token-efficient
repro-omnisift-modality-asymmetric-token-compression-for-efficient-omni-modal-large-language-modrepro-toast-token-channel-selection-and-structured-pruning-for-efficient-vitrepro-efficient-training-free-multi-token-prediction-via-embedding-space-probingrepro-from-token-to-token-pair-efficient-prompt-compression-for-large-language-models-in-clinicarepro-memory-as-a-markov-matrix-sample-efficient-knowledge-expansion-via-token-to-direpro-rtprune-reading-twice-inspired-token-pruning-for-efficient-deepseek-ocr-inference
