This work presents a detailed case study on using Generative AI (GenAI) to develop AI surrogates for simulation models in fusion energy research. The scope includes the methodology, implementation, and results of using GenAI to assist in model development and optimization, comparing these results with previous manually developed models.

We propose an approach to referring expression generation (REG) in visually grounded dialogue that is meant to produce referring expressions (REs) that are both discriminative and discourse-appropriate. Our method constitutes a two-stage process. First, we model REG as a text- and image-conditioned next-token prediction task. REs are autoregressively generated based on their preceding linguistic context and a visual representation of the referent. Second, we propose the use of discourse-aware comprehension guiding as part of a generate-and-rerank strategy through which candidate REs generated with our REG model are reranked based on their discourse-dependent discriminatory power. Results from our human evaluation indicate that our proposed two-stage approach is effective in producing discriminative REs, with higher performance in terms of text-image retrieval accuracy for reranked REs compared to those generated using greedy decoding.

Quality pretraining data is often seen as the key to high-performance language models. However, progress in understanding pretraining data has been slow due to the costly pretraining runs required for data selection experiments. We present a framework that avoids these costs and selects high-quality pretraining data without any LLM training of our own. Our work is based on a simple observation: LLM losses on many pretraining texts are correlated with downstream benchmark performance, and selecting high-correlation documents is an effective pretraining data selection method. We build a new statistical framework for data selection centered around estimates of perplexity-benchmark correlations and perform data selection using a sample of 90 LLMs taken from the Open LLM Leaderboard on texts from tens of thousands of web domains. In controlled pretraining experiments at the 160M parameter scale on 8 benchmarks, our approach outperforms DSIR on every benchmark, while matching the best data selector found in DataComp-LM, a hand-engineered bigram classifier. Dataset curation is increasingly crucial for training high-quality large language models (LLMs). However, data-driven approaches to data selection typically involve expensive model retraining steps that limit their effectiveness, and no algorithm has been reported to consistently beat or match hand-crafted classifiers for data selection (Li et al., 2024). Is training new LLMs necessary for data selection? Instead of training our own models, can we use the growing collection of publicly available, high-performance LLMs (Wolf et al., 2019; Beeching et al., 2023) to perform data valuation and selection? This would have significant benefits: we could leverage the millions of dollars collectively spent on building these LLMs, and we would have coverage over a large, heterogeneous collection of high-performance models varying in size, architectures, and pretraining data distribution. Despite these advantages, using existing models for pretraining data selection is challenging, as the training data for these models are often unknown and heterogeneous.

Human-like large language models (LLMs), especially the most powerful and popular ones in OpenAI's GPT family, have proven to be very helpful for many natural language processing (NLP) related tasks. Therefore, various attempts have been made to apply LLMs to information extraction (IE), which is a fundamental NLP task that involves extracting information from unstructured plain text. To demonstrate the latest representative progress in LLMs' information extraction ability, we assess the information extraction ability of GPT-4 (the latest version of GPT at the time of writing this paper) from four perspectives: Performance, Evaluation Criteria, Robustness, and Error Types. Our results suggest a visible performance gap between GPT-4 and state-of-the-art (SOTA) IE methods. To alleviate this problem, considering the LLMs' human-like characteristics, we propose and analyze the effects of a series of simple prompt-based methods, which can be generalized to other LLMs and NLP tasks. Rich experiments show our methods' effectiveness and some of their remaining issues in improving GPT-4's information extraction ability.

Large language models (LLMs) have demonstrated remarkable capabilities in natural language processing tasks. However, their practical application in high-stake domains, such as fraud and abuse detection, remains an area that requires further exploration. The existing applications often narrowly focus on specific tasks like toxicity or hate speech detection. In this paper, we present a comprehensive benchmark suite designed to assess the performance of LLMs in identifying and mitigating fraudulent and abusive language across various real-world scenarios. Our benchmark encompasses a diverse set of tasks, including detecting spam emails, hate speech, misogynistic language, and more. We evaluated several state-of-the-art LLMs, including models from Anthropic, Mistral AI, and the AI21 family, to provide a comprehensive assessment of their capabilities in this critical domain. The results indicate that while LLMs exhibit proficient baseline performance in individual fraud and abuse detection tasks, their performance varies considerably across tasks, particularly struggling with tasks that demand nuanced pragmatic reasoning, such as identifying diverse forms of misogynistic language. These findings have important implications for the responsible development and deployment of LLMs in high-risk applications. Our benchmark suite can serve as a tool for researchers and practitioners to systematically evaluate LLMs for multi-task fraud detection and drive the creation of more robust, trustworthy, and ethically-aligned systems for fraud and abuse detection.

Recent advancements have significantly improved automated task-solving capabilities using autonomous agents powered by large language models (LLMs). However, most LLM-based agents focus on dialogue, programming, or specialized domains, leaving their potential for addressing generative AI safety tasks largely unexplored. In this paper, we propose Atlas, an advanced LLM-based multi-agent framework targeting generative AI models, specifically focusing on jailbreak attacks against text-to-image (T2I) models with built-in safety filters. Atlas consists of two agents, namely the mutation agent and the selection agent, each comprising four key modules: a vision-language model (VLM) or LLM brain, planning, memory, and tool usage. The mutation agent uses its VLM brain to determine whether a prompt triggers the T2I model's safety filter. It then collaborates iteratively with the LLM brain of the selection agent to generate new candidate jailbreak prompts with the highest potential to bypass the filter. In addition to multi-agent communication, we leverage in-context learning (ICL) memory mechanisms and the chain-of-thought (COT) approach to learn from past successes and failures, thereby enhancing Atlas's performance. Our evaluation demonstrates that Atlas successfully jailbreaks several state-of-the-art T2I models equipped with multi-modal safety filters in a black-box setting. Additionally, Atlas outperforms existing methods in both query efficiency and the quality of generated images. This work convincingly demonstrates the successful application of LLM-based agents in studying the safety vulnerabilities of popular text-to-image generation models. We urge the community to consider advanced techniques like ours in response to the rapidly evolving text-to-image generation field.

It is widely acknowledged that extracting market sentiments from news data benefits market predictions. However, existing methods of using financial sentiments remain simplistic, relying on equal-weight and static aggregation to manage sentiments from multiple news items. This leads to a critical issue termed ``Aggregated Sentiment Homogenization'', which has been explored through our analysis of a large financial news dataset from industry practice. This phenomenon occurs when aggregating numerous sentiments, causing representations to converge towards the mean values of sentiment distributions and thereby smoothing out unique and important information. Consequently, the aggregated sentiment representations lose much predictive value of news data. To address this problem, we introduce the Market Attention-weighted News Aggregation Network (MANA-Net), a novel method that leverages a dynamic market-news attention mechanism to aggregate news sentiments for market prediction. MANA-Net learns the relevance of news sentiments to price changes and assigns varying weights to individual news items. By integrating the news aggregation step into the networks for market prediction, MANA-Net allows for trainable sentiment representations that are optimized directly for prediction. We evaluate MANA-Net using the S&P 500 and NASDAQ 100 indices, along with financial news spanning from 2003 to 2018. Experimental results demonstrate that MANA-Net outperforms various recent market prediction methods, enhancing Profit & Loss by 1.1% and the daily Sharpe ratio by 0.252.

Federated learning (FL), as Previous studies on federated learning (FL) often encounter performance introduced by [31], offers a solution to these privacy challenges degradation due to data heterogeneity among different by enabling collaborative model training across numerous clients clients. In light of the recent advances in multimodal large language under the orchestration of a central server, without sharing the models (MLLMs), such as GPT-4v and LLaVA, which demonstrate raw data. FL systems bring obvious advantages by involving clients their exceptional proficiency in multimodal tasks, such as downloading a global model, performing local updates using their image captioning and multimodal question answering. We introduce data, and then sending these updates back to the server. The server a novel federated learning framework, named Multimodal aggregates these updates to enhance the global model, thereby Large Language Model Assisted Federated Learning (MLLM-FL), preserving data privacy. Aside from the privacy considerations which which employs powerful MLLMs at the server end to address mentioned above, there are two fundamental acknowledgements the heterogeneous and long-tailed challenges. Owing to the about (cross-device) federated learning which have been widely advanced cross-modality representation capabilities and the extensive recognized: data heterogeneity among clients and the limited and open-vocabulary prior knowledge of MLLMs, our framework diverse computational resources on local devices [18, 30, 31, 52]. is adept at harnessing the extensive, yet previously underexploited, Data heterogeneity represents a significant challenge in federated open-source data accessible from websites and powerful server-side learning. It largely stems from the fact that the data across computational resources. Hence, the MLLM-FL not only enhances participating clients are distributed independently, with each client the performance but also avoids increasing the risk of privacy leakage having a different sample distribution.

Large Language Models (LLMs) are becoming very popular and are used for many different purposes, including creative tasks in the arts. However, these models sometimes have trouble with specific reasoning tasks, especially those that involve logical thinking and counting. This paper looks at how well LLMs understand and reason when dealing with musical tasks like figuring out notes from intervals and identifying chords and scales. We tested GPT-3.5 and GPT-4o to see how they handle these tasks. Our results show that while LLMs do well with note intervals, they struggle with more complicated tasks like recognizing chords and scales. This points out clear limits in current LLM abilities and shows where we need to make them better, which could help improve how they think and work in both artistic and other complex areas. We also provide an automatically generated benchmark data set for the described tasks.

While Large Language Models (LLMs) exhibit remarkable generative capabilities, they are not without flaws, particularly in the form of hallucinations. This issue is even more pronounced when LLMs are applied to specific languages and domains. For example, LLMs may generate nonsense information when handling Chinese ancient poetry, proverbs, or idioms, owing to the lack of specific knowledge. To this end, this paper introduces a benchmark for rectifying Chinese knowledge in LLMs via knowledge editing. Specifically, we introduce a new Chinese dataset, CKnowEdit, by collecting seven type of knowledge from various sources, including classical texts, idioms, and content from Baidu Tieba Ruozhiba, thereby accounting for the unique polyphony, antithesis, and logical constructs inherent in the Chinese language. Through the analysis of this dataset, we uncover the challenges faced by current LLMs in mastering Chinese. Furthermore, our evaluation of state-of-the-art knowledge editing techniques on this dataset unveil the substantial scope for advancement in the rectification of Chinese knowledge. Code and dataset are available at https://github.com/zjunlp/EasyEdit.