PPL
Bo Jiang
Papers (4)
Freeze Deep, Train Shallow: Interpretable Layer Allocation for Continued Pre-TrainingLEAP: Unlocking dLLM Parallelism via Lookahead Early-Convergence Token DetectionJoint Consistency: A Unified Test-Time Aggregation Framework via Energy MinimizationDynamic Pondering Sparsity-aware Mixture-of-Experts Transformer for Event Stream based Visual Object Tracking