The remarkable and ever-increasing capabilities of Large Language Models (LLMs) have raised concerns about their potential misuse for creating personalized, convincing misinformation and propaganda. To gain insights into LLMs' persuasive capabilities without directly engaging in experimentation with humans, we propose studying their performance on the related task of detecting convincing arguments. We extend a dataset by Durmus & Cardie (2018) with debates, votes, and user traits and propose tasks measuring LLMs' ability to (1) distinguish between strong and weak arguments, (2) predict stances based on beliefs and demographic characteristics, and (3) determine the appeal of an argument to an individual based on their traits. We show that LLMs perform on par with humans in these tasks and that combining predictions from different LLMs yields significant performance gains, even surpassing human performance. The data and code released with this paper contribute to the crucial ongoing effort of continuously evaluating and monitoring the rapidly evolving capabilities and potential impact of LLMs.

In some applications, edge learning is experiencing a shift in focusing from conventional learning from scratch to new two-stage learning unifying pre-training and task-specific fine-tuning. This paper considers the problem of joint communication and computation resource management in a two-stage edge learning system. In this system, model pre-training is first conducted at an edge server via centralized learning on local pre-stored general data, and then task-specific fine-tuning is performed at edge devices based on the pre-trained model via federated edge learning. For the two-stage learning model, we first analyze the convergence behavior (in terms of the average squared gradient norm bound), which characterizes the impacts of various system parameters such as the number of learning rounds and batch sizes in the two stages on the convergence rate. Based on our analytical results, we then propose a joint communication and computation resource management design to minimize an average squared gradient norm bound, subject to constraints on the transmit power, overall system energy consumption, and training delay. The decision variables include the number of learning rounds, batch sizes, clock frequencies, and transmit power control for both pre-training and fine-tuning stages. Finally, numerical results are provided to evaluate the effectiveness of our proposed design. It is shown that the proposed joint resource management over the pre-training and fine-tuning stages well balances the system performance trade-off among the training accuracy, delay, and energy consumption. The proposed design is also shown to effectively leverage the inherent trade-off between pre-training and fine-tuning, which arises from the differences in data distribution between pre-stored general data versus real-time task-specific data, thus efficiently optimizing overall system performance.

Grounded text generation, encompassing tasks such as long-form question-answering and summarization, necessitates both content selection and content consolidation. Current end-to-end methods are difficult to control and interpret due to their opaqueness. Accordingly, recent works have proposed a modular approach, with separate components for each step. Specifically, we focus on the second subtask, of generating coherent text given pre-selected content in a multi-document setting. Concretely, we formalize Fusion-in-Context (FiC) as a standalone task, whose input consists of source texts with highlighted spans of targeted content. A model then needs to generate a coherent passage that includes all and only the target information. Our work includes the development of a curated dataset of 1000 instances in the reviews domain, alongside a novel evaluation framework for assessing the faithfulness and coverage of highlights, which strongly correlate to human judgment. Several baseline models exhibit promising outcomes and provide insightful analyses. This study lays the groundwork for further exploration of modular text generation in the multi-document setting, offering potential improvements in the quality and reliability of generated content. Our benchmark, FuseReviews, including the dataset, evaluation framework, and designated leaderboard, can be found at https://fusereviews.github.io/.

Single document news summarization has seen substantial progress on faithfulness in recent years, driven by research on the evaluation of factual consistency, or hallucinations. We ask whether these advances carry over to other text summarization domains. We propose a new evaluation benchmark on topic-focused dialogue summarization, generated by LLMs of varying sizes. We provide binary sentence-level human annotations of the factual consistency of these summaries along with detailed explanations of factually inconsistent sentences. Our analysis shows that existing LLMs hallucinate significant amounts of factual errors in the dialogue domain, regardless of the model's size. On the other hand, when LLMs, including GPT-4, serve as binary factual evaluators, they perform poorly and can be outperformed by prevailing state-of-the-art specialized factuality evaluation metrics. Finally, we conducted an analysis of hallucination types with a curated error taxonomy. We find that there are diverse errors and error distributions in model-generated summaries and that non-LLM based metrics can capture all error types better than LLM-based evaluators.

Large language models (LLMs) have become the norm in natural language processing (NLP), excelling in few-shot in-context learning (ICL) with their remarkable abilities. Nonetheless, the success of ICL largely hinges on the choice of few-shot demonstration examples, making the selection process increasingly crucial. Existing methods have delved into optimizing the quantity and semantic similarity of these examples to improve ICL performances. However, our preliminary experiments indicate that the effectiveness of ICL is limited by the length of the input context. Moreover, varying combinations of few-shot demonstration examples can significantly boost accuracy across different test samples. To address this, we propose a novel method named parallel in-context learning (ParaICL) that effectively utilizes all demonstration examples without exceeding the manageable input context length. ParaICL employs parallel batching to distribute demonstration examples into different batches according to the semantic similarities of the questions in the demonstrations to the test question. It then computes normalized batch semantic scores for each batch. A weighted average semantic objective, constrained by adaptive plausibility, is applied to select the most appropriate tokens. Through extensive experiments, we validate the effectiveness of ParaICL and conduct ablation studies to underscore its design rationale. We further demonstrate that ParaICL can seamlessly integrate with existing methods.

Text-to-image (T2I) generative models have attracted significant attention and found extensive applications within and beyond academic research. For example, the Civitai community, a platform for T2I innovation, currently hosts an impressive array of 74,492 distinct models. However, this diversity presents a formidable challenge in selecting the most appropriate model and parameters, a process that typically requires numerous trials. Drawing inspiration from the tool usage research of large language models (LLMs), we introduce DiffAgent, an LLM agent designed to screen the accurate selection in seconds via API calls. DiffAgent leverages a novel two-stage training framework, SFTA, enabling it to accurately align T2I API responses with user input in accordance with human preferences. To train and evaluate DiffAgent's capabilities, we present DABench, a comprehensive dataset encompassing an extensive range of T2I APIs from the community. Our evaluations reveal that DiffAgent not only excels in identifying the appropriate T2I API but also underscores the effectiveness of the SFTA training framework. Codes are available at https://github.com/OpenGVLab/DiffAgent.

Human writers often bookend their writing with ending sentences that relate back to the beginning sentences in order to compose a satisfying narrative that "closes the loop." Motivated by this observation, we propose RENarGen, a controllable story-generation paradigm that generates narratives by ensuring the first and last sentences are related and then infilling the middle sentences. Our contributions include an initial exploration of how various methods of bookending from Narratology affect language modeling for stories. Automatic and human evaluations indicate RENarGen produces better stories with more narrative closure than current autoregressive models.

Large Language Models (LLMs) exhibit remarkable capabilities but are prone to generating inaccurate or hallucinatory responses. This limitation stems from their reliance on vast pretraining datasets, making them susceptible to errors in unseen scenarios. To tackle these challenges, Retrieval-Augmented Generation (RAG) addresses this by incorporating external, relevant documents into the response generation process, thus leveraging non-parametric knowledge alongside LLMs' in-context learning abilities. However, existing RAG implementations primarily focus on initial input for context retrieval, overlooking the nuances of ambiguous or complex queries that necessitate further clarification or decomposition for accurate responses. To this end, we propose learning to Refine Query for Retrieval Augmented Generation (RQ-RAG) in this paper, endeavoring to enhance the model by equipping it with capabilities for explicit rewriting, decomposition, and disambiguation. Our experimental results indicate that our method, when applied to a 7B Llama2 model, surpasses the previous state-of-the-art (SOTA) by an average of 1.9\% across three single-hop QA datasets, and also demonstrates enhanced performance in handling complex, multi-hop QA datasets. Our code is available at https://github.com/chanchimin/RQ-RAG.

How to evaluate Large Language Models (LLMs) in code generation is an open question. Existing benchmarks demonstrate poor alignment with real-world code repositories and are insufficient to evaluate the coding abilities of LLMs. This paper proposes a new benchmark - EvoCodeBench to address the preceding problems, which has three primary advances. (1) EvoCodeBench aligns with real-world repositories in multiple dimensions, e.g., code distributions and dependency distributions. (2) EvoCodeBench offers comprehensive annotations (e.g., requirements, reference code, and reference dependencies), and robust evaluation metrics (e.g., Pass@k and Recall@k). (3) EvoCodeBench is an evolving benchmark to avoid data leakage. We build an automatic pipeline to update EvoCodeBench from the latest repositories. We release the first version - EvoCodeBench-2403, containing 275 samples from 25 real-world repositories. Based on EvoCodeBench, we propose repository-level code generation and evaluate 10 popular LLMs (e.g., gpt-4, gpt-3.5, DeepSeek Coder, StarCoder 2, CodeLLaMa, Gemma, and Qwen 1.5). Our experiments reveal the coding abilities of these LLMs in real-world repositories. For example, the highest Pass@1 of gpt-4 only is 20.73% in our experiments. We also analyze failed cases and summarize the shortcomings of existing LLMs in EvoCodeBench. We release EvoCodeBench, all prompts, and LLMs' completions for further community analysis.

Generative models are rapidly gaining popularity and being integrated into everyday applications, raising concerns over their safety issues as various vulnerabilities are exposed. Faced with the problem, the field of red teaming is experiencing fast-paced growth, which highlights the need for a comprehensive organization covering the entire pipeline and addressing emerging topics for the community. Our extensive survey, which examines over 120 papers, introduces a taxonomy of fine-grained attack strategies grounded in the inherent capabilities of language models. Additionally, we have developed the searcher framework that unifies various automatic red teaming approaches. Moreover, our survey covers novel areas including multimodal attacks and defenses, risks around multilingual models, overkill of harmless queries, and safety of downstream applications. We hope this survey can provide a systematic perspective on the field and unlock new areas of research. Warning: This paper contains examples that may be offensive, harmful, or biased.