Memory-Efficient Looped Transformer: Decoupling Compute from Memory in Looped Language Models
Paper • 2605.07721 • Published • 16
Pratik Kumar
kpratik
·
AI & ML interests
Foundational Models
Recent Activity
upvoted a paper about 4 hours ago
Memory-Efficient Looped Transformer: Decoupling Compute from Memory in Looped Language Models upvoted a paper about 2 months ago
Efficient Reasoning on the Edge upvoted a paper about 2 months ago
Reasoning as Compression: Unifying Budget Forcing via the Conditional Information BottleneckOrganizations
upvoted a paper about 4 hours ago