No article text was recorded for this version.
progressive token pruning
Version 2 of 2825 papers1,143 conceptsfrozen 2026-10-04
Papers in this version
- Blog: DoT: An efficient Double Transformer for NLP tasks with tables
- Blog: Input-length-shortening and text generation via attention values
- Blog: Attend, Select and Eliminate: Accelerating Multi-turn Response Selection with Dual-attention-based Content Elimination
- Blog: RASTP: Representation-Aware Semantic Token Pruning for Generative Recommendation with Semantic Identifiers
- Blog: Learning What to Remember: Adaptive Probabilistic Memory Retention for Memory-Efficient Language Models
- Blog: MrT5: Dynamic Token Merging for Efficient Byte-level Language Models
- Blog: You Only Forward Once: Prediction and Rationalization in A Single Forward Pass
- Blog: SpiralFovea: Input-Adaptive Foveated Tokenization as a Third Lever of Resource-Adaptive Inference
- Blog: SlimInfer: Accelerating Long-Context LLM Inference via Dynamic Token Pruning
- Blog: DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification
- Blog: Block Selective Reprogramming for On-device Training of Vision Transformers
- Blog: Conical Visual Concentration for Efficient Large Vision-Language Models
- Blog: Adaptive Layer Selection for Efficient Vision Transformer Fine-Tuning
- Blog: TR-PTS: Task-Relevant Parameter and Token Selection for Efficient Tuning
- Blog: Sparsified State-Space Models are Efficient Highway Networks
- Blog: Dynamic Tuning Towards Parameter and Inference Efficiency for ViT Adaptation
- Blog: A-ViT: Adaptive Tokens for Efficient Vision Transformer
- Blog: Reinforcement Learning-based Token Pruning in Vision Transformers: A Markov Game Approach
- Blog: Prune Spatio-temporal Tokens by Semantic-aware Temporal Accumulation
- Blog: Practical token pruning for foundation models in few-shot conversational virtual assistant systems
- Blog: Hierarchical Context Merging: Better Long Context Understanding for Pre-trained LLMs
- Blog: Similarity-Aware Token Pruning: Your VLM but Faster
- Blog: Sparsifying Transformer Models with Trainable Representation Pooling
- Blog: Optimizing Retrieval-augmented Reader Models via Token Elimination
- Blog: DoT: An efficient Double Transformer for NLP tasks with tables
- Blog: PIS: Linking Importance Sampling and Attention Mechanisms for Efficient Prompt Compression
- Blog: What Are You Token About? Differentiable Perturbed Top-k Token Selection for Scientific Document Summarization
- Blog: Variation-aware Vision Token Dropping for Faster Large Vision-Language Models
- Blog: Agile-Quant: Activation-Guided Quantization for Faster Inference of LLMs on the Edge
- Blog: AdaViT: Adaptive Tokens for Efficient Vision Transformer
- Blog: Sparsifying Transformer Models with Trainable Representation Pooling
- Blog: Learned Token Pruning for Transformers
- Blog: HaltingVT: Adaptive Token Halting Transformer for Efficient Video Recognition
- Blog: PACT: Pruning and Clustering-Based Token Reduction for Faster Visual Language Models
- Blog: Saliency-driven Dynamic Token Pruning for Large Language Models
- Blog: vid-TLDR: Training Free Token merging for Light-weight Video Transformer
- Blog: An Adaptive Language-Agnostic Pruning Method for Greener Language Models for Code
- Blog: Transkimmer: Transformer Learns to Layer-wise Skim
- Blog: Accelerating Transducers through Adjacent Token Merging
- Blog: CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion
- Blog: DTP: Delta-Guided Two Stage Pruning for Mamba-based Multimodal Large Language Models
- Blog: Input-length-shortening and text generation via attention values
- Blog: TR-BERT: Dynamic Token Reduction for Accelerating BERT Inference
- Blog: MrT5: Dynamic Token Merging for Efficient Byte-level Language Models
- Blog: ST3: Accelerating Multimodal Large Language Model by Spatial-Temporal Visual Token Trimming
- Blog: Skip-Vision: Efficient and Scalable Acceleration of Vision-Language Models via Adaptive Token Skipping
- Blog: Early Attentive Sparsification Accelerates Neural Speech Transcription
- Blog: Chasing Sparsity in Vision Transformers: An End-to-End Exploration
- Blog: Truth in the Few: High-Value Data Selection for Efficient Multi-Modal Reasoning
- Blog: AdapLeR: Speeding up Inference by Adaptive Length Reduction
- Blog: Exploring the Coordination of Frequency and Attention in Masked Image Modeling
- Blog: Chunk, Align, Select: A Simple Long-sequence Processing Method for Transformers
- Blog: ELIP: Efficient Language-Image Pre-training with Fewer Vision Tokens
- Blog: Tricks and Plugins to GBM on Images and Sequences
- Blog: Rethinking Token Reduction for State Space Models
- Blog: GTP-ViT: Efficient Vision Transformers via Graph-based Token Propagation
- Blog: DSPC: Dual-Stage Progressive Compression Framework for Efficient Long-Context Reasoning
- Blog: LeMo: Enabling LEss Token Involvement for MOre Context Fine-tuning
- Blog: CipherPrune: Efficient and Scalable Private Transformer Inference
- Blog: SegViT v2: Exploring Efficient and Continual Semantic Segmentation with Plain Vision Transformers
Showing 60 of 825 papers.
History & citation
Current · Permanent link
- Hierarchical Attention Stability for Token Pruning merged into this direction (57% overlap)
- Hardware Compatible Visual Token Pruning for Multimodal Transformers merged into this direction (83% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_progressive_token_pruning_2cf9f318,
title = {progressive token pruning},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/progressive-token-pruning/rdr_2cf9f318e35e47178fe1846836ab48ef},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-21},
}This exact version (version 2, frozen text):
@misc{lacuna_progressive_token_pruning_2cf9f318_v2,
title = {progressive token pruning},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/progressive-token-pruning/rdr_2cf9f318e35e47178fe1846836ab48ef/v/2},
note = {Version 2, frozen 2026-10-04. Accessed 2026-11-21},
}