Alankar Jain(@alankarjain91):Agent Distillation can be surprisingly data-efficient. We distilled Inkling-Small into Nemotron-3-Nano on incident-diagnosis problems using trajectories from only 19 problems! Very encouraging for post-training task-specific models in data-poor regimes.

Alankar Jain

@alankarjain91

Co-founder, @nexttoken_co, an Applied AI lab focused on post-training open models Prev: Staff SWE @google Search, NLP @carnegiemellon, EECS @iitbombay

参加 June 2010

336 フォロー中 294 ファン

Alankar Jain@alankarjain91

2026.08.04 15:42

Agent Distillation can be surprisingly data-efficient. We distilled Inkling-Small into Nemotron-3-Nano on incident-diagnosis problems using trajectories from only 19 problems! Very encouraging for post-training task-specific models in data-poor regimes.