Minghan Wang from Monash University is presenting his long paper titled "Simultaneous Machine Translation with Large Language Models" online.

➡️ This paper investigates the possibility of applying Large Language Models (#LLM) to #SimulMT tasks by using existing incremental-decoding methods with a newly proposed #RALCP algorithm for latency reduction.
➡️ They conducted experiments using
the #Llama2-7b-chat model on nine different languages from the #MUST-C dataset.

Milindi Kodikara from RMIT University is presenting the long paper titled "Lesser the Shots, Higher the Hallucinations: Exploration of Genetic Information Extraction using Generative Large
Language Models".

This paper systematically evaluates the performance of
generative large language models (#LLMs) on the extraction of specialised genetic information, focusing on end-to-end IE encompassing both named entity recognition and relation extraction.

Kritesh Rauniyar from Delhi Technological University is presenting their paper titled "Which Side Are You On? Investigating Politico-Economic Bias in Nepali Language Models" online.

This paper addresses this gap by investigating the political and economic biases present in five fill-mask models and eleven generative models trained for the Nepali language.

Daniel Cabrera Lozoya from The University of Melbourne is presenting his paper titled "Truth in the Noise: Unveiling Authentic Dementia Self-Disclosure Statements in Social Media with #LLMs

This study implemented a genetic algorithm that evolves prompts using various state-of-the-art prompt engineering techniques, including chain of thought, self-critique, generated knowledge, and expert prompting.

Minkang Liu from the Universiti Sains Malaysia is presenting his papar titled "#GenABSA-Vec: Generative Aspect-Based Sentiment Feature Vectorization for Document-Level Sentiment Classification".

This paper proposes a method to construct a #GenABSA feature vector containing five aspect-sentiment scores to represent each review document.

Emmanuele Chersoni Hong Kong Polytechnic presents joint work with Carina Kauf #MIT on "Comparing Plausibility Estimates in Base and Instruction-Tuned Large Language Models (Abstract)"

This research aims to explore the effect of the instruction tuning process on a model’s knowledge of semantic plausibility.

And as prompting does not suffer from the confounders of log-likelihoods (e.g. frequency, word length etc.), could it turn out to be a better method for extracting plausibility knowledge?

Michael Lambropoulos from the University of Sydney who cannot come alone in person is presenting his paper titled "Towards an Implementation of Rhetorical Structure Theory in Discourse Coherence Modelling" online.

➡️ This paper combines the discourse coherence principles of Elementary Discourse Unit segmentation and Rhetorical Structure Theory parsing to construct meaningful graph-based text representations.

Good morning 😀

Kyla Quinn who is the Technical Director of Data and Analytic Services Branch at the Australian Signals Directorate presented #keynotes titled LLMs are great but …

In this talk, she is
➡️ exploring some of the issues we need to contend with when we put LLMs and other language technologies into an enterprise
➡️ touching on data preprocessing, governance, user trust and interpretation

Long Hei Matthew Lam from Monash University has given the oral presentation titled A Closer Look at Tool-based Logical Reasoning with #LLMs: The Choice of Tool Matters.

In this paper, he fills the gaps in the comparison between symbolic solvers, including #Z3, #Pyke, and
#Prover9 with #LLMs augmented.