Risultati della ricerca - Decoding (algorithm OR (algorithmicsics OR algorithmssicsics))

  1. 201

    Advancing Decoding Strategies: Enhancements in Locally Typical Sampling for LLMs di Sen, Jaydip, Sengupta, Saptarshi, Dasgupta, Subhasis

    Pubblicazione 2025
    Sommario: “...), focusing on enhancing the Locally Typical Sampling (LTS) algorithm. Traditional decoding methods...”
    Testo
    Preprint
  2. 202

    Recurrent Drafter for Fast Speculative Decoding in Large Language Models di Cheng, Yunfei, Zhang, Aonan, Zhang, Xuanyu, Wang, Chong, Wang, Yi

    Pubblicazione 2024
    Sommario: “... on LLM's hidden states, (2) applying a dynamic tree attention algorithm over beam search results...”
    Testo
    Preprint
  3. 203

    Attention-Constrained Inference for Robust Decoder-Only Text-to-Speech di Wang, Hankun, Du, Chenpeng, Guo, Yiwei, Wang, Shuai, Chen, Xie, Yu, Kai

    Pubblicazione 2024
    Sommario: “... algorithm and then applies constraining masks on AEAMs. Our experimental results on decoder-only TTS model...”
    Testo
    Preprint
  4. 204
  5. 205

    Efficient and Asymptotically Unbiased Constrained Decoding for Large Language Models di Ye, Haotian, Jain, Himanshu, You, Chong, Suresh, Ananda Theertha, Lin, Haowei, Zou, James, Yu, Felix

    Pubblicazione 2025
    Sommario: “.... This paper introduces Dynamic Importance Sampling for Constrained Decoding (DISC) with GPU-based Parallel...”
    Testo
    Preprint
  6. 206

    Biomimetic Metamaterial-based Interface for Decoding Heterogeneous Mechanodermal Activity di Xu, Muzi, Zhang, Jiaqi, Dong, Chaoqun, Zhang, Zibo, Li, Duanyang, Yi, Wentian, Zou, Miaomiao, Tang, Chenyu, Malliaras, George G., Occhipinti, Luigi G.

    Pubblicazione 2025
    Sommario: “... various scenarios. Combined with bespoke algorithms, the wireless BMMI device decodes MDA accurately...”
    Testo
    Preprint
  7. 207

    Break the Sequential Dependency of LLM Inference Using Lookahead Decoding di Fu, Yichao, Bailis, Peter, Stoica, Ion, Zhang, Hao

    Pubblicazione 2024
    Sommario: “..., parallel decoding algorithm that accelerates LLM decoding without needing auxiliary models or data stores...”
    Testo
    Preprint
  8. 208

    JDEC: JPEG Decoding via Enhanced Continuous Cosine Coefficients di Han, Woo Kyoung, Im, Sunghoon, Kim, Jaedeok, Jin, Kyong Hwan

    Pubblicazione 2024
    Sommario: “... representation with continuous cosine formulation. The JPEG algorithm significantly quantizes discrete cosine...”
    Testo
    Preprint
  9. 209
  10. 210

    Speculate, then Collaborate: Fusing Knowledge of Language Models during Decoding di Wang, Ziyao, Azmat, Muneeza, Li, Ang, Horesh, Raya, Yurochkin, Mikhail

    Pubblicazione 2025
    Sommario: “... introduce a novel Collaborative Speculative Decoding (CoSD) algorithm that enables efficient LLM knowledge...”
    Testo
    Preprint
  11. 211

    SACM: SEEG-Audio Contrastive Matching for Chinese Speech Decoding di Wang, Hongbin, Jia, Zhihong, Shen, Yuanzhong, Wang, Ziwei, Li, Siyang, Shu, Kai, Hu, Feng, Wu, Dongrui

    Pubblicazione 2025
    Sommario: “... decoding algorithms. Stereo-electroencephalography (SEEG) and synchronized audio data were collected from...”
    Testo
    Preprint
  12. 212

    Optimizing Decoding Paths in Masked Diffusion Models by Quantifying Uncertainty di Chen, Ziyu, Jiang, Xinbei, Sun, Peng, Lin, Tao

    Pubblicazione 2025
    Sommario: “... propose two algorithms designed to optimize the decoding path: a post-hoc selection method and a real-time...”
    Testo
    Preprint
  13. 213

    Improving Temporal Action Segmentation via Constraint-Aware Decoding di Ee, Yeo Keat, Roy, Debaditya, Li, Chen, Zhang, Hao, Fernando, Basura

    Pubblicazione 2026
    Sommario: “... modified Viterbi decoding algorithm, allowing inference-time refinement without retraining or added model...”
    Testo
    Preprint
  14. 214

    Multilingual Contrastive Decoding via Language-Agnostic Layers Skipping di Zhu, Wenhao, Liu, Sizhe, Huang, Shujian, She, Shuaijie, Wendler, Chris, Chen, Jiajun

    Pubblicazione 2024
    Sommario: “...Decoding by contrasting layers (DoLa), is designed to improve the generation quality of large...”
    Testo
    Preprint
  15. 215

    Dependency-Guided Parallel Decoding in Discrete Diffusion Language Models di Ringel, Liran, Ali, Ameen, Romano, Yaniv

    Pubblicazione 2026
    Sommario: “... selection algorithm identifies positions with bounded cumulative dependency for simultaneous unmasking...”
    Testo
    Preprint
  16. 216

    OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure di Wang, Jikai, Su, Yi, Li, Juntao, Xia, Qingrong, Ye, Zi, Duan, Xinyu, Wang, Zhefeng, Zhang, Min

    Pubblicazione 2024
    Sommario: “..., an algorithm to construct adaptive and scalable draft trees. It searches the optimal tree structure...”
    Testo
    Preprint
  17. 217

    Power and Limitations of Linear Programming Decoder for Quantum LDPC Codes di Gu, Shouzhen, Soleimanifar, Mehdi

    Pubblicazione 2025
    Sommario: “...Decoding quantum error-correcting codes is a key challenge in enabling fault-tolerant quantum...”
    Testo
    Preprint
  18. 218

    Wide-In, Narrow-Out: Revokable Decoding for Efficient and Effective DLLMs di Hong, Feng, Yu, Geng, Ye, Yushi, Huang, Haicheng, Zheng, Huangjie, Zhang, Ya, Wang, Yanfeng, Yao, Jiangchao

    Pubblicazione 2025
    Sommario: “...-In, Narrow-Out (WINO), a training-free decoding algorithm that enables revokable decoding in DLLMs. WINO...”
    Testo
    Preprint
  19. 219

    Algorithmic Improvements to List Decoding of Folded Reed-Solomon Codes di Ashvinkumar, Vikrant, Habib, Mursalin, Srivastava, Shashank

    Pubblicazione 2025
    Sommario: “... decoding capacity. In this work, we give improved deterministic and randomized algorithms for list decoding...”
    Testo
    Preprint
  20. 220

    Linear-Time Encodable and Decodable Quantum Error-Correcting Codes di Wills, Adam, Lin, Ting-Chun, Zhang, Rachel Yun, Hsieh, Min-Hsiu

    Pubblicazione 2026
    Sommario: “.... The classical decoding algorithms also run in logarithmic depth and use $\mathcal{O}(n \log n)$ gates...”
    Testo
    Preprint