Large Language Models (LLMs) have recently demonstrated remarkable capabilities in natural language processing tasks and beyond. This success of LLMs has led to a large influx of research contributions in this direction. These works encompass diverse topics such as architectural innovations, better training strategies, context length improvements, fine-tuning, multi-modal LLMs, robotics, datasets, benchmarking, efficiency, and more. With the rapid development of techniques and regular breakthroughs in LLM research, it has become considerably challenging to perceive the bigger picture of the advances in this direction. Considering the rapidly emerging plethora of literature on LLMs, it is imperative that the research community is able to benefit from a concise yet comprehensive overview of the recent developments in this field. This article provides an overview of the existing literature on a broad range of LLM-related concepts. Our self-contained comprehensive overview of LLMs discusses relevant background concepts along with covering the advanced topics at the frontier of research in LLMs. This review article is intended to not only provide a systematic survey but also a quick comprehensive reference for the researchers and practitioners to draw insights from extensive informative summaries of the existing works to advance the LLM research.

Large Language Models (LLMs) with strong abilities in natural language processing tasks have emerged and have been applied in various kinds of areas such as science, finance and software engineering. However, the capability of LLMs to advance the field of chemistry remains unclear. In this paper, rather than pursuing state-of-the-art performance, we aim to evaluate capabilities of LLMs in a wide range of tasks across the chemistry domain. We identify three key chemistry-related capabilities including understanding, reasoning and explaining to explore in LLMs and establish a benchmark containing eight chemistry tasks. Our analysis draws on widely recognized datasets facilitating a broad exploration of the capacities of LLMs within the context of practical chemistry. Five LLMs (GPT-4, GPT-3.5, Davinci-003, Llama and Galactica) are evaluated for each chemistry task in zero-shot and few-shot in-context learning settings with carefully selected demonstration examples and specially crafted prompts. Our investigation found that GPT-4 outperformed other models and LLMs exhibit different competitive levels in eight chemistry tasks. In addition to the key findings from the comprehensive benchmark analysis, our work provides insights into the limitation of current LLMs and the impact of in-context learning settings on LLMs' performance across various chemistry tasks. The code and datasets used in this study are available at https://github.com/ChemFoundationModels/ChemLLMBench.

AI Generated Content (AIGC) has received tremendous attention within the past few years, with content generated in the format of image, text, audio, video, etc. Meanwhile, AIGC has become a double-edged sword and recently received much criticism regarding its responsible usage. In this article, we focus on 8 main concerns that may hinder the healthy development and deployment of AIGC in practice, including risks from (1) privacy; (2) bias, toxicity, misinformation; (3) intellectual property (IP); (4) robustness; (5) open source and explanation; (6) technology abuse; (7) consent, credit, and compensation; (8) environment. Additionally, we provide insights into the promising directions for tackling these risks while constructing generative models, enabling AIGC to be used more responsibly to truly benefit society.

The complex nature of artificial neural networks raises concerns on their reliability, trustworthiness, and fairness in real-world scenarios. The Shapley value -- a solution concept from game theory -- is one of the most popular explanation methods for machine learning models. More traditionally, from a statistical perspective, feature importance is defined in terms of conditional independence. So far, these two approaches to interpretability and feature importance have been considered separate and distinct. In this work, we show that Shapley-based explanation methods and conditional independence testing are closely related. We introduce the SHAPley EXplanation Randomization Test (SHAP-XRT), a testing procedure inspired by the Conditional Randomization Test (CRT) for a specific notion of local (i.e., on a sample) conditional independence. With it, we prove that for binary classification problems, the marginal contributions in the Shapley value provide lower and upper bounds to the expected $p$-values of their respective tests. Furthermore, we show that the Shapley value itself provides an upper bound to the expected $p$-value of a global (i.e., overall) null hypothesis. As a result, we further our understanding of Shapley-based explanation methods from a novel perspective and characterize the conditions under which one can make statistically valid claims about feature importance via the Shapley value.

A Tesla engineer was attacked by a robot during a brutal and bloody malfunction at the company's Giga Texas factory near Austin. Two witnesses watched in horror as their fellow employee was attacked by the machine designed to grab and move freshly cast aluminum car parts. The robot had pinned the man, who was then programming software for two disabled Tesla robots nearby, before sinking its metal claws into the worker's back and arm, leaving a'trail of blood' along the factory surface. The incident - which left the victim with an'open wound' on his left hand - was revealed in a 2021 injury report filed to Travis county and federal regulators, which has been reviewed by DailyMail.com. While no other robot-related injures were reported to regulators by Tesla at the Texas factory in either 2021 or 2022, the incident comes amid years of heightened concerns over the risks of automated robots in the workplace.

We have made it – almost – through another year without being churned into paste by a super-intelligent AI, conscripted into a Martian work camp by an insane billionaire or forced offline by a Carrington event. Even in the absence of civilisation-altering events it's been a busy year. But the advantage of a slow week (I hope that isn't tempting fate) is that you can reflect on the past 12 months and realise that, sometimes, there's only a few stories that really matter. The Guardian has confirmed it was hit by a ransomware attack in December and that the personal data of UK staff members has been accessed in the incident. "We believe this was a criminal ransomware attack, and not the specific targeting of the Guardian as a media organisation," said Guardian Media Group's chief executive, Anna Bateson and the Guardian's editor-in-chief, Katharine Viner.

The prevalent use of commercial and open-source diffusion models (DMs) for text-to-image generation prompts risk mitigation to prevent undesired behaviors. Existing concept erasing methods in academia are all based on full parameter or specification-based fine-tuning, from which we observe the following issues: 1) Generation alternation towards erosion: Parameter drift during target elimination causes alternations and potential deformations across all generations, even eroding other concepts at varying degrees, which is more evident with multi-concept erased; 2) Transfer inability & deployment inefficiency: Previous model-specific erasure impedes the flexible combination of concepts and the training-free transfer towards other models, resulting in linear cost growth as the deployment scenarios increase. To achieve non-invasive, precise, customizable, and transferable elimination, we ground our erasing framework on one-dimensional adapters to erase multiple concepts from most DMs at once across versatile erasing applications. The concept-SemiPermeable structure is injected as a Membrane (SPM) into any DM to learn targeted erasing, and meantime the alteration and erosion phenomenon is effectively mitigated via a novel Latent Anchoring fine-tuning strategy. Once obtained, SPMs can be flexibly combined and plug-and-play for other DMs without specific re-tuning, enabling timely and efficient adaptation to diverse scenarios. During generation, our Facilitated Transport mechanism dynamically regulates the permeability of each SPM to respond to different input prompts, further minimizing the impact on other concepts. Quantitative and qualitative results across ~40 concepts, 7 DMs and 4 erasing applications have demonstrated the superior erasing of SPM. Our code and pre-tuned SPMs will be available on the project page https://lyumengyao.github.io/projects/spm.

Machine Learning has attained high success rates in practically Similarly, zip codes can be correlated with race. Hence, every field, including healthcare, finance, and education, the bias gets embedded in the non-sensitive attributes that based on the accuracy and efficiency of the model's are used in the model training. Based on this hypothesis, a outcome (Dastile, Çelik, and Potsane 2020; Bakator and few initial efforts have been made to mitigate bias in the Radosav 2018). However, these models are biased and exhibit absence of protected attributes (Grari, Lamprier, and Detyniecki a propensity to favor one demographic group over another 2022; Lahoti et al. 2020; Yan, Kao, and Ferrara in various applications, including credit and loan approval, 2020; Zhao et al. 2022). The most recent approach (Zhao criminal justice, and resume-based candidate shortlisting et al. 2022) identifies related features that are correlated with (Mehrabi et al. 2021; Gianfrancesco et al. 2018; Yapo the sensitive attributes and would further minimize the correlation and Weiss 2018). The idea of fairness has received a lot of between the related features and the model's prediction attention recently to combat the discrimination from the outcome to learn a fair classifier with respect to the sensitive of ML models (Dwork et al. 2012; Beutel et al. 2017; attribute. However, identification of related features require Hardt, Price, and Srebro 2016).

In this paper, we aim to align large language models with the ever-changing, complex, and diverse human values (e.g., social norms) across time and locations. This presents a challenge to existing alignment techniques, such as supervised fine-tuning, which internalize values within model parameters. To overcome this, we propose an On-the-fly Preference Optimization (OPO) method, which is a real-time alignment that works in a streaming way. It employs an external memory to store established rules for alignment, which can constrain LLMs' behaviors without further training, allowing for convenient updates and customization of human values. We also introduce a scalable evaluation to assess the proposed method more effectively. Experimental results on both human-annotated and auto-generated questions from legal and moral domains indicate the effectiveness of the proposed OPO method. Our code and data are released at https://github.com/GAIR-NLP/OPO.

The `pre-train, prompt, predict' paradigm of large language models (LLMs) has achieved remarkable success in open-domain question answering (OD-QA). However, few works explore this paradigm in the scenario of multi-document question answering (MD-QA), a task demanding a thorough understanding of the logical associations among the contents and structures of different documents. To fill this crucial gap, we propose a Knowledge Graph Prompting (KGP) method to formulate the right context in prompting LLMs for MD-QA, which consists of a graph construction module and a graph traversal module. For graph construction, we create a knowledge graph (KG) over multiple documents with nodes symbolizing passages or document structures (e.g., pages/tables), and edges denoting the semantic/lexical similarity between passages or intra-document structural relations. For graph traversal, we design an LLM-based graph traversal agent that navigates across nodes and gathers supporting passages assisting LLMs in MD-QA. The constructed graph serves as the global ruler that regulates the transitional space among passages and reduces retrieval latency. Concurrently, the graph traversal agent acts as a local navigator that gathers pertinent context to progressively approach the question and guarantee retrieval quality. Extensive experiments underscore the efficacy of KGP for MD-QA, signifying the potential of leveraging graphs in enhancing the prompt design for LLMs. Our code: https://github.com/YuWVandy/KG-LLM-MDQA.