First-token cueing recovers closed-book facts LLMs fail to start
Prefilling only the gold answer's first token fixes a fifth to a half of multi-token English QA errors across 16 open models.
1 stories, newest first · one card per story
Prefilling only the gold answer's first token fixes a fifth to a half of multi-token English QA errors across 16 open models.
Today 1.5K readers · 2.4K views · 757 reads · all time 189K views