I and J

By W. Marsden. 2012

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Sequence to Sequence Learning with Neural Networks (2014, background): “While it could work in pr i ciple since the RNN is provided with all the relevant information, it would be difficult to tr ain the RNNs due to the resulting long term dependencies [14, 4] (figure 1) [16, 15].”
One billion word benchmark for measuring progress in statistical language modeling (2013, neutral): “…mention some of the most popular ones: • N-grams with Modified Kneser-Ney smoothing (Chen and Goodman, 1996) • Cache (Jelinek et al., 1991) • Class-based (Brown et al., 1992) • Maximum entropy (Rosenfeld, 1994) • Structured (Chelba and Jelinek, 2000) • Neural net based (Bengio et al., 2003) •…”
Very Deep Convolutional Networks for Large-Scale Image Recognition (2014, methodology): “Following [1, 9 , 25], on Caltech-101 we generated 3 random splits into training and test data, so that each split con ains 30 training images per class, and up to 50 test images per class.”
Direct Preference Optimization: Your Language Model is Secretly a Reward Model (2023, methodology): “, which has also been adopted in subsequent work [35, 1, 23].”
Adaptive Computation Time for Recurrent Neural Networks (2016, background): “make every network very deep and design its architecture in such a way as to mitigate the vanishing gradient problem [13] associated with long chains of iteration [29, 17].”
The Capacity for Moral Self-Correction in Large Language Models (2023, background): “The efficacy of CoT prompting on model capabilities on complex reasoning tasks emerges [18, 58] with model size [28, 57, 60] which is consistent with our results.”
Beyond Skip Connections: Top-Down Modulation for Object Detection (2016, background): “Contemporary works [5, 12, 32, 43], which incorporate top-down feedback loop in ConvNets, have used ‘unrolling’ to train these networks stagewise.”
Convolutional Pose Machines (2016, background): “Networks with many layers have been shown to be prone to the problem of vanishing gradients[4, 5, 9, 10] during training.”
Training Generative Adversarial Networks with Limited Data (2020, background): “Bora et al. [4] consider a similar problem in training GANs under corrupted measurements, and show that the training implicitly undoes the corruptions and finds the correct distribution, as long as the corruption process is represented by an invertible transformation of probability distributions ove”
ImageBART: Bidirectional Context with Multinomial Diffusion for Autoregressive Image Synthesis (2021, background): “The success of this approach is evident in domains as diverse as language modeling [7], music generation [16], neural machine translation [46, 76], and (conditional) image synthesis [54, 8].”
A Note on Data Biases in Generative Models (2020, background): “1 which demonstrates even more drastic improvements when transitioning from the AnimalFaces [12] dataset to its corresponding high-quality version, AnimalFacesHQ [9].”
node2vec: Scalable Feature Learning for Networks (2016, background): “The subgraph corresponds to the graph induced by nodes for which we could obtain labels from the hallmark gene sets [19] and represent biological states.”
A Simple Language Model for Task-Oriented Dialogue (2020, methodology): “[37] used GPT-2 to learn a response generator conditioned on oracle dialogue acts which did not evaluate on dialogue state tracking.”
Learning multiple visual domains with residual adapters (2017, background): “The first one [37, 33] is to freeze the network parameters for the old tasks and learn a new task by adding extra parameters.”
Dense-Captioning Events in Videos (2017, methodology): “In an effort to capture longer events, MSRVTT [20], MSVD [2] and YouCook [3] collected a dataset with slightly longer length, at the cost of a few descriptions than previous datasets.”
MovieQA: Understanding Stories in Movies through Question-Answering (2015, background): “For TV series in particular, the majority of work aims at recognizing and tracking characters in the videos [2, 4, 24, 31].”
The Ubuntu Dialogue Corpus: A Large Dataset for Research in Unstructured Multi-Turn Dialogue Systems (2015, background): “Switchboard [8] Human-human Various 2,400 — 3,000,000 Telephone conversations spoken on pre-specified topics DSTC1 [32] Human-computer State 15,000 210,000 Bus ride information spoken tracking system DSTC2 [10] Human-computer State 3,000 24,000 — Restaurant booking spoken tracking system DSTC3 [9] H”
Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019 (2019, background): “[92] 20 stocks in S&P500 2010-2015 Price data AE + LSTM Weekly Returns -”