Top Research Papers on LLMs

Large Language ModelsIn the latest edition of Stats, STAT!, Fralick and colleagues explain the statistics behind large language models - used in chat bots like ChatGPT and Bard. While these new tools may seem remarkably intelligent, at their core they just assemble sentences based on statistics from large amounts of text.

This essay argues that such failures—so-called hallucinations—are not accidental glitches but are instead a by-product of the design of the transformer architecture on which large language models are built, given its foundation on the distributional hypothesis, a nonreferential approach to meaning.

This work firstly converts the input image into multiple suitable text-based abstraction spaces, then utilises the associative power of LLMs to derive the input-output relationship and map this to actions in the form of a working program, similar to Voyager / Ghost in the MineCraft.

It is suggested that large language models can effectively extract genomic characteristics from unstructured reports and can potentially improve genomic literacy among prostate cancer patients by providing easy-to-interpret variant summaries, however, targeted prompting may be required to increase lexical diversity for improved engagement.

It is argued that the importance of representation as well as multi-modality are likely key to making the new generation of systems more powerful, usable, accessible and utile for all.

This work presents a method which further expands the capabilities of an LLM by embedding it within an algorithm or program, and presents an illustrative example of evidence-supported question-answering.

It is implausible to think that LLMs possess communicative desires and perform speech acts, yet it is implausible not to think that LLMs possess communicative desires and perform speech acts.

Interacting with a contemporary LLM-based conversational agent can create an illusion of being in the presence of a thinking creature, yet such systems are fundamentally not like us.

This work presents LLM fundamentals, including model architecture, pre-training, fine-tuning, inference and utilization, model evaluation, and telecom deployment, and introduces LLM-enabled key techniques and telecom applications in terms of generation, classification, optimization, and prediction problems.

LLaSM is an end-to-end trained large multi-modal speech-language model with cross- modal conversational abilities, capable of following speech-and-language instructions.

GPT-4 performed well on lower-order as well as higher-order questions, making ChatGPT-4 a potentially very useful tool for reviewing radiology exam questions, and Radiologists should be aware of ChatGPT's limitations, including its tendency to confidently provide incorrect responses.

This talk will discuss the recent progress on LLM features and the potential of LLM in enabling intelligent wireless communication systems.

The use of large language models (LLMs) is now spreading in several areas of research and development. This work is concerned with systematically reviewing LLMs’ involvement in engineering education. Starting from a general research question, two queries were used to select 370 papers from the literature. Filtering them through several inclusion/exclusion criteria led to the selection of 20 papers. These were investigated based on eight dimensions to identify areas of engineering disciplines that involve LLMs, where they are most present, how this involvement takes place, and which LLM-based t...

An electronic survey for students across North American medical colleges to gauge their views on and current use of ChatGPT and similar technologies found that 96% of respondents had heard of ChatGPT and 52% had used it for medical school coursework.

The LLMs studied show competence comparable to that of medical specialists in the interpretation of clinical reports, even in complex and confusingly worded texts, and represent a preferred option over human analysis for data mining and structuring information in extensive sets of clinical reports.

This paper proposes a novel approach based on active inference for LLMs inference task offloading and resource allocation in cloud-edge computing that has superior performance over mainstream DRLs, improves in data utilization efficiency, and is more adaptable to changing task load scenarios.

A hybrid approach that combines the strengths of open-source large and small-scale language models (LLMs and LMs) with traditional forecasting methods, outperforming existing methods by significant margins in terms of forecast accuracy.

,

This study specifically focuses on evaluating the performance of LLMs for Named Entity Recognition and Part-of-Speech tagging for a low-resource language, Nepali, and compares their performance with that of alternative approaches deployed for the tasks.

We analyze the security implications of large language models (LLMs) from their use as security tools for both attackers and defenders and the security of LLMs. We discuss how LLMs increase the scale of traditional threats such as social engineering and add new ones such as prompt injections.