Risultati della ricerca - (endedededeendecoding OR (encoding OR Decoding)) (algorithmics OR (algorithm OR algorithm))

  1. 4321

    Automatic Pronunciation Error Detection and Correction of the Holy Quran's Learners Using Deep Learning di Abdelfattah, Abdullah, Khalil, Mahmoud I., Abbas, Hazem

    Pubblicazione 2025
    Sommario: “... Tasmeea algorithm; (2) 850+ hours of audio (~300K annotated utterances); (3) A novel ASR-based approach...”
    Testo
    Preprint
  2. 4322

    TrimTokenator: Towards Adaptive Visual Token Pruning for Large Multimodal Models di Zhang, Hao, Lyu, Mengsi, He, Chenrui, Ao, Yulong, Lin, Yonghua

    Pubblicazione 2025
    Sommario: “... usually encode visual inputs into dense token sequences, which are then concatenated with textual tokens...”
    Testo
    Preprint
  3. 4323
  4. 4324

    TransMPC: Transformer-based Explicit MPC with Variable Prediction Horizon di Wu, Sichao, Wu, Jiang, Cao, Xingyu, Zhang, Fawang, Yu, Guangyuan, Zhao, Junjie, Qu, Yue, Ma, Fei, Duan, Jingliang

    Pubblicazione 2025
    Sommario: “... TransMPC, a novel Transformer-based explicit MPC algorithm capable of generating highly accurate control...”
    Testo
    Preprint
  5. 4325

    Agents of Discovery di Diefenbacher, Sascha, Hallin, Anna, Kasieczka, Gregor, Krämer, Michael, Lauscher, Anne, Lukas, Tim

    Pubblicazione 2025
    Sommario: “..., these tools are typically special-purpose algorithms that rely, for example, on encoded physics knowledge...”
    Testo
    Preprint
  6. 4326

    DATE: Dynamic Absolute Time Enhancement for Long Video Understanding di Yuan, Chao, Yang, Yang, Yang, Yehui, Cheng, Zach

    Pubblicazione 2025
    Sommario: “... approaches typically adopt uniform frame sampling and rely on implicit position encodings to model temporal...”
    Testo
    Preprint
  7. 4327

    CollaPipe: Adaptive Segment-Optimized Pipeline Parallelism for Collaborative LLM Training in Heterogeneous Edge Networks di Chen, Jiewei, Deng, Xiumei, Xiong, Zehui, Guo, Shaoyong, Qiu, Xuesong, Wang, Ping, Niyato, Dusit

    Pubblicazione 2025
    Sommario: “... networks. In CollaPipe, the encoder part is adaptively partitioned into variable-sized segments...”
    Testo
    Preprint
  8. 4328

    Hallucination reduction with CASAL: Contrastive Activation Steering For Amortized Learning di Wannan, Yang, Qiu, Xinchi, Yu, Lei, Zhang, Yuchen, Yang, Aobo, Kokhlikyan, Narine, Cancedda, Nicola, Garcia-Olano, Diego

    Pubblicazione 2025
    Sommario: “... providing incorrect answers instead of admitting ignorance. Prior work has shown that models encode linear...”
    Testo
    Preprint
  9. 4329

    Partial Information Decomposition via Normalizing Flows in Latent Gaussian Distributions di Zhao, Wenyuan, Balachandran, Adithya, Tian, Chao, Liang, Paul Pu

    Pubblicazione 2025
    Sommario: “... as Gaussian PID (GPID). We propose a new gradient-based algorithm that substantially improves...”
    Testo
    Preprint
  10. 4330

    Automaton Constrained Q-Learning di Manganaris, Anastasios, Giammarino, Vittorio, Qureshi, Ahmed H.

    Pubblicazione 2025
    Sommario: “...-Learning (ACQL), an algorithm that addresses this gap by combining goal-conditioned value learning...”
    Testo
    Preprint
  11. 4331

    Xiaoice: Training-Free Video Understanding via Self-Supervised Spatio-Temporal Clustering of Semantic Features di Ji, Shihao, Song, Zihui

    Pubblicazione 2025
    Sommario: “...-trained VLMs with classic machine learning algorithms for pattern discovery. Our core idea is to reframe...”
    Testo
    Preprint
  12. 4332

    Vocab Diet: Reshaping the Vocabulary of LLMs via Vector Arithmetic di Reif, Yuval, Kaplan, Guy, Schwartz, Roy

    Pubblicazione 2025
    Sommario: “...Large language models (LLMs) often encode word-form variation (e.g., walk vs. walked) as linear...”
    Testo
    Preprint
  13. 4333

    Exploring Spiking Neural Networks for Binary Classification in Multivariate Time Series at the Edge di Ghawaly, James, Nicholson, Andrew, Schuman, Catherine, Diez, Dalton, Young, Aaron, Witherspoon, Brett

    Pubblicazione 2025
    Sommario: “... alarm rates. The approach uses the Evolutionary Optimization of Neuromorphic Systems (EONS) algorithm...”
    Testo
    Preprint
  14. 4334
  15. 4335

    Explaining and Mitigating Crosslingual Tokenizer Inequities di Arnett, Catherine, Chang, Tyler A., Biderman, Stella, Bergen, Benjamin K.

    Pubblicazione 2025
    Sommario: “...The number of tokens it takes to encode parallel text in different languages is known to vary...”
    Testo
    Preprint
  16. 4336

    Learning Reconfigurable Representations for Multimodal Federated Learning with Missing Data di Nguyen, Duong M., Hoang, Trong Nghia, Huynh, Thanh Trung, Nguyen, Quoc Viet Hung, Nguyen, Phi Le

    Pubblicazione 2025
    Sommario: “... on learnable client-side embedding controls that encode each client's data-missing patterns. These embeddings...”
    Testo
    Preprint
  17. 4337

    FoldPath: End-to-End Object-Centric Motion Generation via Modulated Implicit Paths di Rabino, Paolo, Tiboni, Gabriele, Tommasi, Tatiana

    Pubblicazione 2025
    Sommario: “... and welding. To realize effective automation, robust algorithms are essential for generating extended, object...”
    Testo
    Preprint
  18. 4338

    Influence of Data Dimensionality Reduction Methods on the Effectiveness of Quantum Machine Learning Models di Shinde, Aakash Ravindra, Nurminen, Jukka K.

    Pubblicazione 2025
    Sommario: “... machine algorithms, quantum data encoding methods, and data reduction methods. All these models were...”
    Testo
    Preprint
  19. 4339

    RAS: A Bit-Exact rANS Accelerator For High-Performance Neural Lossless Compression di Qin, Yuchao, Fan, Anjunyi, Yan, Bonan

    Pubblicazione 2025
    Sommario: “... Asymmetric Numeral System Acceleration System, a hardware architecture that integrates the rANS algorithm...”
    Testo
    Preprint
  20. 4340

    Factorization-in-Loop: Proximal Fill-in Minimization for Sparse Matrix Reordering di Li, Ziwei, Niu, Shuzi, Yuan, Tao, Li, Huiyuan, Wu, Wenjia

    Pubblicazione 2025
    Sommario: “... number of fill-ins. The reordering network utilizes a graph encoder to predict row or column node scores...”
    Testo
    Preprint