This paper presents a dataset for closest opposite questions in Arabic language. The dataset is the first of its kind for the Arabic language. It is beneficial for the assessment of systems on the aspect of antonymy detection. The structure is similar to that of the Graduate Record Examination (GRE) closest opposite questions dataset for the English language. The introduced dataset consists of 500 questions, each contains a query word for which the closest opposite needs to be determined from among a set of candidate words. Each question is also associated with the correct answer. We publish the dataset publicly in addition to providing standard splits of the dataset into development and test sets. Moreover, the paper provides a benchmark for the performance of different Arabic word embedding models on the introduced dataset.

Text sanitization is the task of redacting a document to mask all occurrences of (direct or indirect) personal identifiers, with the goal of concealing the identity of the individual(s) referred in it. In this paper, we consider a two-step approach to text sanitization and provide a detailed analysis of its empirical performance on two recently published datasets: the Text Anonymization Benchmark (Pil\'an et al., 2022) and a collection of Wikipedia biographies (Papadopoulou et al., 2022). The text sanitization process starts with a privacy-oriented entity recognizer that seeks to determine the text spans expressing identifiable personal information. This privacy-oriented entity recognizer is trained by combining a standard named entity recognition model with a gazetteer populated by person-related terms extracted from Wikidata. The second step of the text sanitization process consists in assessing the privacy risk associated with each detected text span, either isolated or in combination with other text spans. We present five distinct indicators of the re-identification risk, respectively based on language model probabilities, text span classification, sequence labelling, perturbations, and web search. We provide a contrastive analysis of each privacy indicator and highlight their benefits and limitations, notably in relation to the available labeled data.

This paper addresses the problem of generating questions from a given context and an answer, specifically focusing on questions that require multi-hop reasoning across an extended context. Previous studies have suggested that key phrase selection is essential for question generation (QG), yet it is still challenging to connect such disjointed phrases into meaningful questions, particularly for long context. To mitigate this issue, we propose MultiFactor, a novel QG framework based on multi-level content planning. Specifically, MultiFactor includes two components: FA-model, which simultaneously selects key phrases and generates full answers, and Q-model which takes the generated full answer as an additional input to generate questions. Here, full answer generation is introduced to connect the short answer with the selected key phrases, thus forming an answer-aware summary to facilitate QG. Both FA-model and Q-model are formalized as simple-yet-effective Phrase-Enhanced Transformers, our joint model for phrase selection and text generation. Experimental results show that our method outperforms strong baselines on two popular QG datasets. Our code is available at https://github.com/zeaver/MultiFactor.

Open large language models (LLMs) with great performance in various tasks have significantly advanced the development of LLMs. However, they are far inferior to commercial models such as ChatGPT and GPT-4 when acting as agents to tackle complex tasks in the real world. These agent tasks employ LLMs as the central controller responsible for planning, memorization, and tool utilization, necessitating both fine-grained prompting methods and robust LLMs to achieve satisfactory performance. Though many prompting methods have been proposed to complete particular agent tasks, there is lack of research focusing on improving the agent capabilities of LLMs themselves without compromising their general abilities. In this work, we present AgentTuning, a simple and general method to enhance the agent abilities of LLMs while maintaining their general LLM capabilities. We construct AgentInstruct, a lightweight instruction-tuning dataset containing high-quality interaction trajectories. We employ a hybrid instruction-tuning strategy by combining AgentInstruct with open-source instructions from general domains. AgentTuning is used to instruction-tune the Llama 2 series, resulting in AgentLM. Our evaluations show that AgentTuning enables LLMs' agent capabilities without compromising general abilities. The AgentLM-70B is comparable to GPT-3.5-turbo on unseen agent tasks, demonstrating generalized agent capabilities. We open source the AgentInstruct and AgentLM-7B, 13B, and 70B models at https://github.com/THUDM/AgentTuning, serving open and powerful alternatives to commercial LLMs for agent tasks.

In many real-world scenarios (e.g., academic networks, social platforms), different types of entities are not only associated with texts but also connected by various relationships, which can be abstracted as Text-Attributed Heterogeneous Graphs (TAHGs). Current pretraining tasks for Language Models (LMs) primarily focus on separately learning the textual information of each entity and overlook the crucial aspect of capturing topological connections among entities in TAHGs. In this paper, we present a new pretraining framework for LMs that explicitly considers the topological and heterogeneous information in TAHGs. Firstly, we define a context graph as neighborhoods of a target node within specific orders and propose a topology-aware pretraining task to predict nodes involved in the context graph by jointly optimizing an LM and an auxiliary heterogeneous graph neural network. Secondly, based on the observation that some nodes are text-rich while others have little text, we devise a text augmentation strategy to enrich textless nodes with their neighbors' texts for handling the imbalance issue. We conduct link prediction and node classification tasks on three datasets from various domains. Experimental results demonstrate the superiority of our approach over existing methods and the rationality of each design. Our code is available at https://github.com/Hope-Rita/THLM.

A common practice in knowledge-grounded dialogue generation is to explicitly utilize intermediate steps (e.g., web-search, memory retrieval) with modular approaches. However, data for such steps are often inaccessible compared to those of dialogue responses as they are unobservable in an ordinary dialogue. To fill in the absence of these data, we develop a self-improving method to improve the generative performances of intermediate steps without the ground truth data. In particular, we propose a novel bootstrapping scheme with a guided prompt and a modified loss function to enhance the diversity of appropriate self-generated responses. Through experiments on various benchmark datasets, we empirically demonstrate that our method successfully leverages a self-improving mechanism in generating intermediate and final responses and improves the performances on the task of knowledge-grounded dialogue generation. Along with the progress of Language Model (LM) pretraining, open-domain dialogue models have evolved to leverage the advantage of the transformer architecture's generalization ability (Zhang et al., 2019; Freitas et al., 2020; Roller et al., 2021; Xu et al., 2022a; Shuster et al., 2022b; Thoppilan et al., 2022). While model scaling also improves the dialogue quality (Freitas et al., 2020) as seen in large LMs, relying on sole LMs casts limitations such as hallucination and the lack of faithfulness by outdated training data (Brown et al., 2020; Thoppilan et al., 2022; Chowdhery et al., 2022). In order to overcome the limitations, prior works have adopted a modular design where multiple modules generate intermediate texts (e.g., to retrieve documents) before the final response (Lewis et al., 2020; Adolphs et al., 2021; Zhang et al., 2021; Shuster et al., 2022a). Among them, Komeili et al. (2022); Shuster et al. (2022b) have shown promising results in dialogue generation. Specifically, they adopted a modular design to integrate external knowledge (e.g., internet) and internal knowledge (e.g., memory) in dialogue models. For example, in Komeili et al. (2022), a LM first decides whether to access a knowledge in a form of text generation. Upon deciding to access knowledge, the LM generates an appropriate query for knowledge retrieval from external sources such as search engines. Then, the LM generates a response based on extracted knowledge from the accessed data. See Figure 2 of Appendix A for an illustrative example. Regarding each intermediate phase as a separate module, a convenient method of training these modules would be to apply supervised learning on each module using individual datasets (Dinan et al., 2019; Shuster et al., 2022a; Glass et al., 2022; Shuster et al., 2022b).

Pretrained language models have achieved remarkable success in natural language understanding. However, fine-tuning pretrained models on limited training data tends to overfit and thus diminish performance. This paper presents Bi-Drop, a fine-tuning strategy that selectively updates model parameters using gradients from various sub-nets dynamically generated by dropout. The sub-net estimation of Bi-Drop is performed in an in-batch manner, so it overcomes the problem of hysteresis in sub-net updating, which is possessed by previous methods that perform asynchronous sub-net estimation. Also, Bi-Drop needs only one mini-batch to estimate the sub-net so it achieves higher utility of training data. Experiments on the GLUE benchmark demonstrate that Bi-Drop consistently outperforms previous fine-tuning methods. Furthermore, empirical results also show that Bi-Drop exhibits excellent generalization ability and robustness for domain transfer, data imbalance, and low-resource scenarios.

Topic models are used to make sense of large text collections. However, automatically evaluating topic model output and determining the optimal number of topics both have been longstanding challenges, with no effective automated solutions to date. This paper proposes using large language models to evaluate such output. We find that large language models appropriately assess the resulting topics, correlating more strongly with human judgments than existing automated metrics. We then investigate whether we can use large language models to automatically determine the optimal number of topics. We automatically assign labels to documents and choosing configurations with the most pure labels returns reasonable values for the optimal number of topics.

Large language models (LLMs), such as ChatGPT, are prone to generate hallucinations, i.e., content that conflicts with the source or cannot be verified by the factual knowledge. To understand what types of content and to which extent LLMs are apt to hallucinate, we introduce the Hallucination Evaluation benchmark for Large Language Models (HaluEval), a large collection of generated and human-annotated hallucinated samples for evaluating the performance of LLMs in recognizing hallucination. To generate these samples, we propose a ChatGPT-based two-step framework, i.e., sampling-then-filtering. Besides, we also hire some human labelers to annotate the hallucinations in ChatGPT responses. The empirical results suggest that ChatGPT is likely to generate hallucinated content in specific topics by fabricating unverifiable information (i.e., about $19.5\%$ responses). Moreover, existing LLMs face great challenges in recognizing the hallucinations in texts. However, our experiments also prove that providing external knowledge or adding reasoning steps can help LLMs recognize hallucinations. Our benchmark can be accessed at https://github.com/RUCAIBox/HaluEval.

Generative AI models have shown impressive performance on many Natural Language Processing tasks such as language understanding, reasoning, and language generation. An important question being asked by the AI community today is about the capabilities and limits of these models, and it is clear that evaluating generative AI is very challenging. Most studies on generative LLMs have been restricted to English and it is unclear how capable these models are at understanding and generating text in other languages. We present the first comprehensive benchmarking of generative LLMs - MEGA, which evaluates models on standard NLP benchmarks, covering 16 NLP datasets across 70 typologically diverse languages. We compare the performance of generative LLMs including Chat-GPT and GPT-4 to State of the Art (SOTA) non-autoregressive models on these tasks to determine how well generative models perform compared to the previous generation of LLMs. We present a thorough analysis of the performance of models across languages and tasks and discuss challenges in improving the performance of generative LLMs on low-resource languages. We create a framework for evaluating generative LLMs in the multilingual setting and provide directions for future progress in the field.