Results 11 to 20 of about 205,061 (335)
Symbolic Chain-of-Thought Distillation: Small Models Can Also “Think” Step-by-Step [PDF]
Annual Meeting of the Association for Computational Linguistics, 2023 Chain-of-thought prompting (e.g., “Let’s think step-by-ste”) primes large language models to verbalize rationalization for their predictions. While chain-of-thought can lead to dramatic performance gains, benefits appear to emerge only for sufficiently ...
Liunian Harold Li +5 more
semanticscholar +1 more source
International Journal of Coal Science & Technology, 2022 The development trend of Fischer–Tropsch (F–T) technology is to develop high value-added products. The separation of linear α-olefins with low cost is an effective method.
Zongchao Liu +8 more
doaj +1 more source
Dataset Distillation: A Comprehensive Review [PDF]
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023 Recent success of deep learning is largely attributed to the sheer amount of data used for training deep neural networks. Despite the unprecedented success, the massive data, unfortunately, significantly increases the burden on storage and transmission ...
Ruonan Yu, Songhua Liu, Xinchao Wang
semanticscholar +1 more source
Frontiers in Chemistry, 2022 Surfactants adsorption onto carbonate reservoirs would cause surfactants concentrations decrease in surfactant flooding, which would decrease surfactant efficiency in practical applications of enhanced oil recovery (EOR) processes.
Jinjian Hou +8 more
doaj +1 more source
A Comprehensive Survey of Dataset Distillation [PDF]
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023 Deep learning technology has developed unprecedentedly in the last decade and has become the primary choice in many application domains. This progress is mainly attributed to a systematic collaboration in which rapidly growing computing resources ...
Shiye Lei, Dacheng Tao
semanticscholar +1 more source
DistillSpec: Improving Speculative Decoding via Knowledge Distillation [PDF]
International Conference on Learning Representations, 2023 Speculative decoding (SD) accelerates large language model inference by employing a faster draft model for generating multiple tokens, which are then verified in parallel by the larger target model, resulting in the text generated according to the target
Yongchao Zhou +7 more
semanticscholar +1 more source
TRACT: Denoising Diffusion Models with Transitive Closure Time-Distillation [PDF]
arXiv.org, 2023 Denoising Diffusion models have demonstrated their proficiency for generative sampling. However, generating good samples often requires many iterations.
David Berthelot +8 more
semanticscholar +1 more source
Knowledge Distillation from A Stronger Teacher [PDF]
Neural Information Processing Systems, 2022 Unlike existing knowledge distillation methods focus on the baseline settings, where the teacher models and training strategies are not that strong and competing as state-of-the-art approaches, this paper presents a method dubbed DIST to distill better ...
Tao Huang +4 more
semanticscholar +1 more source
Fine-tuning Global Model via Data-Free Knowledge Distillation for Non-IID Federated Learning [PDF]
Computer Vision and Pattern Recognition, 2022 Federated Learning (FL) is an emerging distributed learning paradigm under privacy constraint. Data heterogeneity is one of the main challenges in FL, which results in slow convergence and degraded performance.
Lin Zhang +4 more
semanticscholar +1 more source
Towards Lossless Dataset Distillation via Difficulty-Aligned Trajectory Matching [PDF]
International Conference on Learning Representations, 2023 The ultimate goal of Dataset Distillation is to synthesize a small synthetic dataset such that a model trained on this synthetic set will perform equally well as a model trained on the full, real dataset.
Ziyao Guo +5 more
semanticscholar +1 more source