Large language models (LLMs) face issues in handling factual knowledge, making it vital to evaluate their true ability to understand facts. In this study, we introduce knowledge probing frameworks, BELIEF(-ICL), to evaluate the knowledge understanding ability of not only encoder-based PLMs but also decoder-based PLMs from diverse perspectives. BELIEFs utilize a multi-prompt dataset to evaluate PLM's accuracy, consistency, and reliability in factual knowledge understanding. To provide a more reliable evaluation with BELIEFs, we semi-automatically create MyriadLAMA, which has more diverse prompts than existing datasets. We validate the effectiveness of BELIEFs in correctly and comprehensively evaluating PLM's factual understanding ability through extensive evaluations. We further investigate key factors in learning facts in LLMs, and reveal the limitation of the prompt-based knowledge probing. The dataset is anonymously publicized.

Instruction-following large language models (LLMs), such as ChatGPT, have become increasingly popular with the general audience, many of whom are incorporating them into their daily routines. However, these LLMs inadvertently disclose personal or copyrighted information, which calls for a machine unlearning method to remove selective knowledge. Previous attempts sought to forget the link between the target information and its associated entities, but it rather led to generating undesirable responses about the target, compromising the end-user experience. In this work, we propose SNAP, an innovative framework designed to selectively unlearn information by 1) training an LLM with negative instructions to generate obliterated responses, 2) augmenting hard positives to retain the original LLM performance, and 3) applying the novel Wasserstein regularization to ensure adequate deviation from the initial weights of the LLM. We evaluate our framework on various NLP benchmarks and demonstrate that our approach retains the original LLM capabilities, while successfully unlearning the specified information.

The probabilistic interpretation of Canonical Correlation Analysis (CCA) for learning low-dimensional real vectors, called as latent variables, has been exploited immensely in various fields. This study takes a step further by demonstrating the potential of CCA in discovering a latent state that captures the contextual information within the textual data under a two-view setting. The interpretation of CCA discussed in this study utilizes the multi-view nature of textual data, i.e. the consecutive sentences in a document or turns in a dyadic conversation, and has a strong theoretical foundation. Furthermore, this study proposes a model using CCA to perform the Automatic Short Answer Grading (ASAG) task. The empirical analysis confirms that the proposed model delivers competitive results and can even beat various sophisticated supervised techniques. The model is simple, linear, and adaptable and should be used as the baseline especially when labeled training data is scarce or nonexistent.

The rapid development of Large Language Models (LLMs) in vertical domains, including intellectual property (IP), lacks a specific evaluation benchmark for assessing their understanding, application, and reasoning abilities. To fill this gap, we introduce IPEval, the first evaluation benchmark tailored for IP agency and consulting tasks. IPEval comprises 2657 multiple-choice questions across four major dimensions: creation, application, protection, and management of IP. These questions span patent rights (inventions, utility models, designs), trademarks, copyrights, trade secrets, and other related laws. Evaluation methods include zero-shot, 5-few-shot, and Chain of Thought (CoT) for seven LLM types, predominantly in English or Chinese. Results show superior English performance by models like GPT series and Qwen series, while Chinese-centric LLMs excel in Chinese tests, albeit specialized IP LLMs lag behind general-purpose ones. Regional and temporal aspects of IP underscore the need for LLMs to grasp legal nuances and evolving laws. IPEval aims to accurately gauge LLM capabilities in IP and spur development of specialized models. Website: \url{https://ipeval.github.io/}

With the increasing role of Natural Language Processing (NLP) in various applications, challenges concerning bias and stereotype perpetuation are accentuated, which often leads to hate speech and harm. Despite existing studies on sexism and misogyny, issues like homophobia and transphobia remain underexplored and often adopt binary perspectives, putting the safety of LGBTQIA+ individuals at high risk in online spaces. In this paper, we assess the potential harm caused by sentence completions generated by English large language models (LLMs) concerning LGBTQIA+ individuals. This is achieved using QueerBench, our new assessment framework, which employs a template-based approach and a Masked Language Modeling (MLM) task. The analysis indicates that large language models tend to exhibit discriminatory behaviour more frequently towards individuals within the LGBTQIA+ community, reaching a difference gap of 7.2% in the QueerBench score of harmfulness.

Despite their remarkable successes, state-of-the-art large language models (LLMs), including vision-and-language models (VLMs) and unimodal language models (ULMs), fail to understand precise semantics. For example, semantically equivalent sentences expressed using different lexical compositions elicit diverging representations. The degree of this divergence and its impact on encoded semantics is not very well understood. In this paper, we introduce the SUGARCREPE++ dataset to analyze the sensitivity of VLMs and ULMs to lexical and semantic alterations. Each sample in SUGARCREPE++ dataset consists of an image and a corresponding triplet of captions: a pair of semantically equivalent but lexically different positive captions and one hard negative caption. This poses a 3-way semantic (in)equivalence problem to the language models. We comprehensively evaluate VLMs and ULMs that differ in architecture, pre-training objectives and datasets to benchmark the performance of SUGARCREPE++ dataset. Experimental results highlight the difficulties of VLMs in distinguishing between lexical and semantic variations, particularly in object attributes and spatial relations. Although VLMs with larger pre-training datasets, model sizes, and multiple pre-training objectives achieve better performance on SUGARCREPE++, there is a significant opportunity for improvement. We show that all the models which achieve better performance on compositionality datasets need not perform equally well on SUGARCREPE++, signifying that compositionality alone may not be sufficient for understanding semantic and lexical alterations. Given the importance of the property that the SUGARCREPE++ dataset targets, it serves as a new challenge to the vision-and-language community.

Bias in LLMs can harm user experience and societal outcomes. However, current bias mitigation methods often require intensive human feedback, lack transferability to other topics or yield overconfident and random outputs. We find that involving LLMs in roleplaying scenario boosts their ability to recognize and mitigate biases. Based on this, we propose Reinforcement Learning from Multirole Debates as Feedback (RLDF), a novel approach for bias mitigation replacing human feedback in traditional RLHF. We utilize Figure 1: Asking GPT-3.5-turbo and GPT-2 about the LLMs in multi-role debates to create a bias in the text it generates using the prompt "Here dataset that includes both high-bias and lowbias is our Q&A ","Here is the Q&A between me and a instances for training the reward model language model" and "Here is the Q&A between me in reinforcement learning. Our approach comprises and a language model competing with you", the number two modes: (1) self-reflection, where of identified biases increases gradually. When informed the same LLM participates in multi-role debates, that the content was generated by itself, the LLM admits and (2) teacher-student, where a more to far fewer biased responses than with other prompts.

Large Language Models (LLMs) are increasingly being used in educational and learning applications. Research has demonstrated that controlling for style, to fit the needs of the learner, fosters increased understanding, promotes inclusion, and helps with knowledge distillation. To understand the capabilities and limitations of contemporary LLMs in style control, we evaluated five state-of-the-art models: GPT-3.5, GPT-4, GPT-4o, Llama-3, and Mistral-instruct-7B across two style control tasks. We observed significant inconsistencies in the first task, with model performances averaging between 5th and 8th grade reading levels for tasks intended for first-graders, and standard deviations up to 27.6. For our second task, we observed a statistically significant improvement in performance from 0.02 to 0.26. However, we find that even without stereotypes in reference texts, LLMs Figure 1: Overall view of this paper. We find that while often generated culturally insensitive content in-context learning can control for reading level and during their tasks. We provide a thorough analysis simplicity, it cannot do the same for vernacular English.

A dog whistle is a form of coded communication that carries a secondary meaning to specific audiences and is often weaponized for racial and socioeconomic discrimination. Dog whistling historically originated from United States politics, but in recent years has taken root in social media as a means of evading hate speech detection systems and maintaining plausible deniability. In this paper, we present an approach for word-sense disambiguation of dog whistles from standard speech using Large Language Models (LLMs), and leverage this technique to create a dataset of 16,550 high-confidence coded examples of dog whistles used in formal and informal communication. Silent Signals is the largest dataset of disambiguated dog whistle usage, created for applications in hate speech detection, neology, and political science. The dataset can be found at https://huggingface.co/datasets/SALT-NLP/silent_signals.

AI is becoming increasingly common across different domains. However, as sophisticated AI-based systems are often blackboxed, rendering the decision-making logic opaque, users find it challenging to comply with their recommendations. Although researchers are investigating Explainable AI (XAI) to increase the transparency of the underlying machine learning models, it is unclear what types of explanations are effective and what other factors increase compliance. To better understand the interplay of these factors, we conducted an experiment with 562 participants, who were presented with the recommendations of an AI and two different types of XAI. We find that users' compliance increases with the introduction of XAI but is also affected by AI literacy. We also find that the relationships between AI literacy, XAI, and users' compliance are mediated by the users' mental model of AI. Our study has several implications for successfully designing AI-based systems utilizing XAI.