Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2609.22633v1 Announce Type: new Abstract: Bilingual language models (LMs) offer a controlled setting for studying how training conditions shape second-language (L2) behaviour, but prior work ty...
arXiv:2609.23039v1 Announce Type: new Abstract: LLM-based AI systems answer political questions for hundreds of millions of people. Current audits measure what they say to an average user, but their...
arXiv:2609.23191v1 Announce Type: new Abstract: Speech-text space alignment is a multimodal representation learning method consisting to map different speech and text into a shared representation spa...
arXiv:2609.23267v1 Announce Type: new Abstract: The visual modality, i.e., images, plays a key role in multi-modal entity alignment (MMEA). Existing approaches often directly fuse the image with othe...
arXiv:2609.23726v1 Announce Type: new Abstract: Large language models have shown strong performance across a range of legal tasks, but existing benchmarks rarely evaluate the ability to take and defe...
arXiv:2609.23955v1 Announce Type: new Abstract: Previous work on Egyptian Arabic in NLP has focused largely on the prestigious Cairene Egyptian Arabic (CEA) dialect, resulting in a lack of representa...
arXiv:2609.23951v1 Announce Type: new Abstract: Expressive speech synthesis has advanced through prosody modeling, yet generating structured poetic speech, such as haiku, remains challenging. Prior w...
arXiv:2609.24194v1 Announce Type: new Abstract: Evaluation scores used around LLM systems -- including reward models, rerankers, and LLM judges -- can track surface form instead of the quality they c...
arXiv:2609.24372v1 Announce Type: new Abstract: In-context learning (ICL) based on large language models (LLMs) has shown promising potential in alleviating performance bottlenecks caused by the limi...
arXiv:2609.24516v1 Announce Type: new Abstract: In recent years, large language models (LLMs) have emerged as a popular alternative for evaluation. Often referred to as LLMs as judges (LLJs), these s...
arXiv:2609.24932v1 Announce Type: new Abstract: Despite the success of neural models in natural language processing, their black-box nature limits interpretability and conceals the linguistic phenome...
arXiv:2609.23071v1 Announce Type: cross Abstract: This thesis proposes a framework for "responsible intelligence" to address AI's critical challenges in safety, ethics, and cultural sensitivity. It a...
arXiv:2609.22949v1 Announce Type: cross Abstract: Existing prompt injection research focuses on single-model chatbot scenarios, where an attacker manipulates one LLM through crafted input. Multi-agen...
arXiv:2609.23416v1 Announce Type: cross Abstract: Long-form audio performance is often summarized by context length and aggregate accuracy, obscuring how language, evidence, and task jointly shape di...
arXiv:2609.23193v1 Announce Type: cross Abstract: As Large Language Model (LLM) APIs become increasingly integrated into privacy-sensitive workflows, ensuring inference-time privacy without compromis...
arXiv:2609.23979v1 Announce Type: cross Abstract: Attention encoder-decoder (AED) Speech Foundation Models achieve strong ASR performance but can generate acoustically unsupported text when inputs co...
arXiv:2404.05880v3 Announce Type: replace Abstract: Jailbreaking attacks can enable Large Language Models (LLMs) to bypass the safeguard and generate harmful content. Existing jailbreaking defense me...
arXiv:2411.02455v3 Announce Type: replace Abstract: The rapid adoption of generative AI has created new opportunities for teaching, learning, and quality assurance. Existing applications, however, re...
arXiv:2506.12576v3 Announce Type: replace Abstract: Sparse autoencoders (SAEs) can enable inference-time topic steering by modifying latent feature activations, but existing steering methods often fa...
arXiv:2601.03981v3 Announce Type: replace Abstract: To efficiently combat the spread of LLM-generated misinformation in the news domain, we present RADAR, a Retrieval-Augmented Detector with Adversar...