State of the Field
Current research in natural language processing is increasingly focused on enhancing the reliability and contextual understanding of language models. Recent work on child language assessment introduces metrics that evaluate the quality of children's utterances based on their contextual contributions, moving beyond traditional length-based measures. This shift aims to improve educational tools and developmental assessments. Concurrently, advancements in selective abstraction techniques for long-form text generation are addressing the issue of factual inaccuracies in language models, particularly in high-stakes applications, by allowing models to balance specificity and reliability. Additionally, studies comparing linear and quadratic attention mechanisms are refining our understanding of in-context learning, while efforts to improve multilingual embeddings through multi-way parallel text alignment are enhancing cross-lingual performance across diverse languages. Together, these developments signal a maturation of the field, emphasizing the importance of context, reliability, and cross-lingual capabilities in practical applications.
Papers
1–10 of 13A Fusion of context-aware based BanglaBERT and Two-Layer Stacked LSTM Framework for Multi-Label Cyberbullying Detection
Cyberbullying has become a serious and growing concern in todays virtual world. When left unnoticed, it can have adverse consequences for social and mental health. Researchers have explored various ty...
When Should LLMs Be Less Specific? Selective Abstraction for Reliable Long-Form Text Generation
LLMs are widely used, yet they remain prone to factual errors that erode user trust and limit adoption in high-risk settings. One approach to mitigate this risk is to equip models with uncertainty est...
T2S-Bench & Structure-of-Thought: Benchmarking and Prompting Comprehensive Text-to-Structure Reasoning
Think about how human handles complex reading tasks: marking key points, inferring their relationships, and structuring information to guide understanding and responses. Likewise, can a large language...
Learning to Generate and Extract: A Multi-Agent Collaboration Framework For Zero-shot Document-level Event Arguments Extraction
Document-level event argument extraction (DEAE) is essential for knowledge acquisition, aiming to extract participants of events from documents.In the zero-shot setting, existing methods employ LLMs t...
Beyond Length: Context-Aware Expansion and Independence as Developmentally Sensitive Evaluation in Child Utterances
Evaluating the quality of children's utterances in adult-child dialogue remains challenging due to insufficient context-sensitive metrics. Common proxies such as Mean Length of Utterance (MLU), lexica...
Humans and LLMs Diverge on Probabilistic Inferences
Human reasoning often involves working over limited information to arrive at probabilistic conclusions. In its simplest form, this involves making an inference that is not strictly entailed by a premi...
Task-Centric Acceleration of Small-Language Models
Small language models (SLMs) have emerged as efficient alternatives to large language models for task-specific applications. However, they are often employed in high-volume, low-latency settings, wher...
In-Context Learning in Linear vs. Quadratic Attention Models: An Empirical Study on Regression Tasks
Recent work has demonstrated that transformers and linear attention models can perform in-context learning (ICL) on simple function classes, such as linear regression. In this paper, we empirically st...
Ask don't tell: Reducing sycophancy in large language models
Sycophancy, the tendency of large language models to favour user-affirming responses over critical engagement, has been identified as an alignment failure, particularly in high-stakes advisory and soc...
Enhancing Multilingual Embeddings via Multi-Way Parallel Text Alignment
Multilingual pretraining typically lacks explicit alignment signals, leading to suboptimal cross-lingual alignment in the representation space. In this work, we show that training standard pretrained ...