer PDE solvers. Integrating physical insights with recent Transformer advances, Unisolver achieves consistent state-of-the-art results on three challenging large-scale benchmarks, showing impressive gains and endowing favorable generalizability and scalability.

A week after its algorithms advised people to eat rocks and put glue on pizza, Google admitted Thursday that it needed to make adjustments to its bold new generative AI search feature. The episode highlights the risks of Google's aggressive drive to commercialize generative AI--and also the treacherous and fundamental limitations of that technology. Google's AI Overviews feature draws on Gemini, a large language model like the one behind OpenAI's ChatGPT, to generate written answers to some search queries by summarizing information found online. The current AI boom is built around LLMs' impressive fluency with text, but the software can also use that facility to put a convincing gloss on untruths or errors. Using the technology to summarize online information promises can make search results easier to digest, but it is hazardous when online sources are contractionary or when people may use the information to make important decisions.

Apple is working on a version of its Siri voice assistant that will use advanced AI powered by large language models (LLMs), Bloomberg has reported. The technology will allow users to perform specific app functions with their voices, such as opening documents, sending emails and more. The new version of Siri will only work on Apple's own apps to start with. It won't arrive with iOS 18 but may be released subsequently as an update early next year, the report states. The assistant will be able to analyze your phone's activity and automatically enable Siri-controlled features. It'll support "hundreds" of commands but will only be able to process one at a time at first, according to the article.

Unfortunately, AI systems are inherently unreliable. Within days of AI Overviews' release in the US, users were sharing examples of responses that were strange at best. It suggested that users add glue to pizza or eat at least one small rock a day, and that former US president Andrew Johnson earned university degrees between 1947 and 2012, despite dying in 1875. On Thursday, Liz Reid, head of Google Search, announced that the company has been making technical improvements to the system to make it less likely to generate incorrect answers, including better detection mechanisms for nonsensical queries. It is also limiting the inclusion of satirical, humorous, and user-generated content in responses, since such material could result in misleading advice.

Artificial intelligence company OpenAI has announced that it disrupted covert influence campaigns originating from Russia, China, Israel and Iran. The ChatGPT maker said on Thursday that it identified five campaigns involving "deceptive attempts to manipulate public opinion or influence political outcomes without revealing the true identity or intentions of the actors behind them". The campaigns used OpenAI's models to generate text and images that were posted across social media platforms such as Telegram, X, and Instagram, in some cases exploiting the tools to produce content with "fewer language errors than would have been possible for human operators," OpenAI said. Open AI said it terminated accounts associated with two Russian operations, dubbed Bad Grammer and Doppelganger; a Chinese campaign known as Spamouflage; an Iranian network called International Union of Virtual Media; and an Israeli operation dubbed Zero Zeno. "We are committed to developing safe and responsible AI, which involves designing our models with safety in mind and proactively intervening against malicious use," the California-based start-up said in a statement posted on its website.

Entity matching (EM) is a critical task in data integration, aiming to identify records across different datasets that refer to the same real-world entities. Traditional methods often rely on manually engineered features and rule-based systems, which struggle with diverse and unstructured data. The emergence of Large Language Models (LLMs) such as GPT-4 offers transformative potential for EM, leveraging their advanced semantic understanding and contextual capabilities. This vision paper explores the application of LLMs to EM, discussing their advantages, challenges, and future research directions. Additionally, we review related work on applying weak supervision and unsupervised approaches to EM, highlighting how LLMs can enhance these methods.

Natural language processing (NLP) is rapidly developing for applications in educational assessment. In this paper, I describe an NLP-based procedure that can be used to support subject matter experts in establishing a crosswalk between item specifications and content standards. This paper extends recent work by proposing and demonstrating the use of multivariate similarity based on embedding vectors for sentences or texts. In particular, a hybrid regression procedure is demonstrated for establishing the match of each content standard to multiple item specifications. The procedure is used to evaluate the match of the Common Core State Standards (CCSS) for mathematics at grade 4 to the corresponding item specifications for the 2026 National Assessment of Educational Progress (NAEP).

The current gold standard for evaluating generated chest x-ray (CXR) reports is through radiologist annotations. However, this process can be extremely time-consuming and costly, especially when evaluating large numbers of reports. In this work, we present FineRad-Score, a Large Language Model (LLM)-based automated evaluation metric for generated CXR reports. Given a candidate report and a ground-truth report, FineRadScore gives the minimum number of line-by-line corrections required to go from the candidate to the ground-truth report. Additionally, FineRadScore provides an error severity rating with each correction and generates comments explaining why the correction was needed. We demonstrate that FineRadScore's corrections and error severity scores align with radiologist opinions. We also show that, when used to judge the quality of the report as a whole, FineRadScore aligns with radiologists as well as current state-of-the-art automated CXR evaluation metrics. Finally, we analyze FineRadScore's shortcomings to provide suggestions for future improvements. Code to run FineRadScore can be found here.

Deep neural networks have exhibited substantial success in the field of Natural Language Processing and ensuring their safety and reliability is crucial: there are safety critical contexts where such models must be robust to variability or attack, and give guarantees over their output. Unlike Computer Vision, NLP lacks a unified verification methodology and, despite recent advancements in literature, they are often light on the pragmatical issues of NLP verification. In this paper, we attempt to distil and evaluate general components of an NLP verification pipeline, that emerges from the progress in the field to date. Our contributions are two-fold. Firstly, we give a general (i.e. algorithm-independent) characterisation of verifiable subspaces that result from embedding sentences into continuous spaces. We identify, and give an effective method to deal with, the technical challenge of semantic generalisability of verified subspaces; and propose it as a standard metric in the NLP verification pipelines (alongside with the standard metrics of model accuracy and model verifiability). Secondly, we propose a general methodology to analyse the effect of the embedding gap -- a problem that refers to the discrepancy between verification of geometric subspaces, and the semantic meaning of sentences which the geometric subspaces are supposed to represent. In extreme cases, poor choices in embedding of sentences may invalidate verification results. We propose a number of practical NLP methods that can help to quantify the effects of the embedding gap; and in particular we propose the metric of falsifiability of semantic subspaces as another fundamental metric to be reported as part of the NLP verification pipeline. We believe that together these general principles pave the way towards a more consolidated and effective development of this new domain.