Teaching Pretrained Language Models to Think Deeper with Retrofitted Recurrence
Paper • 2511.07384 • Published • 21
LLM Training Methodologies
-
-
Believe Your Model: Distribution-Guided Confidence Calibration
Paper • 2603.03872 • Published • 40 -
WiT: Waypoint Diffusion Transformers via Trajectory Conflict Navigation
Paper • 2603.15132 • Published • 35 -
Long Context Pre-Training with Lighthouse Attention
Paper • 2605.06554 • Published • 31 -
Darwin Family: MRI-Trust-Weighted Evolutionary Merging for Training-Free Scaling of Language-Model Reasoning
Paper • 2605.14386 • Published • 69 -
DataFlow-Harness: A Grounded Code-Agent Platform for Constructing Editable LLM Data Pipelines
Paper • 2607.16617 • Published • 135