33 accepted papers
Large language models (LLMs) have shown impressive multilingual capabilities through pretraining on diverse corpora. While these models show strong reasoning abilities, their performance varies significantly across languages due to imbalanced training data distribution. Existing approaches using sam
Refusal on harmful prompts is a key safety behaviour in instruction‑tuned large language models (LLMs), yet the internal causes of this behaviour remain poorly understood. We study two public instruction tuned models—Gemma‑2-2B‑IT and LLaMA‑3.1-8B‑IT using sparse autoencoders (SAEs) trained on resid
Ultra-long generation by large language models (LLMs) is a widely demanded scenario, yet it remains a significant challenge due to their maximum generation length limit and overall quality degradation as sequence length increases. Previous approaches, exemplified by LongWriter, typically rely on ''t…
Accurate detection of offensive content on social media demands high-quality labeled data; however, such data is often scarce due to the low prevalence of offensive instances and the high cost of manual annotation. To address this low-resource challenge, we propose a self-training framework that lev
We present SafeLens, a lightweight segment-level video moderation system that fuses speech, text, and visual frames to produce hateful content detection for each segment. For every segment, SafeLens returns a structured prediction: label, prediction confidence, reasons for flag, harm categories. The
We present a comprehensive empirical study on how backdoor attacks affect CLIP by analyzing the representations of backdoor images. Specifically, based on the methodology of representation decomposing, image representations can be decomposed into a sum of representations across individual image patc…
Multilingual Large Language Models (MLLMs) demonstrate strong generalization across languages, yet they remain prone to hallucinations, especially in low-resource languages, due to training data imbalances. These hallucinations, which include inaccurate or fabricated outputs, are particularly proble
Iterative data generation and model retraining are widely used to align large language models (LLMs).It typically involves a policy model to generate on-policy responses and a reward model to guide training data selection. Direct Preference Optimization (DPO) further enhances this process by constru…
Understanding humor is a core aspect of social intelligence, yet it remains a significant challenge for Large Multimodal Models (LMMs). We introduce PixelHumor, a benchmark dataset of 2,800 annotated multi-panel comics designed to evaluate LMMs’ ability to interpret multimodal humor and recognize na
Large Language Models (LLMs) have become essential for offensive language detection, yet their ability to handle annotation disagreement remains underexplored. Disagreement samples, which arise from subjective interpretations, pose a unique challenge due to their ambiguous nature. Understanding how…
Current benchmarks like ``$\textit{Needle-in-a-Haystack}$'' ($\textit{NIAH}$), $\textit{Ruler}$, and $\textit{Needlebench}$ focus on models' ability to understand long-context input sequences but fail to capture a critical dimension: the generation of high-quality long-form text. Applications such a…
Large Language Models (LLMs) can struggle to balance gullibility to misinformation and resistance to valid corrections in persuasive dialogues, a critical challenge for reliable deployment. We introduce **DuET-PD** (**Du**al **E**valuation for **T**rust in **P**ersuasive **D**ialogues), a framework
Large Language Models (LLMs) augmented with retrieval mechanisms have demonstrated significant potential in fact-checking tasks by integrating external knowledge. However, their reliability decreases when confronted with conflicting evidence from sources of varying credibility. This paper presents t
The advancement of Large Language Models (LLMs) has transformed natural language processing; however, their safety mechanisms remain under-explored in low-resource, multilingual settings. Here, we aim to bridge this gap. In particular, we introduce SGToxicGuard, a novel dataset and evaluation framew
Refusal is a key safety behavior in aligned language models, yet the internal mechanisms driving refusals remain opaque. In this work, we conduct a mechanistic study of refusal in instruction-tuned LLMs using sparse autoencoders to identify latent features that causally mediate refusal behaviors. We
Large language models (LLMs) have demonstrated remarkable capabilities in simulating human behaviour and social intelligence. However, they risk perpetuating societal biases, especially when demographic information is involved. We introduce a novel framework using cosine distance to measure semantic…
The widespread presence of hate speech on the internet, including formats such as text-based tweets and multimodal memes, poses a significant challenge to digital platform safety. Recent research has developed detection models tailored to specific modalities; however, there is a notable gap in trans…
Large language models (LLMs) have demonstrated impressive reasoning capabilities, particularly in textual mathematical problem-solving. However, existing open-source image instruction fine-tuning datasets, containing limited question-answer pairs per image, do not fully exploit visual information to…
Moderating hate speech (HS) in the evolving online landscape is a complex challenge, compounded by the multimodal nature of digital content. This survey examines recent advancements in HS moderation, focusing on the burgeoning role of large language models (LLMs) and large multimodal models (LMMs) i…
Detecting hate speech and offensive language is essential for maintaining a safe and respectful digital environment. This study examines the limitations of state-of-the-art large language models (LLMs) in identifying offensive content within systematically perturbed data, with a focus on Chinese, a…
Adapting a large language model for multiple-attribute text style transfer via fine-tuning can be challenging due to the substantial amount of computational resources and labeled data required for the specific downstream task. In this paper, we address this challenge by introducing \textsf{Adapter-T…
Recent studies have proposed models that yielded promising performance for the hateful meme classification task. Nevertheless, these proposed models do not generate interpretable explanations that uncover the underlying meaning and support the classification output. A major reason for the lack of ex…
Recent research has focused on using large language models (LLMs) to generate explanations for hate speech through fine-tuning or prompting. Despite the growing interest in this area, these generated explanations' effectiveness and potential limitations remain poorly understood. A key concern is tha…
Toxic content is a global problem, but most resources for detecting toxic content are in English. When datasets are created in other languages, they often focus exclusively on one language or dialect. In many cultural and geographical settings, however, it is common to code-mix languages, combining…
The success of large language models (LLMs), like GPT-4 and ChatGPT, has led to the development of numerous cost-effective and accessible alternatives that are created by finetuning open-access LLMs with task-specific data (e.g., ChatDoctor) or instruction data (e.g., Alpaca). Among the various fine…
Visual question answering (VQA) is the task of answering questions about an image. The task assumes an understanding of both the image and the question to provide a natural language answer. VQA has gained popularity in recent years due to its potential applications in a wide range of fields, includi…
Large language models (LLMs) have recently been shown to deliver impressive performance in various NLP tasks. To tackle multi-step reasoning tasks, Few-shot chain-of-thought (CoT) prompting includes a few manually crafted step-by-step reasoning demonstrations which enable LLMs to explicitly generate…
Authorship verification (AV) is a fundamental task in natural language processing (NLP) and computational linguistics, with applications in forensic analysis, plagiarism detection, and identification of deceptive content. Existing AV techniques, including traditional stylometric and deep learning ap…
Hateful meme classification is a challenging multimodal task that requires complex reasoning and contextual background knowledge. Ideally, we could leverage an explicit external knowledge base to supplement contextual and cultural information in hateful memes. However, there is no known explicit ext…
Language models such as GPT-2 have performed well on constructing syntactically sound sentences for text auto-completion tasks. However, such models often require considerable training effort to adapt to specific writing domains (e.g., medical). In this paper, we propose an intermediate training str…
Orthogonality constraints encourage matrices to be orthogonal for numerical stability. These plug-and-play constraints, which can be conveniently incorporated into model training, have been studied for popular architectures in natural language processing, such as convolutional neural networks and re…
Academia and industry have developed machine learning and natural language processing models to detect online hate speech automatically. However, most of these existing methods adopt a supervised approach that heavily depends on labeled datasets for training. This results in the methods’ poor detect…
Math word problem (MWP) is challenging due to the limitation in training data where only one “standard” solution is available. MWP models often simply fit this solution rather than truly understand or solve the problem. The generalization of models (to diverse word scenarios) is thus limited. To add…