Research
18 entries
| Entry | Summary | As of |
|---|---|---|
| Attention Is All You Needresearch paper | The Transformer paper proposes an encoder-decoder architecture based on attention without recurrent or convolutional sequence layers. Its design enables greater training parallelism while modeling token-to-token interactions directly. | 2026-07-13 |
| BERTresearch paper | BERT pretrains deep bidirectional Transformer representations using masked language modeling. Fine-tuning the same pretrained encoder produced strong reported results across multiple language-understanding tasks. | 2026-07-13 |
| Computing Machinery and Intelligenceresearch paper | Alan Turing's 1950 paper asks whether machines can think and reframes the question through an imitation game. The paper also anticipates objections and discusses learning machines rather than defining intelligence through one internal mechanism. | 2026-07-13 |
| DOAJ open article metadataopen research infrastructure | DOAJ makes article-level metadata available under the CC0 waiver. Individual articles indexed by DOAJ retain the license selected by their authors or publishers, which must be checked item by item. | 2026-07-13 |
| Denoising Diffusion Probabilistic Modelsresearch paper | The DDPM paper presents high-quality image synthesis through learned reversal of a diffusion process. Its formulation connects denoising score matching with a probabilistic latent-variable objective. | 2026-07-13 |
| Direct Preference Optimizationresearch paper | DPO derives a direct preference-learning objective from the relationship between a policy and implicit reward. The method avoids training a separate reward model and running reinforcement-learning optimization in its core pipeline. | 2026-07-13 |
| Generative Adversarial Networksresearch paper | GANs train a generator and discriminator in a two-player minimax framework. The generator learns through the discriminator's signal rather than an explicitly specified likelihood for each output. | 2026-07-13 |
| Highly accurate protein structure prediction with AlphaFoldresearch paper | The AlphaFold paper reports a neural system achieving high accuracy in the CASP14 protein-structure prediction assessment. The architecture incorporates evolutionary, geometric, and physical reasoning signals while outputting confidence estimates. | 2026-07-13 |
| ImageNet Classification with Deep Convolutional Neural Networksresearch paper | AlexNet reports a deep convolutional network trained for large-scale ImageNet classification. Its results helped demonstrate the combined value of learned representations, GPUs, large datasets, and regularization. | 2026-07-13 |
| Language Models are Few-Shot Learnersresearch paper | The GPT-3 paper studies autoregressive language-model scaling and task performance from instructions or examples in context. The work reports few-shot behavior without gradient updates for each evaluated task. | 2026-07-13 |
| Learning representations by back-propagating errorsresearch paper | The 1986 Nature paper demonstrates backpropagation for learning internal representations in multilayer networks. Its importance lies in efficiently assigning error signals through composed differentiable layers. | 2026-07-13 |
| Long Short-Term Memoryresearch paper | The LSTM paper introduces gated recurrent units designed to preserve learning signals over longer intervals. Input, output, and memory mechanisms address failure modes of conventional recurrent learning. | 2026-07-13 |
| OpenAlex research graphopen research infrastructure | OpenAlex makes its website, API, and data snapshot available without charge and dedicates its data to CC0. The CC0 status applies to OpenAlex data; it does not erase copyright or license terms attached to the scholarly works described by that metadata. | 2026-07-13 |
| OpenCitations open citation dataopen research infrastructure | OpenCitations publishes the data in its datasets under a Creative Commons CC0 public-domain dedication. Citation metadata describes relationships among works and does not grant republication rights to the cited papers' full text. | 2026-07-13 |
| Retrieval-Augmented Generationresearch paper | The RAG paper combines parametric sequence generation with retrieved documents from a non-parametric index. It evaluates knowledge-intensive tasks where access to external passages can support generation. | 2026-07-13 |
| Sequence to Sequence Learning with Neural Networksresearch paper | Sequence-to-sequence learning uses an encoder and decoder to map variable-length input sequences to output sequences. The reported system uses multilayer LSTMs and reverses source sentences to improve optimization. | 2026-07-13 |
| The Perceptronresearch paper | Rosenblatt's perceptron paper presents a probabilistic model for information storage and organization. The perceptron became foundational to the history of learned linear decision rules and neural-network research. | 2026-07-13 |
| Training language models to follow instructions with human feedbackresearch paper | The InstructGPT paper applies demonstrations and preference feedback to align model outputs with user intent. It reports that a smaller aligned model can be preferred to a much larger base model on the studied prompts. | 2026-07-13 |