Computation and Language (cs.CL)

Abstract: Recently, the development and progress of Large Language Models (LLMs) have amazed the entire Artificial Intelligence community. As an outstanding representative of LLMs and the foundation model that set off this wave of research on LLMs, ChatGPT has attracted more and more researchers to study its capabilities and performance on various downstream Natural Language Processing (NLP) tasks. While marveling at ChatGPT's incredible performance on kinds of tasks, we notice that ChatGPT also has excellent multilingual processing capabilities, such as Chinese. To explore the Chinese processing ability of ChatGPT, we focus on Chinese Text Correction, a fundamental and challenging Chinese NLP task. Specifically, we evaluate ChatGPT on the Chinese Grammatical Error Correction (CGEC) and Chinese Spelling Check (CSC) tasks, which are two main Chinese Text Correction scenarios. From extensive analyses and comparisons with previous state-of-the-art fine-tuned models, we empirically find that the ChatGPT currently has both amazing performance and unsatisfactory behavior for Chinese Text Correction. We believe our findings will promote the landing and application of LLMs in the Chinese NLP community.

Abstract: GPT-3.5 and GPT-4 are the two most widely used large language model (LLM) services. However, when and how these models are updated over time is opaque. Here, we evaluate the March 2023 and June 2023 versions of GPT-3.5 and GPT-4 on four diverse tasks: 1) solving math problems, 2) answering sensitive/dangerous questions, 3) generating code and 4) visual reasoning. We find that the performance and behavior of both GPT-3.5 and GPT-4 can vary greatly over time. For example, GPT-4 (March 2023) was very good at identifying prime numbers (accuracy 97.6%) but GPT-4 (June 2023) was very poor on these same questions (accuracy 2.4%). Interestingly GPT-3.5 (June 2023) was much better than GPT-3.5 (March 2023) in this task. GPT-4 was less willing to answer sensitive questions in June than in March, and both GPT-4 and GPT-3.5 had more formatting mistakes in code generation in June than in March. Overall, our findings shows that the behavior of the same LLM service can change substantially in a relatively short amount of time, highlighting the need for continuous monitoring of LLM quality.

Abstract: Dialogue-based human-AI collaboration can revolutionize collaborative problem-solving, creative exploration, and social support. To realize this goal, the development of automated agents proficient in skills such as negotiating, following instructions, establishing common ground, and progressing shared tasks is essential. This survey begins by reviewing the evolution of dialogue management paradigms in collaborative dialogue systems, from traditional handcrafted and information-state based methods to AI planning-inspired approaches. It then shifts focus to contemporary data-driven dialogue management techniques, which seek to transfer deep learning successes from form-filling and open-domain settings to collaborative contexts. The paper proceeds to analyze a selected set of recent works that apply neural approaches to collaborative dialogue management, spotlighting prevailing trends in the field. This survey hopes to provide foundational background for future advancements in collaborative dialogue management, particularly as the dialogue systems community continues to embrace the potential of large language models.

Abstract: The goal of Text-to-image person retrieval is to retrieve person images from a large gallery that match the given textual descriptions. The main challenge of this task lies in the significant differences in information representation between the visual and textual modalities. The textual modality conveys abstract and precise information through vocabulary and grammatical structures, while the visual modality conveys concrete and intuitive information through images. To fully leverage the expressive power of textual representations, it is essential to accurately map abstract textual descriptions to specific images. To address this issue, we propose a novel framework to Unleash the Imagination of Text (UIT) in text-to-image person retrieval, aiming to fully explore the power of words in sentences. Specifically, the framework employs the pre-trained full CLIP model as a dual encoder for the images and texts , taking advantage of prior cross-modal alignment knowledge. The Text-guided Image Restoration auxiliary task is proposed with the aim of implicitly mapping abstract textual entities to specific image regions, facilitating alignment between textual and visual embeddings. Additionally, we introduce a cross-modal triplet loss tailored for handling hard samples, enhancing the model's ability to distinguish minor differences. To focus the model on the key components within sentences, we propose a novel text data augmentation technique. Our proposed methods achieve state-of-the-art results on three popular benchmark datasets, and the source code will be made publicly available shortly.

Abstract: Despite being the current de-facto models in most NLP tasks, transformers are often limited to short sequences due to their quadratic attention complexity on the number of tokens. Several attempts to address this issue were studied, either by reducing the cost of the self-attention computation or by modeling smaller sequences and combining them through a recurrence mechanism or using a new transformer model. In this paper, we suggest to take advantage of pre-trained sentence transformers to start from semantically meaningful embeddings of the individual sentences, and then combine them through a small attention layer that scales linearly with the document length. We report the results obtained by this simple architecture on three standard document classification datasets. When compared with the current state-of-the-art models using standard fine-tuning, the studied method obtains competitive results (even if there is no clear best model in this configuration). We also showcase that the studied architecture obtains better results when freezing the underlying transformers. A configuration that is useful when we need to avoid complete fine-tuning (e.g. when the same frozen transformer is shared by different applications). Finally, two additional experiments are provided to further evaluate the relevancy of the studied architecture over simpler baselines.

Abstract: Gender bias in artificial intelligence (AI) and natural language processing has garnered significant attention due to its potential impact on societal perceptions and biases. This research paper aims to analyze gender bias in Large Language Models (LLMs) with a focus on multiple comparisons between GPT-2 and GPT-3.5, some prominent language models, to better understand its implications. Through a comprehensive literature review, the study examines existing research on gender bias in AI language models and identifies gaps in the current knowledge. The methodology involves collecting and preprocessing data from GPT-2 and GPT-3.5, and employing in-depth quantitative analysis techniques to evaluate gender bias in the generated text. The findings shed light on gendered word associations, language usage, and biased narratives present in the outputs of these Large Language Models. The discussion explores the ethical implications of gender bias and its potential consequences on social perceptions and marginalized communities. Additionally, the paper presents strategies for reducing gender bias in LLMs, including algorithmic approaches and data augmentation techniques. The research highlights the importance of interdisciplinary collaborations and the role of sociological studies in mitigating gender bias in AI models. By addressing these issues, we can pave the way for more inclusive and unbiased AI systems that have a positive impact on society.

Abstract: We analyze sentiment analysis and toxicity detection models to detect the presence of explicit bias against people with disability (PWD). We employ the bias identification framework of Perturbation Sensitivity Analysis to examine conversations related to PWD on social media platforms, specifically Twitter and Reddit, in order to gain insight into how disability bias is disseminated in real-world social settings. We then create the \textit{Bias Identification Test in Sentiment} (BITS) corpus to quantify explicit disability bias in any sentiment analysis and toxicity detection models. Our study utilizes BITS to uncover significant biases in four open AIaaS (AI as a Service) sentiment analysis tools, namely TextBlob, VADER, Google Cloud Natural Language API, DistilBERT and two toxicity detection models, namely two versions of Toxic-BERT. Our findings indicate that all of these models exhibit statistically significant explicit bias against PWD.

Abstract: As the probability (and thus perplexity) of a text is calculated based on the product of the probabilities of individual tokens, it may happen that one unlikely token significantly reduces the probability (i.e., increase the perplexity) of some otherwise highly probable input, while potentially representing a simple typographical error. Also, given that perplexity is a scalar value that refers to the entire input, information about the probability distribution within it is lost in the calculation (a relatively good text that has one unlikely token and another text in which each token is equally likely they can have the same perplexity value), especially for longer texts. As an alternative to scalar perplexity this research proposes a simple algorithm used to calculate vector values based on n-gram perplexities within the input. Such representations consider the previously mentioned aspects, and instead of a unique value, the relative perplexity of each text token is calculated, and these values are combined into a single vector representing the input.

Abstract: Relative positional encoding is widely used in vanilla and linear transformers to represent positional information. However, existing encoding methods of a vanilla transformer are not always directly applicable to a linear transformer, because the latter requires a decomposition of the query and key representations into separate kernel functions. Nevertheless, principles for designing encoding methods suitable for linear transformers remain understudied. In this work, we put together a variety of existing linear relative positional encoding approaches under a canonical form and further propose a family of linear relative positional encoding algorithms via unitary transformation. Our formulation leads to a principled framework that can be used to develop new relative positional encoding methods that preserve linear space-time complexity. Equipped with different models, the proposed linearized relative positional encoding (LRPE) family derives effective encoding for various applications. Experiments show that compared with existing methods, LRPE achieves state-of-the-art performance in language modeling, text classification, and image classification. Meanwhile, it emphasizes a general paradigm for designing broadly more relative positional encoding methods that are applicable to linear transformers. The code is available at https://github.com/OpenNLPLab/Lrpe.

Abstract: Siamese networks have gained popularity as a method for modeling text semantic similarity. Traditional methods rely on pooling operation to compress the semantic representations from Transformer blocks in encoding, resulting in two-dimensional semantic vectors and the loss of hierarchical semantic information from Transformer blocks. Moreover, this limited structure of semantic vectors is akin to a flattened landscape, which restricts the methods that can be applied in downstream modeling, as they can only navigate this flat terrain. To address this issue, we propose a novel 3D Siamese network for text semantic similarity modeling, which maps semantic information to a higher-dimensional space. The three-dimensional semantic tensors not only retains more precise spatial and feature domain information but also provides the necessary structural condition for comprehensive downstream modeling strategies to capture them. Leveraging this structural advantage, we introduce several modules to reinforce this 3D framework, focusing on three aspects: feature extraction, attention, and feature fusion. Our extensive experiments on four text semantic similarity benchmarks demonstrate the effectiveness and efficiency of our 3D Siamese Network.

Abstract: In this work, we develop and release Llama 2, a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Our models outperform open-source chat models on most benchmarks we tested, and based on our human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models. We provide a detailed description of our approach to fine-tuning and safety improvements of Llama 2-Chat in order to enable the community to build on our work and contribute to the responsible development of LLMs.

Abstract: We present the Multi-Modal Discussion Transformer (mDT), a novel multi-modal graph-based transformer model for detecting hate speech in online social networks. In contrast to traditional text-only methods, our approach to labelling a comment as hate speech centers around the holistic analysis of text and images. This is done by leveraging graph transformers to capture the contextual relationships in the entire discussion that surrounds a comment, with interwoven fusion layers to combine text and image embeddings instead of processing different modalities separately. We compare the performance of our model to baselines that only process text; we also conduct extensive ablation studies. We conclude with future work for multimodal solutions to deliver social value in online contexts, arguing that capturing a holistic view of a conversation greatly advances the effort to detect anti-social behavior.

Abstract: For real-world language applications, detecting an out-of-distribution (OOD) sample is helpful to alert users or reject such unreliable samples. However, modern over-parameterized language models often produce overconfident predictions for both in-distribution (ID) and OOD samples. In particular, language models suffer from OOD samples with a similar semantic representation to ID samples since these OOD samples lie near the ID manifold. A rejection network can be trained with ID and diverse outlier samples to detect test OOD samples, but explicitly collecting auxiliary OOD datasets brings an additional burden for data collection. In this paper, we propose a simple but effective method called Pseudo Outlier Exposure (POE) that constructs a surrogate OOD dataset by sequentially masking tokens related to ID classes. The surrogate OOD sample introduced by POE shows a similar representation to ID data, which is most effective in training a rejection network. Our method does not require any external OOD data and can be easily implemented within off-the-shelf Transformers. A comprehensive comparison with state-of-the-art algorithms demonstrates POE's competitiveness on several text classification benchmarks.

Abstract: Human-AI interactivity is a critical aspect that reflects the usability of multimodal large language models (MLLMs). However, existing end-to-end MLLMs only allow users to interact with them through language instructions, leading to the limitation of the interactive accuracy and efficiency. In this study, we present precise referring instructions that utilize diverse reference representations such as points and boxes as referring prompts to refer to the special region. This enables MLLMs to focus on the region of interest and achieve finer-grained interaction. Based on precise referring instruction, we propose ChatSpot, a unified end-to-end multimodal large language model that supports diverse forms of interactivity including mouse clicks, drag-and-drop, and drawing boxes, which provides a more flexible and seamless interactive experience. We also construct a multi-grained vision-language instruction-following dataset based on existing datasets and GPT-4 generating. Furthermore, we design a series of evaluation tasks to assess the effectiveness of region recognition and interaction. Experimental results showcase ChatSpot's promising performance.