Risultati della ricerca - decoding (algorithmics OR (algorithm OR (Algorithmsssics OR (algorithm OR algorithm))))

EcoServe: Enabling Cost-effective LLM Serving with Proactive Intra- and Inter-Instance Orchestration di Du, Jiangsu, Zhang, Hongbin, Wei, Taosheng, Zheng, Zhenyi, Wu, Kaiyi, Chen, Zhiguang, Lu, Yutong

Sommario: “...Existing LLM serving strategies can be categorized based on whether prefill and decode phases...”
Testo

AMBEDKAR-A Multi-level Bias Elimination through a Decoding Approach with Knowledge Augmentation for Robust Constitutional Alignment of Language Models di Mukhopadhyay, Snehasis, Kasat, Aryan, Dubey, Shivam, Karthikeyan, Rahul, Sood, Dhruv, Jain, Vinija, Chadha, Aman, Das, Amitava

Sommario: “... speculative decoding algorithm that proactively reduces casteist and communal bias during generation...”
Testo

GapDNER: A Gap-Aware Grid Tagging Model for Discontinuous Named Entity Recognition di Yang, Yawen, Ma, Fukun, Meng, Shiao, Liu, Aiwei, Wen, Lijie

Sommario: “... or internal tokens, which face challenges of error propagation and decoding ambiguity due to the wide variety...”
Testo

Semantic representations emerge in biologically inspired ensembles of cross-supervising neural networks di Urbach, Roy, Schneidman, Elad

Sommario: “... or decoding, but typically rely on biologically-implausible implementations. We suggest that interactions...”
Testo

Leak@$k$: Unlearning Does Not Make LLMs Forget Under Probabilistic Decoding di Reisizadeh, Hadi, Ruan, Jiajun, Chen, Yiwei, Pal, Soumyadeep, Liu, Sijia, Hong, Mingyi

Sommario: “... deterministic (greedy) decoding often suggest successful knowledge removal using standard benchmarks, we show...”
Testo

Support Recovery in One-bit Compressed Sensing with Near-Optimal Measurements and Sublinear Time di Li, Xiaxin, Mazumdar, Arya

Sommario: “... methods minimize the number of measurements but often incur $Ω(n)$ decoding complexity, limiting large...”
Testo

AI/ML based Joint Source and Channel Coding for HARQ-ACK Payload di Doshi, Akash, Sen, Pinar, Ivanov, Kirill, Yang, Wei, Namgoong, June, Wang, Runxin, Wang, Rachel, Yoo, Taesang, Jiang, Jing, Ji, Tingfang

Sommario: “..., we learn a transformer-based encoder using a novel "free-lunch" training algorithm and propose per...”
Testo

Toward Robust, Reproducible, and Widely Accessible Intracranial Language Brain-Computer Interfaces: A Comprehensive Review of Neural Mechanisms, Hardware, Algorithms, Evaluation, C... di He, Dongyi, Siok, Wai Ting, Wang, Nizhuan

Sommario: “... by fragmented evidence and unresolved design trade-offs across neuroscience, hardware, algorithm, evaluation...”
Testo

Polar Complexity: A New Descriptive Complexity with Applications to Source and Joint Source-Channel Coding di Yao, Xinyuanmeng, Ma, Xiao

Sommario: “... compression and successive cancellation decoding (SCD), and define the polar complexity of a sequence...”
Testo

KVDrive: A Holistic Multi-Tier KV Cache Management System for Long-Context LLM Inference di Lin, Jian, Mi, Jiazhi, Hong, Zicong, Wang, Haodong, Liu, Qianli, Zhang, Haodyue, Li, Peng, Guo, Song

Sommario: “... critical entries during decoding, but this strategy quickly hits a ceiling: sparsity cannot be pushed...”
Testo

Shared autonomy between human electroencephalography and TD3 deep reinforcement learning: A multi‐agent copilot approach di Chun‐Ren Phang, Akimasa Hirata

Sommario: “... Deep reinforcement learning (RL) algorithms enable the development of fully autonomous agents that can...”
Testo

Roll Out and Roll Back: Diffusion LLMs are Their Own Efficiency Teachers di Zeng, Fanqin, Hong, Feng, Yu, Geng, Zheng, Huangjie, Cao, Xiaofeng, Zhang, Ya, Han, Bo, Wang, Yanfeng, Yao, Jiangchao

Sommario: “... still face a severe quality-speed trade-off: accelerating decoding by revealing multiple tokens often...”
Testo

A Spiking Neural Network based on Neural Manifold for Augmenting Intracortical Brain-Computer Interface Data di Zheng, Shengjie, Li, Wenyi, Qian, Lang, He, Chenggang, Li, Xiaojian

Sommario: “... of the algorithms oriented to general artifi-cial intelligence because it borrows the neural information processing...”
Testo

Asynchronous Multiuser Detection for SCMA with Unknown Delays: A Compressed Sensing Approach di Wheeler, Dylan, Tripp, Erin E., Natarajan, Balasubramaniam

Sommario: “... access system. The proposed technique can be used with nearly all of the numerous decoding algorithms...”
Testo

Foundational Policy Acquisition via Multitask Learning for Motor Skill Generation di Yamamori, Satoshi, Morimoto, Jun

Sommario: “...In this study, we propose a multitask reinforcement learning algorithm for foundational policy...”
Testo

One-Bit-Aided Modulo Sampling for DOA Estimation di Zhang, Qi, Zhu, Jiang, Qu, Fengzhong, Soh, De Wen

Sommario: “.... The estimate of the normalized covariance matrix is used to implement blind integer-forcing (BIF) decoder...”
Testo

Optimal Construction of N-bit-delay Almost Instantaneous Fixed-to-Variable-Length Codes di Sugiura, Ryosuke, Nishino, Masaaki, Yasuda, Norihito, Kamamoto, Yutaka, Moriya, Takehiro

Sommario: “...-length (AIFV) codes, the general form of binary codes we can make when finite bits of decoding delay...”
Testo

Living Scenes: Multi-object Relocalization and Reconstruction in Changing 3D Environments di Zhu, Liyuan, Huang, Shengyu, Schindler, Konrad, Armeni, Iro

Sommario: “... representation in a single encoder-decoder network, trained on synthetic data. This representation enables us...”
Testo

Secure ISAC MIMO Systems: Exploiting Interference With Bayesian Cramér-Rao Bound Optimization di Su, Nanchi, Liu, Fan, Masouros, Christos, Alexandropoulos, George C., Xiong, Yifeng, Zhang, Qinyu

Sommario: “... their decoding probability, thus enhancing the ISAC's system physical-layer security (PLS) capability. To tackle...”
Testo

MISC: Ultra-low Bitrate Image Semantic Compression Driven by Large Multimodal Model di Li, Chunyi, Lu, Guo, Feng, Donghui, Wu, Haoning, Zhang, Zicheng, Liu, Xiaohong, Zhai, Guangtao, Lin, Weisi, Zhang, Wenjun

Sommario: “... become a highly demanding topic. However, existing compression algorithms must sacrifice either...”
Testo