Paper treats long-context retrieval as value-indexed local traversal
On MiniGrid Memory, a five-token traversal head reached 0.95 mean return after 0.59M steps, ahead of absolute addressing.
1 stories, newest first · one card per story
On MiniGrid Memory, a five-token traversal head reached 0.95 mean return after 0.59M steps, ahead of absolute addressing.
Today 1.5K readers · 2.4K views · 757 reads · all time 189K views