Can Language Models Learn When to Abstain? Policy Learning and Calibrated Answer Selection
A learned answer selector on Qwen3-14B improves TriviaQA coverage and lowers scored-answer errors versus confidence filtering.
1 stories, newest first · one card per story
A learned answer selector on Qwen3-14B improves TriviaQA coverage and lowers scored-answer errors versus confidence filtering.
Today 2.2K readers · 2.7K views · 397 reads · all time 108K views