large language models

A comparative study of large language models in responding to breast cancer–related questions

No single model outperformed the others in all the test areas, so many different LLMs were chosen to provide support for breast cancer health information.

Digital Twins in the Architectural Design Stage for Sustainable Net-Zero Buildings: A Systematic Review of Frameworks, Tools and Research Gaps

Early architectural decisions shape building energy demand and life cycle carbon, yet digital support remains fragmented across modelling, simulation and performance workflows. This systematic review examines how digital twin (DT) frameworks, tools and workflows are applied at the architectural design stage to support...

NZACU-Core Text

This document is a white paper that decodes the source code of our universe. Anyone can intuitively understand the validity of this theory by feeding this text into a Large Language Model (LLM) such as Gemini to discuss and debug it. For humanity's next OS update, I hereby release this 'Specification of the Universe' a...

Can Large Language Models (LLMs) be Trusted for Power Analysis? An Empirical Evaluation

It is suggested that while LLMs may offer support in structuring and initiating power analysis, they cannot substitute for expert judgment and underscores the importance of critically evaluating LLM performance in statistically demanding tasks.

Emulation Diagnostics for Model Self-Reports: A Bridge Between Interpretability and Welfare Assessment

Two adjacent but methodologically separate substrates currently coexist inside Anthropic's published record. The first is a body of mechanistic-interpretability work, most prominently On the Biology of a Large Language Model, that produces attribution graphs over sparse activation features and explicitly cautions that...

The Ghost Pattern Codebook: A Public-Surface Taxonomy of LLM Output Patterns

Large language model outputs fail in ways that the standard evaluation categories, factual error, refusal, hallucination, sycophancy, do not adequately capture. A separate class of failure appears at the level of output form: text that remains fluent, structured, and superficially credible while abandoning the inferent...

Can Large Language Models Provide Safe and Evidence-Based Nutritional Recommendations for Dental Implant Patients? A Comparative Benchmarking Study

If It's Not Buggy, Don't Fix It: On the Dynamics of Iterative Bug-fixing with LLMs

This report explores the iterative blind use of LLMs as bug-fixers, and unveils the existence of a steering vector which controls the editing propensity, suggesting that LLMs have an internal representation of ``buggy code", and that this representation is what is falsely activated to induce pseudo-bug fixing.

5-Dialects-BN: Unmasking the Impact of Transliteration on Bangla Dialectal LLMs

Large Language Models (LLMs) have achieved remarkable progress across natural language processing (NLP) tasks, yet their capabilities degrade sharply for low-resource languages and dialectally diverse settings. Bangla, the world's sixth most spoken language, exemplifies this gap: existing resources overwhelmingly targe...

$S^3$-Bench: Evaluating Speech Interaction Models as Scientific Voice Assistants

S$^3$-Bench is introduced, a systematic evaluation framework covering 10 major disciplines, consisting of a Knowledge set for speech question-answering and a Dialogue set for multi-turn progressive interactions with simulated user agents that systematically characterize the common challenges and performance tradeoffs o...

Do LLMs Make More Mistakes If They Do Not Believe the Input Data?

How faithfulness of LLMs to provided context depends on how plausible they perceive the context to be (context-memory conflict) is analyzed, which makes use of the increased difficulty of non-English and low-resource language text generation and input data based on local knowledge.

Vague2Detect: Handling Ambiguous Prompts in Knowledge-Based Open-World Detection

This work proposes Vague2Detect, a hybrid pipeline in which a fine-tuned Sentence-BERT retrieves candidates from a structured household Knowledge Base, and YOLO-World verifies their presence in the image, and improves performance to 61% VPSR with high precision, and up to 85% when augmented with GPT fallback.