English
Related papers

Related papers: Sub-Token Routing in LoRA for Adaptation and Query…

200 papers

Modern language models are trained almost exclusively on token sequences produced by a fixed tokenizer, an external lossless compressor often over UTF-8 byte sequences, thereby coupling the model to that compressor. This work introduces…

Computation and Language · Computer Science 2026-05-15 Lin Zheng , Xinyu Li , Qian Liu , Xiachong Feng , Lingpeng Kong

Although LoRa is predominantly employed with the single-hop LoRaWAN protocol, recent advancements have extended its application to multi-hop mesh topologies. Designing efficient routing for LoRa mesh networks remains challenging due to…

Networking and Internet Architecture · Computer Science 2025-10-07 Nalith Udugampola , Xiaoyu Ai , Binghao Li , Henry Gong , Aruna Seneviratne

Learning, prediction, and compression are intimately connected: a model that accurately predicts the next symbol in a sequence can be coupled with a source coder to compress that sequence near its information-theoretic limit. When tokenized…

Information Theory · Computer Science 2026-05-05 Vishnu Teja Kunde , Jean-Francois Chamberland , Krishna R. Narayanan , Jamison Ebert

Encoder transformer models compress information from all tokens in a sequence into a single [CLS] token to represent global context. This approach risks diluting fine-grained or hierarchical features, leading to information loss in…

Computation and Language · Computer Science 2025-09-23 Asif Shahriar , Rifat Shahriyar , M Saifur Rahman

Large language model (LLM) tokenizers act as structured compressors: by mapping text to discrete token sequences, they determine token count (and thus compute and context usage) and the statistical structure seen by downstream models.…

Information Theory · Computer Science 2026-01-15 Mete Erdogan , Abhiram Gorle , Shubham Chandak , Mert Pilanci , Tsachy Weissman

Large language models rely on kv-caches to avoid redundant computation during autoregressive decoding, but as context length grows, reading and writing the cache can quickly saturate GPU memory bandwidth. Recent work has explored KV-cache…

Computation and Language · Computer Science 2026-02-10 Jian Chen , Zhuoran Wang , Jiayu Qin , Ming Li , Meng Wang , Changyou Chen , Yin Chen , Qizhen Weng , Yirui Liu

Large language models (LLMs) exhibit strong reasoning capabilities but typically require expensive post-training to reach high performance. Recent test-time alignment methods offer a lightweight alternative, but have been explored mainly…

Computation and Language · Computer Science 2026-03-20 Arushi Rai , Qiang Zhang , Hanqing Zeng , Yunkai Zhang , Dipesh Tamboli , Xiangjun Fan , Zhuokai Zhao , Lizhu Zhang

Low-Rank Adaptation (LoRA) is widely recognized for its parameter-efficient fine-tuning of large-scale neural models. However, standard LoRA independently optimizes low-rank matrices, which inherently limits its expressivity and…

Machine Learning · Computer Science 2025-06-23 Jun Qi , Chen-Yu Liu , Sabato Marco Siniscalchi , Chao-Han Huck Yang , Min-Hsiu Hsieh

Transformers have revolutionized Computer Vision (CV) through self-attention mechanisms. However, their complexity makes latent token representations difficult to interpret. We introduce ULTra, a framework for interpreting Transformer…

Computer Vision and Pattern Recognition · Computer Science 2026-01-06 Hesam Hosseini , Ghazal Hosseini Mighan , Amirabbas Afzali , Sajjad Amini , Amir Houmansadr

We propose Quantum-informed Tensor Adaptation (QuanTA), a novel, easy-to-implement, fine-tuning method with no inference overhead for large-scale pre-trained language models. By leveraging quantum-inspired methods derived from quantum…

Machine Learning · Computer Science 2025-11-11 Zhuo Chen , Rumen Dangovski , Charlotte Loh , Owen Dugan , Di Luo , Marin Soljačić

Vision-Language-Action (VLA) models have emerged as a powerful paradigm in Embodied AI. However, the significant computational overhead of processing redundant visual tokens remains a critical bottleneck for real-time robotic deployment.…

Robotics · Computer Science 2025-11-25 Juntao Gao , Feiyang Ye , Jing Zhang , Wenjing Qian

The advent of pre-trained large language models (LLMs) has revolutionized various natural language processing tasks. These models predominantly employ an auto-regressive decoding mechanism that utilizes Key-Value (KV) caches to eliminate…

Computation and Language · Computer Science 2024-06-12 Hao Yu , Zelan Yang , Shen Li , Yong Li , Jianxin Wu

Low-Rank Adaptation (LoRA) has become a widely adopted parameter-efficient fine-tuning method for large language models, with its effectiveness largely influenced by the allocation of ranks and scaling factors, as well as initialization.…

Computation and Language · Computer Science 2026-04-21 Weicheng Lin , Yi Zhang , Jiawei Dang , Liang-Jie Zhang

Recent work on KV cache quantization, culminating in TurboQuant, has approached the Shannon entropy limit for per-vector compression of transformer key-value caches. We observe that this limit applies to a strictly weaker problem than the…

Machine Learning · Computer Science 2026-04-20 Gregory Magarshak

Huge memory consumption has been a major bottleneck for deploying high-throughput large language models in real-world applications. In addition to the large number of parameters, the key-value (KV) cache for the attention mechanism in the…

Computation and Language · Computer Science 2024-06-05 Haoyi Wu , Kewei Tu

In fine-tuning large language models (LLMs), conserving computational resources while maintaining effectiveness and improving outcomes within the same computational constraints is crucial. The Low-Rank Adaptation (LoRA) strategy balances…

Machine Learning · Computer Science 2024-09-05 Xiaojun Xiao , Sen Shen , Qiming Bao , Hongfei Rong , Kairui Liu , Zhongsheng Wang , Jiamou Liu

Despite its huge number of variants, standard Low-Rank Adaptation (LoRA) is still a dominant technique for parameter-efficient fine-tuning (PEFT). Nonetheless, it faces persistent challenges, including the pre-selection of an optimal rank…

Computation and Language · Computer Science 2026-02-20 Ivan Vulić , Adam Grycner , Quentin de Laroussilhe , Jonas Pfeiffer

Language Models (LMs) recently incorporate mixture-of-experts layers consisting of a router and a collection of experts to scale up their parameter count given a fixed computational budget. Building on previous efforts indicating that…

Computation and Language · Computer Science 2024-09-24 Stefan Arnold , Marian Fietta , Dilara Yesilbas

Current Large Language Models (LLMs) face three major challenges: context length limitations, high inference costs, and catastrophic forgetting during continual learning. While Mixture-of-Experts (MoE) architectures mitigate some of these…

Machine Learning · Computer Science 2025-12-23 Zhongpan Tang

Approaches for compressing large-language models using low-rank decomposition have made strides, particularly with the introduction of activation and loss-aware SVD, which improves the trade-off between decomposition rank and downstream…

Machine Learning · Computer Science 2025-12-17 Sidhant Sundrani , Francesco Tudisco , Pasquale Minervini