This paper proposes a simple yet effective jailbreak attack named FlipAttack against black-box LLMs. First, from the autoregressive nature, we reveal that LLMs tend to understand the text from left to right and find that they struggle to comprehend the text when noise is added to the left side. Motivated by these insights, we propose to disguise the harmful prompt by constructing left-side noise merely based on the prompt itself, then generalize this idea to 4 flipping modes. Second, we verify the strong ability of LLMs to perform the text-flipping task, and then develop 4 variants to guide LLMs to denoise, understand, and execute harmful behaviors accurately. These designs keep FlipAttack universal, stealthy, and simple, allowing it to jailbreak black-box LLMs within only 1 query. Experiments on 8 LLMs demonstrate the superiority of FlipAttack. Remarkably, it achieves $\sim$98\% attack success rate on GPT-4o, and $\sim$98\% bypass rate against 5 guardrail models on average. The codes are available at GitHub\footnote{https://github.com/yueliu1999/FlipAttack}.

Hate speech has grown into a pervasive phenomenon, intensifying during times of crisis, elections, and social unrest. Multiple approaches have been developed to detect hate speech using artificial intelligence, but a generalized model is yet unaccomplished. The challenge for hate speech detection as text classification is the cost of obtaining high-quality training data. This study focuses on detecting bilingual hate speech in YouTube comments and measuring the impact of using additional data from other platforms in the performance of the classification model. We examine the value of additional training datasets from cross-platforms for improving the performance of classification models. We also included factors such as content similarity, definition similarity, and common hate words to measure the impact of datasets on performance. Our findings show that adding more similar datasets based on content similarity, hate words, and definitions improves the performance of classification models. The best performance was obtained by combining datasets from YouTube comments, Twitter, and Gab with an F1-score of 0.74 and 0.68 for English and German YouTube comments.

Knowledge tracing (KT) is a popular approach for modeling students' learning progress over time, which can enable more personalized and adaptive learning. However, existing KT approaches face two major limitations: (1) they rely heavily on expert-defined knowledge concepts (KCs) in questions, which is timeconsuming and prone to errors; and (2) KT methods tend to overlook the semantics of both questions and the given KCs. In this work, we address these challenges and present KCQRL, a framework for automated knowledge concept annotation and question representation learning that can improve the effectiveness of any existing KT model. First, we propose an automated KC annotation process using large language models (LLMs), which generates question solutions and then annotates KCs in each solution step of the questions. Second, we introduce a contrastive learning approach to generate semantically rich embeddings for questions and solution steps, aligning them with their associated KCs via a tailored false negative elimination approach. These embeddings can be readily integrated into existing KT models, replacing their randomly initialized embeddings. We demonstrate the effectiveness of KCQRL across 15 KT models on two large real-world Math learning datasets, where we achieve consistent performance improvements. The recent years have witnessed a surge in online learning platforms (Adedoyin & Soykan, 2023; Gros & García-Peñalvo, 2023), where students learn new knowledge concepts, which are then tested through exercises. Needless to say, personalization is crucial for effective learning: it allows that new knowledge concepts are carefully tailored to the current knowledge state of the student, which is more effective than one-size-fits-all approaches to learning (Cui & Sachan, 2023; Xu et al., 2024). However, such personalization requires that the knowledge of students is continuously assessed, which highlights the need for knowledge tracing (KT). In KT, one models the temporal dynamics of students' learning processes (Corbett & Anderson, 1994) in terms of a core set of skills, which are called knowledge concepts (KCs). KT models are typically time-series models that receive the past interactions of the learner as input (e.g., her previous exercises) in order to predict response of the learner to the next exercise. Yet, existing KT models have two main limitations that hinder their applicability in practice (see Figure 1). They require a comprehensive mapping between KCs and questions, which is typically done through manual annotations by experts. However, such KC annotation is both time-intensive and prone to errors (Clark, 2014; Bier et al., 2019). KT models overlook the semantics of both questions and KCs.

Open-ended short-answer questions (SAGs) have been widely recognized as a powerful tool for providing deeper insights into learners' responses in the context of learning analytics (LA). However, SAGs often present challenges in practice due to the high grading workload and concerns about inconsistent assessments. With recent advancements in natural language processing (NLP), automatic short-answer grading (ASAG) offers a promising solution to these challenges. Despite this, current ASAG algorithms are often limited in generalizability and tend to be tailored to specific questions. In this paper, we propose a unified multi-agent ASAG framework, GradeOpt, which leverages large language models (LLMs) as graders for SAGs. More importantly, GradeOpt incorporates two additional LLM-based agents - the reflector and the refiner - into the multi-agent system. This enables GradeOpt to automatically optimize the original grading guidelines by performing self-reflection on its errors. Through experiments on a challenging ASAG task, namely the grading of pedagogical content knowledge (PCK) and content knowledge (CK) questions, GradeOpt demonstrates superior performance in grading accuracy and behavior alignment with human graders compared to representative baselines. Finally, comprehensive ablation studies confirm the effectiveness of the individual components designed in GradeOpt.

Dataset distillation (DD) generates small synthetic datasets that can efficiently train deep networks with a limited amount of memory and compute. Despite the success of DD methods for supervised learning, DD for self-supervised pre-training of deep models has remained unaddressed. Pre-training on unlabeled data is crucial for efficiently generalizing to downstream tasks with limited labeled data. In this work, we propose the first effective DD method for SSL pre-training. First, we show, theoretically and empirically, that naive application of supervised DD methods to SSL fails, due to the high variance of the SSL gradient. Then, we address this issue by relying on insights from knowledge distillation (KD) literature. Specifically, we train a small student model to match the representations of a larger teacher model trained with SSL. Then, we generate a small synthetic dataset by matching the training trajectories of the student models. As the KD objective has considerably lower variance than SSL, our approach can generate synthetic datasets that can successfully pre-train high-quality encoders. Through extensive experiments, we show that our distilled sets lead to up to 13% higher accuracy than prior work, on a variety of downstream tasks, in the presence of limited labeled data.

Large language models (LLMs) deployed as agents solve user-specified tasks over multiple steps while keeping the required manual engagement to a minimum. Crucially, such LLMs need to ground their generations in any feedback obtained to reliably achieve desired outcomes. We propose an end-to-end reinforcement learning method for teaching models to leverage execution feedback in the realm of code synthesis, where state-of-the-art LLMs struggle to improve code iteratively compared to independent sampling. We benchmark on competitive programming tasks, where we achieve new start-of-the art results with both small (8B parameters) and large (70B) models while reducing the amount of samples required by an order of magnitude. Our analysis of inference-time behavior demonstrates that our method produces LLMs that effectively leverage automatic feedback over multiple steps.

Despite significant advancements in the reasoning capabilities of Large Language Models (LLMs), the lack of diverse reasoning solutions often makes them trapped in a limited solution search area. In this paper, we propose TypedThinker, a novel framework that enhances LLMs' problem-solving abilities by incorporating multiple reasoning types (deductive, inductive, abductive, and analogical). Our analysis across four benchmarks reveals that different reasoning types uniquely solve distinct sets of problems, highlighting the importance of diverse thinking approaches. TypedThinker addresses two key challenges: selecting appropriate reasoning types for given problems and effectively implementing specific reasoning types. Through self-training on successful experiences, TypedThinker learns an implicit policy for reasoning type selection and application. Experimental results demonstrate significant improvements over baseline models, with accuracy increases of 3.4% for Mistral 7B and 16.7% for LLaMA3 8B across four reasoning benchmarks. Notably, TypedThinker shows effective generalization to new benchmarks and can further enhance the reasoning capability of powerful models like GPT-4o. The code is released at https://github.com/dqwang122/ThinkHub.

In tackling the challenges of large language model (LLM) performance for Text-to-SQL tasks, we introduce CHASE-SQL, a new framework that employs innovative strategies, using test-time compute in multi-agent modeling to improve candidate generation and selection. CHASE-SQL leverages LLMs' intrinsic knowledge to generate diverse and high-quality SQL candidates using different LLM generators with: (1) a divide-and-conquer method that decomposes complex queries into manageable sub-queries in a single LLM call; (2) chain-of-thought reasoning based on query execution plans, reflecting the steps a database engine takes during execution; and (3) a unique instance-aware synthetic example generation technique, which offers specific few-shot demonstrations tailored to test questions.To identify the best candidate, a selection agent is employed to rank the candidates through pairwise comparisons with a fine-tuned binary-candidates selection LLM. This selection approach has been demonstrated to be more robust over alternatives. The proposed generators-selector framework not only enhances the quality and diversity of SQL queries but also outperforms previous methods. Overall, our proposed CHASE-SQL achieves the state-of-the-art execution accuracy of 73.0% and 73.01% on the test set and development set of the notable BIRD Text-to-SQL dataset benchmark, rendering CHASE-SQL the top submission of the leaderboard (at the time of paper submission).

We study the depth of grade-school math (GSM) problem-solving capabilities of LLMs. To this end, we evaluate their performance on pairs of existing math word problems together so that the answer to the second problem depends on correctly answering the first problem. Our findings reveal a significant reasoning gap in most LLMs, that is performance difference between solving the compositional pairs and solving each question independently. This gap is more pronounced in smaller, more cost-efficient, and math-specialized models. Moreover, instruction-tuning recipes and code generation have varying effects across LLM sizes, while finetuning on GSM can lead to task overfitting. Our analysis indicates that large reasoning gaps are not because of test-set leakage, but due to distraction from additional context and poor second-hop reasoning. Overall, LLMs exhibit systematic differences in their reasoning abilities, despite what their performance on standard benchmarks indicates.

Various visual foundation models have distinct strengths and weaknesses, both of which can be improved through heterogeneous multi-teacher knowledge distillation without labels, termed "agglomerative models." We build upon this body of work by studying the effect of the teachers' activation statistics, particularly the impact of the loss function on the resulting student model quality. We explore a standard toolkit of statistical normalization techniques to better align the different distributions and assess their effects. Further, we examine the impact on downstream teacher-matching metrics, which motivates the use of Hadamard matrices. With these matrices, we demonstrate useful properties, showing how they can be used for isotropic standardization, where each dimension of a multivariate distribution is standardized using the same scale. We call this technique "PHI Standardization" (PHI-S) and empirically demonstrate that it produces the best student model across the suite of methods studied.