As large language models (LLMs) constantly evolve, ensuring their safety remains a critical research issue. Previous red teaming approaches for LLM safety have primarily focused on single prompt attacks or goal hijacking. To the best of our knowledge, we are the first to study LLM safety in multi-turn dialogue coreference. We created a dataset of 1,400 questions across 14 categories, each featuring multi-turn coreference safety attacks. We then conducted detailed evaluations on five widely used open-source LLMs. The results indicated that under multi-turn coreference safety attacks, the highest attack success rate was 56% with the LLaMA2-Chat-7b model, while the lowest was 13.9% with the Mistral-7B-Instruct model. These findings highlight the safety vulnerabilities in LLMs during dialogue coreference interactions.

Despite being spoken by a large population of speakers worldwide, Cantonese is under-resourced in terms of the data scale and diversity compared to other major languages. This limitation has excluded it from the current “pre-training and fine-tuning” paradigm that is dominated by Transformer architectures.In this paper, we provide a comprehensive review on the existing resources and methodologies for Cantonese Natural Language Processing, covering the recent progress in language understanding, text generation and development of language models.We finally discuss two aspects of the Cantonese language that could make it potentially challenging even for state-of-the-art architectures: colloquialism and multilinguality.

Aspect words, indicating opinion targets, are essential in expressing and understanding human opinions. To identify aspects, most previous efforts focus on using sequence tagging models trained on human-annotated data. This work studies unsupervised aspect extraction and explores how words appear in global context (on sentence level) and local context (conveyed by neighboring words). We propose a novel neural model, capable of coupling global and local representation to discover aspect words. Experimental results on two benchmarks, laptop and restaurant reviews, show that our model significantly outperforms the state-of-the-art models from previous studies evaluated with varying metrics. Analysis on model output show our ability to learn meaningful and coherent aspect representations. We further investigate how words distribute in global and local context, and find that aspect and non-aspect words do exhibit different context, interpreting our superiority in unsupervised aspect extraction.

We present a system called ACE for Automatic Colloquialism and Errors detection for written Chinese. ACE is based on the combination of N-gram model and rule-base model. Although it focuses on detecting colloquial Cantonese (a dialect of Chinese) at the current stage, it can be extended to detect other dialects. We chose Cantonese becauase it has many interesting properties, such as unique grammar system and huge colloquial terms, that turn the detection task extremely challenging. We conducted experiments using real data and synthetic data. The results indicated that ACE is highly reliable and effective.