2 papers
cs.CL2026
PolyFact: Comparing Consistency-Driven Post-training Methods for Cross-Lingual Factual Recall
Jonathan von Rad, Louis Arts, George Burgess +6
Large language models (LLMs) trained predominantly on English data encode substantial world knowledge, yet often fail to express it reliably in other languages, a phenomenon known…
astro-ph.GA2024
Evaluating the Predictive Capacity of FLARES Simulations for High Redshift "Little Red Dots"
Louis M. T. Arts
The recent discovery of little red dots - a population of extremely compact and highly dust-reddened high redshift galaxies - by the James Webb Space Telescope presents a new chall…