On the Design of Hyper-Connections: Simplification, Stability, and Scalability Paper • 2611.05984 • Published 13 days ago • 31
Truncate or Not? A Scaling Study of Looped Transformers with Truncated Backpropagation Paper • 2610.26030 • Published 28 days ago • 33
Towards Looped Models Done Right, Part I: Topology, Input Injection, Recurrent-State Design Paper • 2610.23036 • Published Oct 16 • 19
In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks Paper • 2609.38173 • Published Sep 29 • 49
WTF GENIUS PAPERS Collection Papers that made me appreciate my major and my life a little more. obs=Observation, innov=Innovation. Most papers are abt improving tiny models. • 541 items • Updated 1 day ago • 112
Towards Looped Models Done Right, Part II: Rethinking at Fixed Points Paper • 2610.06833 • Published Oct 5 • 26
WM-VLM: Probing Internal World Models for Interleaved Visual-Textual Reasoning Paper • 2609.34826 • Published Sep 28 • 9
Towards Looped Models Done Right Collection Looped language model checkpoints from Part I and Part II (Rethinking at Fixed Points). Native xLLM format; loading requires the xLLM code. • 49 items • Updated Oct 15 • 3
Improving Test-Time Scaling with Adaptive Looped Transformers Paper • 2609.35748 • Published Sep 28 • 64
Unlocking Lossless Speedups in LLMs via Discrete Diffusion Paper • 2609.04010 • Published Sep 3 • 112
SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers Paper • 2609.01343 • Published Sep 1 • 63
Contrasting Adversarial Perturbations: The Space of Harmless Perturbations Paper • 2402.02095 • Published Feb 3, 2024 • 1
Equilibrium Reasoners: Learning Attractors Enables Scalable Reasoning Paper • 2605.21488 • Published May 20 • 7
Efficient Agentic Reasoning Through Self-Regulated Simulative Planning Paper • 2605.22138 • Published May 21 • 10