2 papers
cs.CL2026
Misaligned by Reward: Socially Undesirable Preferences in LLMs
Gayane Ghazaryan, Esra Dönmez
Reward models are a key component of large language model alignment, serving as proxies for human preferences during training. However, existing evaluations focus primarily on broa…
cs.CL2024
SynDARin: Synthesising Datasets for Automated Reasoning in Low-Resource Languages
Gayane Ghazaryan, Erik Arakelyan, Pasquale Minervini +1
Question Answering (QA) datasets have been instrumental in developing and evaluating Large Language Model (LLM) capabilities. However, such datasets are scarce for languages other…