Dimitris Roussis
droussis
AI & ML interests
All things data for LLMs, NMT, evaluation, safety, alignment, and more
Recent Activity
upvoted a paper 3 days ago
Co-Evolving Policy DistillationOrganizations
models 0
None public yet
datasets 5
droussis/quantization_calibration_data
Updated • 45
droussis/multilingual_multiturn_agentic_calibration_data
Viewer • Updated • 102 • 59 • 1
droussis/euroblocks_sft_1sample_per_lang
Viewer • Updated • 34 • 121 • 1
droussis/test_translation
Viewer • Updated • 1k • 17
droussis/UltraSafety_binarized-orpo-dpo
Viewer • Updated • 2.26k • 12 • 2