Recently, large language models (LLMs) have shown surprising performance in task-specific workloads as well as general tasks with the given prompts. However, to achieve unprecedented performance, recent LLMs use billions to trillions of parameters, which hinder the wide adaptation of those models due to their extremely large compute and memory requirements. To resolve the issue, various model compression methods are being actively investigated. In this work, we propose SDQ (Sparse Decomposed Quantization) to exploit both structured sparsity and quantization to achieve both high compute and memory efficiency. From our evaluations, we observe that SDQ can achieve 4 effective compute throughput with <1% quality drop. Previous efforts (Hoefler et al., 2021) have shown how to compress classic DNNs by more Large Language Models (LLMs) (Brown et al., 2020; than 90% (10 computation reduction) with limited loss of Chowdhery et al., 2022; Touvron et al., 2023b) with billions model quality; however, when it comes to LLMs, compressing or trillions of parameters have gained extensive attention as beyond 50% (2 computation reduction) with a limited they show promising quality in various domains.

Email continues to be a pivotal and extensively utilized communication medium within professional and commercial domains. Nonetheless, the prevalence of spam emails poses a significant challenge for users, disrupting their daily routines and diminishing productivity. Consequently, accurately identifying and filtering spam based on content has become crucial for cybersecurity. Recent advancements in natural language processing, particularly with large language models like ChatGPT, have shown remarkable performance in tasks such as question answering and text generation. However, its potential in spam identification remains underexplored. To fill in the gap, this study attempts to evaluate ChatGPT's capabilities for spam identification in both English and Chinese email datasets. We employ ChatGPT for spam email detection using in-context learning, which requires a prompt instruction and a few demonstrations. We also investigate how the number of demonstrations in the prompt affects the performance of ChatGPT. For comparison, we also implement five popular benchmark methods, including naive Bayes, support vector machines (SVM), logistic regression (LR), feedforward dense neural networks (DNN), and BERT classifiers. Through extensive experiments, the performance of ChatGPT is significantly worse than deep supervised learning methods in the large English dataset, while it presents superior performance on the low-resourced Chinese dataset.

Despite significant advancements in computer vision, understanding complex scenes, particularly those involving humor, remains a substantial challenge. This paper introduces HumorDB, a novel image-only dataset specifically designed to advance visual humor understanding. HumorDB consists of meticulously curated image pairs with contrasting humor ratings, emphasizing subtle visual cues that trigger humor and mitigating potential biases. The dataset enables evaluation through binary classification(Funny or Not Funny), range regression(funniness on a scale from 1 to 10), and pairwise comparison tasks(Which Image is Funnier?), effectively capturing the subjective nature of humor perception. Initial experiments reveal that while vision-only models struggle, vision-language models, particularly those leveraging large language models, show promising results. HumorDB also shows potential as a valuable zero-shot benchmark for powerful large multimodal models.

This paper investigates the possibility of intuitive human-robot interaction through the application of Natural Language Processing (NLP) and Large Language Models (LLMs) in mobile robotics. We aim to explore the feasibility of using these technologies for edge-based deployment, where traditional cloud dependencies are eliminated. The study specifically contrasts the performance of GPT-4-Turbo, which requires cloud connectivity, with an offline-capable, quantized version of LLaMA 2 (LLaMA 2-7B.Q5 K M). Our results show that GPT-4-Turbo delivers superior performance in interpreting and executing complex commands accurately, whereas LLaMA 2 exhibits significant limitations in consistency and reliability of command execution. Communication between the control computer and the mobile robot is established via a Raspberry Pi Pico W, which wirelessly receives commands from the computer without internet dependency and transmits them through a wired connection to the robot's Arduino controller. This study highlights the potential and challenges of implementing LLMs and NLP at the edge, providing groundwork for future research into fully autonomous and network-independent robotic systems. For video demonstrations and source code, please refer to: https://tinyurl.com/MobileRobotGPT4LLaMA2024.

While large language models exhibit certain cross-lingual generalization capabilities, they suffer from performance degradation (PD) on unseen closely-related languages (CRLs) and dialects relative to their high-resource language neighbour (HRLN). However, we currently lack a fundamental understanding of what kinds of linguistic distances contribute to PD, and to what extent. Furthermore, studies of cross-lingual generalization are confounded by unknown quantities of CRL language traces in the training data, and by the frequent lack of availability of evaluation data in lower-resource related languages and dialects. To address these issues, we model phonological, morphological, and lexical distance as Bayesian noise processes to synthesize artificial languages that are controllably distant from the HRLN. We analyse PD as a function of underlying noise parameters, offering insights on model robustness to isolated and composed linguistic phenomena, and the impact of task and HRL characteristics on PD. We calculate parameter posteriors on real CRL-HRLN pair data and show that they follow computed trends of artificial languages, demonstrating the viability of our noisers. Our framework offers a cheap solution to estimating task performance on an unseen CRL given HRLN performance using its posteriors, as well as for diagnosing observed PD on a CRL in terms of its linguistic distances from its HRLN, and opens doors to principled methods of mitigating performance degradation.

Words have been represented in a high-dimensional vector space that encodes their semantic similarities, enabling downstream applications such as retrieving synonyms, antonyms, and relevant contexts. However, despite recent advances in multilingual language models (LMs), the effectiveness of these models' representations in semantic retrieval contexts has not been comprehensively explored. To fill this gap, this paper introduces the MINERS, a benchmark designed to evaluate the ability of multilingual LMs in semantic retrieval tasks, including bitext mining and classification via retrieval-augmented contexts. We create a comprehensive framework to assess the robustness of LMs in retrieving samples across over 200 diverse languages, including extremely low-resource languages in challenging cross-lingual and code-switching settings. Our results demonstrate that by solely retrieving semantically similar embeddings yields performance competitive with state-of-the-art approaches, without requiring any fine-tuning.

Text embedding methods have become increasingly popular in both industrial and academic fields due to their critical role in a variety of natural language processing tasks. The significance of universal text embeddings has been further highlighted with the rise of Large Language Models (LLMs) applications such as Retrieval-Augmented Systems (RAGs). While previous models have attempted to be general-purpose, they often struggle to generalize across tasks and domains. However, recent advancements in training data quantity, quality and diversity; synthetic data generation from LLMs as well as using LLMs as backbones encourage great improvements in pursuing universal text embeddings. In this paper, we provide an overview of the recent advances in universal text embedding models with a focus on the top performing text embeddings on Massive Text Embedding Benchmark (MTEB). Through detailed comparison and analysis, we highlight the key contributions and limitations in this area, and propose potentially inspiring future research directions.

The opacity in developing large language models (LLMs) is raising growing concerns about the potential contamination of public benchmarks in the pre-training data. Existing contamination detection methods are typically based on the text overlap between training and evaluation data, which can be too superficial to reflect deeper forms of contamination. In this paper, we first present a cross-lingual form of contamination that inflates LLMs' performance while evading current detection methods, deliberately injected by overfitting LLMs on the translated versions of benchmark test sets. Then, we propose generalization-based approaches to unmask such deeply concealed contamination. Specifically, we examine the LLM's performance change after modifying the original benchmark by replacing the false answer choices with correct ones from other questions. Contaminated models can hardly generalize to such easier situations, where the false choices can be \emph{not even wrong}, as all choices are correct in their memorization. Experimental results demonstrate that cross-lingual contamination can easily fool existing detection methods, but not ours. In addition, we discuss the potential utilization of cross-lingual contamination in interpreting LLMs' working mechanisms and in post-training LLMs for enhanced multilingual capabilities. The code and dataset we use can be obtained from \url{https://github.com/ShangDataLab/Deep-Contam}.

How can we construct an automated debate judge to evaluate an extensive, vibrant, multi-turn debate? This task is challenging, as judging a debate involves grappling with lengthy texts, intricate argument relationships, and multi-dimensional assessments. At the same time, current research mainly focuses on short dialogues, rarely touching upon the evaluation of an entire debate. In this paper, by leveraging Large Language Models (LLMs), we propose Debatrix, which makes the analysis and assessment of multi-turn debates more aligned with majority preferences. Specifically, Debatrix features a vertical, iterative chronological analysis and a horizontal, multi-dimensional evaluation collaboration. To align with real-world debate scenarios, we introduced the PanelBench benchmark, comparing our system's performance to actual debate outcomes. The findings indicate a notable enhancement over directly using LLMs for debate evaluation. Source code and benchmark data are available online at https://github.com/ljcleo/debatrix .

In this paper, we identify a new category of bias that induces input-conflicting hallucinations, where large language models (LLMs) generate responses inconsistent with the content of the input context. This issue we have termed the false negative problem refers to the phenomenon where LLMs are predisposed to return negative judgments when assessing the correctness of a statement given the context. In experiments involving pairs of statements that contain the same information but have contradictory factual directions, we observe that LLMs exhibit a bias toward false negatives. Specifically, the model presents greater overconfidence when responding with False. Furthermore, we analyze the relationship between the false negative problem and context and query rewriting and observe that both effectively tackle false negatives in LLMs.