Results 11 to 20 of about 28,651,668 (327)
MBR and QE Finetuning: Training-time Distillation of the Best and Most Expensive Decoding Methods [PDF]
International Conference on Learning Representations, 2023 Recent research in decoding methods for Natural Language Generation (NLG) tasks has shown that MAP decoding is not optimal, because model probabilities do not always align with human preferences.
M. Finkelstein, Markus Freitag
semanticscholar +1 more source
Objective comparison of methods to decode anomalous diffusion [PDF]
Nature Communications, 2021 AbstractDeviations from Brownian motion leading to anomalous diffusion are found in transport dynamics from quantum physics to life sciences. The characterization of anomalous diffusion from the measurement of an individual trajectory is a challenging task, which traditionally relies on calculating the trajectory mean squared displacement.
Diego Krapf +37 more
openaire +10 more sources
Mainstream encoding–decoding methods of DNA data storage
CCF Transactions on High Performance Computing, 2022 DNA storage is a new digital data storage technology based on specific encoding and decoding methods between 0 and 1 binary codes of digital data and A-T-C-G quaternary codes of DNAs, which and is expected to develop into a major data storage form in the
Chenyang Wang +12 more
semanticscholar +1 more source
PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models [PDF]
Conference on Empirical Methods in Natural Language Processing, 2021 Large pre-trained language models for textual data have an unconstrained output space; at each decoding step, they can produce any of 10,000s of sub-word tokens. When fine-tuned to target constrained formal languages like SQL, these models often generate
Torsten Scholak +2 more
semanticscholar +1 more source
A Novel Decoding Method for the Erasure Codes [PDF]
Security and Communication Networks, 2021 The erasure codes are widely used in the distributed storage with low redundancy compared to the replication method. However, the current research studies about the erasure codes mainly focus on the encoding methods, while there are few studies on the decoding methods. In this paper, a novel erasure decoding method is proposed; it is a general decoding
Dan Tang, Hongliang Cai
openaire +2 more sources
Accelerating Transformer Inference for Translation via Parallel Decoding [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 Autoregressive decoding limits the efficiency of transformers for Machine Translation (MT). The community proposed specific network architectures and learning-based methods to solve this issue, which are expensive and require changes to the MT model ...
Andrea Santilli +6 more
semanticscholar +1 more source
Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning [PDF]
Conference on Empirical Methods in Natural Language Processing, 2023 Despite their impressive performance, large language models (LMs) still struggle with reliably generating complex output structures when not finetuned to follow the required output format exactly. To address this issue, grammar-constrained decoding (GCD)
Saibo Geng +3 more
semanticscholar +1 more source
Cascade Decoder: A Universal Decoding Method For Biomedical Image Segmentation [PDF]
2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019), 2019 The Encoder-Decoder architecture is a main stream deep learning model for biomedical image segmentation. The encoder fully compresses the input and generates encoded features, and the decoder then produces dense predictions using encoded features. However, decoders are still under-explored in such architectures.
Danny Z. Chen +5 more
openaire +3 more sources
New Rotor Position Redundancy Decoding Method Based on Resolver Decoder
Micromachines, 2022 In view of the frequent safety problems of electric vehicles, the research on accurately obtaining the rotor position of the motor through the resolver is an important means to improve the functional safety of the system. The commonly used resolver decoding method involves the resolver decoding chip method and software decoding method, but few studies ...
Anbo Yu +5 more
openaire +4 more sources
Comparison of Diverse Decoding Methods from Conditional Language Models [PDF]
Annual Meeting of the Association for Computational Linguistics, 2019 While conditional language models have greatly improved in their ability to output high quality natural language, many NLP applications benefit from being able to generate a diverse set of candidate sequences. Diverse decoding strategies aim to, within a
Daphne Ippolito +4 more
semanticscholar +1 more source