The generation of text that is increasingly human-like has been enabled by the advent of large language models (LLMs). As the detection of AI-generated content holds significant importance in the fight against issues such as misinformation and academic cheating, numerous studies have been conducted to develop reliable LLM detectors. While promising results have been demonstrated by such detectors on test data, recent research has revealed that they can be circumvented by employing different techniques. In this article, homoglyph-based ($a \rightarrow {\alpha}$) attacks that can be used to circumvent existing LLM detectors are presented. The efficacy of the attacks is illustrated by analizing how homoglyphs shift the tokenization of the text, and thus its token loglikelihoods. A comprehensive evaluation is conducted to assess the effectiveness of homoglyphs on state-of-the-art LLM detectors, including Binoculars, DetectGPT, OpenAI's detector, and watermarking techniques, on five different datasets. A significant reduction in the efficiency of all the studied configurations of detectors and datasets, down to an accuracy of 0.5 (random guessing), is demonstrated by the proposed approach. The results show that homoglyph-based attacks can effectively evade existing LLM detectors, and the implications of these findings are discussed along with possible defenses against such attacks.

The widespread presence of hateful languages on social media has resulted in adverse effects on societal well-being. As a result, addressing this issue with high priority has become very important. Hate speech or offensive languages exist in both explicit and implicit forms, with the latter being more challenging to detect. Current research in this domain encounters several challenges. Firstly, the existing datasets primarily rely on the collection of texts containing explicit offensive keywords, making it challenging to capture implicitly offensive contents that are devoid of these keywords. Secondly, common methodologies tend to focus solely on textual analysis, neglecting the valuable insights that community information can provide. In this research paper, we introduce a novel dataset OffensiveLang, a community based implicit offensive language dataset generated by ChatGPT 3.5 containing data for 38 different target groups. Despite limitations in generating offensive texts using ChatGPT due to ethical constraints, we present a prompt-based approach that effectively generates implicit offensive languages. To ensure data quality, we evaluate the dataset with human. Additionally, we employ a prompt-based zero-shot method with ChatGPT and compare the detection results between human annotation and ChatGPT annotation. We utilize existing state-of-the-art models to see how effective they are in detecting such languages. The dataset is available here: https://github.com/AmitDasRup123/OffensiveLang

Benchmarks for large multimodal language models (MLMs) now serve to simultaneously assess the general capabilities of models instead of evaluating for a specific capability. As a result, when a developer wants to identify which models to use for their application, they are overwhelmed by the number of benchmarks and remain uncertain about which benchmark's results are most reflective of their specific use case. This paper introduces Task-Me-Anything, a benchmark generation engine which produces a benchmark tailored to a user's needs. Task-Me-Anything maintains an extendable taxonomy of visual assets and can programmatically generate a vast number of task instances. Additionally, it algorithmically addresses user queries regarding MLM performance efficiently within a computational budget. It contains 113K images, 10K videos, 2K 3D object assets, over 365 object categories, 655 attributes, and 335 relationships. It can generate 750M image/video question-answering pairs, which focus on evaluating MLM perceptual capabilities. Task-Me-Anything reveals critical insights: open-source MLMs excel in object and attribute recognition but lack spatial and temporal understanding; each model exhibits unique strengths and weaknesses; larger models generally perform better, though exceptions exist; and GPT4o demonstrates challenges in recognizing rotating/moving objects and distinguishing colors.

Sequential decision making addresses the problem of adapting an agent to an unknown environment, where the agent learns through a feedback loop by repeatedly receiving contexts, selecting actions, and observing feedback. This approach has been widely applied in real-world scenarios, including recommendation systems (Li et al., 2010; Agarwal et al., 2016), healthcare Tewari and Murphy (2017); Svensson (2023), and dialogue systems (Li et al., 2016). With the significant success of large language models (LLMs) in natural language processing (Brown et al., 2020; Ouyang et al., 2022; Achiam et al., 2023), an important next step is to extend this success to sequential decision making and enhance applications therein. Existing efforts to leverage LLMs for sequential decision making focus on two directions: (i) viewing decision making as sequence modeling and re-training or finetuning large models to adapt them to unknown environments (Chen et al., 2021; Zheng et al., 2022; Reid et al., 2022; Sun et al., 2023; Raparthy et al., 2023; Lee et al., 2024), and (ii) utilizing prompt engineering and in-context learning to adapt pretrained large models to sequential decision making problems (Krishnamurthy et al., 2024). While the first approach usually achieves promising empirical results, it is hindered by the substantial computational burden associated with re-training or finetuning large models, which often contain hundreds of billions of parameters. The second approach (Krishnamurthy et al., 2024), on the other hand, has demonstrated that most in-context learning and prompt engineering methods fail to effectively adapt LLMs to sequential decision making environments, except when employing the most advanced models, i.e., GPT-4 (Achiam et al., 2023), with sophisticated prompt designs. In this paper, we propose a new approach to efficiently incorporate large pretrained models into sequential decision making environments, without the need for expensive model re-training or finetuning. We run experiments (see Figure 1 and its caption for settings) on the AmazonCat-13K dataset (Bhatia et al., 2016) and observe that: Vanilla LLMs as decision making agents exhibit strong initial performance thanks to their significant commonsense knowledge and remarkable reasoning ability.

The effective alignment of Large Language Models (LLMs) with precise instructions is essential for their application in diverse real-world scenarios. Current methods focus on enhancing the diversity and complexity of training and evaluation samples, yet they fall short in accurately assessing LLMs' ability to follow similar instruction variants. We introduce an effective data augmentation technique that decomposes complex instructions into simpler sub-components, modifies these, and reconstructs them into new variants, thereby preserves the original instruction's context and complexity while introducing variability, which is critical for training and evaluating LLMs' instruction-following precision. We developed the DeMoRecon dataset using this method to both fine-tune and evaluate LLMs. Our findings show that LLMs fine-tuned with DeMoRecon will gain significant performance boost on both ours and commonly used instructions-following benchmarks.

Aspect Sentiment Triplet Extraction (ASTE) aims to co-extract the sentiment triplets in a given corpus. Existing approaches within the pretraining-finetuning paradigm tend to either meticulously craft complex tagging schemes and classification heads, or incorporate external semantic augmentation to enhance performance. In this study, we, for the first time, re-evaluate the redundancy in tagging schemes and the internal enhancement in pretrained representations. We propose a method to improve and utilize pretrained representations by integrating a minimalist tagging scheme and a novel token-level contrastive learning strategy. The proposed approach demonstrates comparable or superior performance compared to state-of-the-art techniques while featuring a more compact design and reduced computational overhead. Additionally, we are the first to formally evaluate GPT-4's performance in few-shot learning and Chain-of-Thought scenarios for this task. The results demonstrate that the pretraining-finetuning paradigm remains highly effective even in the era of large language models.

We propose Ruby Teaming, a method that improves on Rainbow Teaming by including a memory cache as its third dimension. The memory dimension provides cues to the mutator to yield better-quality prompts, both in terms of attack success rate (ASR) and quality diversity. The prompt archive generated by Ruby Teaming has an ASR of 74%, which is 20% higher than the baseline. In terms of quality diversity, Ruby Teaming outperforms Rainbow Teaming by 6% and 3% on Shannon's Evenness Index (SEI) and Simpson's Diversity Index (SDI), respectively.

Semantic code search, retrieving code that matches a given natural language query, is an important task to improve productivity in software engineering. Existing code search datasets are problematic: either using unrealistic queries, or with mismatched codes, and typically using one-to-one query-code pairing, which fails to reflect the reality that a query might have multiple valid code matches. This paper introduces CoSQA+, pairing high-quality queries (reused from CoSQA) with multiple suitable codes. We collect code candidates from diverse sources and form candidate pairs by pairing queries with these codes. Utilizing the power of large language models (LLMs), we automate pair annotation, filtering, and code generation for queries without suitable matches. Through extensive experiments, CoSQA+ has demonstrated superior quality over CoSQA. Models trained on CoSQA+ exhibit improved performance. Furthermore, we propose a new metric Mean Multi-choice Reciprocal Rank (MMRR), to assess one-to-N code search performance. We provide the code and data at https://github.com/DeepSoftwareAnalytics/CoSQA_Plus.

Language models typically tokenize raw text into sequences of subword identifiers from a predefined vocabulary, a process inherently sensitive to typographical errors, length variations, and largely oblivious to the internal structure of tokens-issues we term the curse of tokenization. In this study, we delve into these drawbacks and demonstrate that large language models (LLMs) remain susceptible to these problems. This study systematically investigates these challenges and their impact on LLMs through three critical research questions: (1) complex problem solving, (2) token structure probing, and (3) resilience to typographical variation. Our findings reveal that scaling model parameters can mitigate the issue of tokenization; however, LLMs still suffer from biases induced by typos and other text format variations. Our experiments show that subword regularization such as BPE-dropout can mitigate this issue. We will release our code and data to facilitate further research.

Large Language Models trained on code corpora (code-LLMs) have demonstrated impressive performance in various coding assistance tasks. However, despite their increased size and training dataset, code-LLMs still have limitations such as suggesting codes with syntactic errors, variable misuse etc. Some studies argue that code-LLMs perform well on coding tasks because they use self-attention and hidden representations to encode relations among input tokens. However, previous works have not studied what code properties are not encoded by code-LLMs. In this paper, we conduct a fine-grained analysis of attention maps and hidden representations of code-LLMs. Our study indicates that code-LLMs only encode relations among specific subsets of input tokens. Specifically, by categorizing input tokens into syntactic tokens and identifiers, we found that models encode relations among syntactic tokens and among identifiers, but they fail to encode relations between syntactic tokens and identifiers. We also found that fine-tuned models encode these relations poorly compared to their pre-trained counterparts. Additionally, larger models with billions of parameters encode significantly less information about code than models with only a few hundred million parameters.