arXiv:2607. 02612v1 Announce Type: cross Abstract: Vision Transformers achieve strong image classification accuracy but process all image regions with nearly the same computation, even when many regions are redundant or uninformative.
arXiv:2608. 13141v1 Announce Type: cross Abstract: Vision Transformers (ViTs) demonstrate exceptional performance in computer vision but suffer from large parameter counts and quadratic computational complexity, severely limiting their deployment on resource-constrained edge hardware.
arXiv:2610.07758v1 Announce Type: cross Abstract: Training-free token reduction accelerates vision transformers by removing redundant tokens across layers, recovering most of the original accuracy at...
arXiv:2610.33688v1 Announce Type: new Abstract: Visual token pruning substantially reduces the inference cost of multimodal large language models (MLLMs), yet existing methods are mostly training-fre...
The paper studies Graph-Guided Token Merging (G2TM), a module that reduces token count in Vision Transformers. It evaluates G2TM across multiple segmentation frameworks and decoder types, finding that its performance gains are tied to the encoder rather than the decoder. The authors report consistent reductions in GFLOPs (22‑47%) and throughput improvements (up to 74%) on ADE20K, with optimal hyperparameters depending mainly on backbone pre‑training and target dataset.
LapMerge is a new differentiable token merging operator that partitions an ordered sequence into exactly N equal‑mass segments using a learned Laplace‑kernel density, then aggregates each segment with prefix/suffix scans, achieving an exact output budget in linear time. The method preserves information while allowing end‑to‑end differentiability and can change the budget without retraining. Experiments on vision and vision‑language tasks show LapMerge matches or outperforms existing reducers and can be up to two orders of magnitude faster for long sequences, especially when token order reflects spatial or temporal locality.
arXiv:2608.22526v1 Announce Type: new Abstract: We introduce RS$^3$-Prune, a training-free token-pruning recipe that instantiates as a small set of inference time hooks atop existing video object seg...
arXiv:2604. 25421v2 Announce Type: replace-cross Abstract: Federated fine-tuning provides a practical route to adapt large language models (LLMs) on edge devices without centralizing private data, yet in mobile deployments the training wall-clock is often bottlenecked by straggler-limited uplink communication under heterogeneous bandwidth and intermittent participation.
CatchUp is a training‑free framework that improves efficiency in large vision‑language models by depth‑aware re‑admission of visual tokens. It repeatedly evaluates query‑conditioned relevance, processes only selected tokens per decoder interval, and preserves deferred tokens for later re‑admission, using reference‑to‑current transition signals to approximate missed representation evolution. Experiments on three referring‑expression grounding datasets and six multimodal benchmarks show that CatchUp retains over 92% of full‑model performance with 88.9% fewer active tokens and reduces end‑to‑end latency by 8.0% at a 64‑token budget.
arXiv:2610.23455v1 Announce Type: cross Abstract: LLMs and LVLMs are increasingly deployed across heterogeneous hardware, ranging from data-center GPUs to resource-constrained edge devices. Their fix...
The paper introduces a token‑oriented semantic communication framework that transmits only task‑relevant image latents instead of full token embeddings, reducing communication cost and improving interoperability. It leverages a spatial alignment between vision transformer patch tokens and learned image compression latents, enabling token‑level relevance estimation and selective transmission. Experiments on ImageNet demonstrate a superior rate–accuracy trade‑off compared to existing semantic communication methods and hand‑crafted codecs.
LearnPrunerV2 is a fully learnable token‑pruning framework for Vision‑Language Models that replaces heuristic pruning with lightweight modules in every backbone layer. These modules estimate token importance by combining visual saliency and query relevance via a shadow token, revealing a consistent staircase redundancy pattern across layers. The method introduces Budget‑Deterministic Pruning to set fixed per‑layer budgets, achieving state‑of‑the‑art efficiency while preserving 99.8% of performance even after pruning 77.8% of visual tokens on LLaVA‑v1.5.