Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks390 from public data
- Learning to rank2,499
It is proposed that learning to rank should adopt the listwise approach in which lists of objects are used as 'instances' in learning, and introduces two probability models, respectively referred to as permutation probability and top k probability, to define a listwise loss function for learning.
- BioGPT: generative pre-trained transformer for biomedical text generation and mining1,580
This paper proposes BioGPT, a domain-specific generative Transformer language model pre-trained on large-scale biomedical literature and evaluates it on six biomedical natural language processing tasks and demonstrates that the model outperforms previous models on most tasks.
- 733
- 636
- 559
- LETOR: Benchmark Dataset for Research on Learning to Rank for Information Retrieval528
This paper has constructed a benchmark dataset referred to as LETOR, derived the LETOR data from the existing data sets widely used in IR, namely, OHSUMED and TREC data and provided the results of several state-ofthe-arts learning to rank algorithms on the data.
- LETOR: A benchmark collection for research on learning to rank for information retrieval527
The details of the LETOR collection are described and it is shown how it can be used in different kinds of researches, and several state-of-the-art learning to rank algorithms on LETOR are compared.
- 521
- 513
- 499
- NaturalSpeech 3: Zero-Shot Speech Synthesis with Factorized Codec and Diffusion Models447
This work proposes NaturalSpeech 3, a TTS system with novel factorized diffusion models to generate natural speech in a zero-shot way that outperforms the state-of-the-art TTS systems on quality, similarity, prosody, and intelligibility, and achieves on-par quality with human recordings.
- NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers431
This paper develops NaturalSpeech 2, a TTS system that leverages a neural audio codec with residual vector quantizers to get the quantized latent vectors and uses a diffusion model to generate these latent vectors conditioned on text input.
- 427
- AdaRNN389
This paper proposes Adaptive RNNs (AdaRNN) to tackle the TCS problem by building an adaptive model that generalizes well on the unseen test data and proposes Temporal Distribution Characterization to better characterize the distribution information in the TS.
- Image-to-Image Translation: Methods and Applications377
The key techniques of the existing I2I works are analyzed and clarified to clarify the main progress the community has made and point out remaining challenges in related fields.
- NaturalSpeech: End-to-End Text-to-Speech Synthesis With Human-Level Quality359
A variational auto-encoder (VAE) for end-to-end text-to-waveform generation, with several key modules to enhance the capacity of the prior from text and reduce the complexity of the posterior from speech, including phoneme pre-training, differentiable duration modeling, bidirectional prior/posterior modeling, and a memory mechanism in VAE.
- An Empirical Study on Learning to Rank of Tweets 1310
This paper proposes a new ranking strategy which uses not only the content relevance of a tweet, but also the account authority and tweet-specific features such as whether a URL link is included in the tweet.
- 308
- Feature selection for ranking286
This paper proposes a new feature selection method that uses its value to rank the training instances, and defines the ranking accuracy in terms of a performance measure or a loss function as the importance of the feature.
- A general approximation framework for direct optimization of information retrieval measures285
A general framework for direct optimization of IR measures, which enjoys several theoretical advantages, and experiments on benchmark datasets show that the algorithms deduced from the framework are very effective when compared to existing methods.
- FRank277
An algorithm named FRank is proposed based on a generalized additive model for the sake of minimizing the fedelity loss and learning an effective ranking function and the experimental results show that the proposed algorithm outperforms other learning-based ranking methods on both conventional IR problem and Web search.
- MusicBERT: Symbolic Music Understanding with Large-Scale Pre-Training224
This paper develops MusicBERT, a large-scale pre-trained model for music understanding that contains more than 1 million music songs and designs several mechanisms, including OctupleMIDI encoding and bar-level masking strategy, to enhance pre-training with symbolic music data.
- Deliberation Networks: Sequence Generation Beyond One-Pass Decoding223
This work introduces the deliberation process into the encoder-decoder framework and proposes deliberation networks for sequence generation and establishes a new state-of-the-art BLEU score on the WMT 2014 English-to-French translation task.
- Efficient Training of BERT by Progressively Stacking198
This paper explores an efficient training method for the state-of-the-art bidirectional Transformer (BERT) model and pro-poses the stacking algorithm to transfer knowledge from a shallow model to a deep model; then the algorithm is applied progressively to accelerate BERT training.
- A Study of Reinforcement Learning for Neural Machine Translation192
A systematic study on how to train better NMT models using reinforcement learning, providing a comprehensive comparison of several important factors and proposing a new method to leverage RL to further boost the performance of NMT systems trained with source/target monolingual data.
- 189
- Supervised rank aggregation188
Experimental results on meta-searches show that Supervised Rank Aggregation can significantly outperform existing unsupervised methods and it is proved that the optimization problem can be transformed into that of Semidefinite Programming and solve it efficiently.
- 179
- Web image clustering by consistent utilization of visual features and surrounding texts171
A novel method named consistent bipartite graph co-partitioning is proposed, which can cluster Web images based on the consistent fusion of the information contained in both low-level features and surrounding texts and can be efficiently solved by semi-definite programming (SDP).
- Query-level loss functions for information retrieval168
A query-level loss function based on the cosine similarity between a ranking list and the corresponding ground truth is proposed and a coordinate descent algorithm is designed, referred to as RankCosine, which utilizes the proposed loss function to create a generalized additive ranking model.
- Multi-Armed Bandit with Budget Constraint and Variable Costs166
It is shown that when applying the proposed algorithms to a previous setting with fixed costs, one can improve the previously obtained regret bound, and results on real-time bidding in ad exchange verify the effectiveness of the algorithms and are consistent with the theoretical analysis.
- Non-Autoregressive Machine Translation with Auxiliary Regularization165
This paper proposes to address the issues of repeated translations and incomplete translations in NAT models by improving the quality of decoder hidden representations via two auxiliary regularization terms in the training process of an NAT model.
- Query dependent ranking using K-nearest neighbor165
This paper proposes a K-Nearest Neighbor (KNN) method for query-dependent ranking, and proves a theory which indicates that the approximations are accurate in terms of difference in loss of prediction, if the learning algorithm used is stable with respect to minor changes in training examples.
- 158
- Global Ranking Using Continuous Conditional Random Fields155
The paper shows how the Continuous CRF method can perform global ranking better than baselines and can naturally represent the content information of objects as well as the relation information between objects, necessary for global ranking.
- 152
- Conditional Image-to-Image Translation149
This paper twists two conditional translation models together for inputs combination and reconstruction while preserving domain independent features and carries out experiments on men's faces from-to women's faces translation and edges to shoes&bags translations.
- PopMAG148
A novel MUlti-track MIDI representation (MuMIDI), which enables simultaneous multi-track generation in a single sequence and explicitly models the dependency of the notes from different tracks and is called PopMAG, which largely outperforms other state-of-the-art music accompaniment generation models and multi- track MIDI representations in terms of subjective and objective metrics.
- Deep Learning for Prediction of the Air Quality Response to Emission Changes147
A novel method that combines deep-learning approaches with chemical indicators of pollutant formation to quickly estimate the coefficients of air quality response functions using ambient concentrations of 18 chemical indicators simulated with a comprehensive atmospheric chemical transport model (CTM).
- Adversarial Neural Machine Translation139
Adversarial-NMT is a new learning paradigm for Neural Machine Translation that can achieve significantly better translation quality than several strong baselines and is inspired by the recent success of generative adversarial networks.
- Soft Contextual Data Augmentation for Neural Machine Translation137
This work softly augments a randomly chosen word in a sentence by its contextual mixture of multiple related words, replacing the one-hot representation of a word by a distribution (provided by a language model) over the vocabulary.
- MBNET: MOS Prediction for Synthesized Speech with Mean-Bias Network136
MBNet is proposed, a MOS predictor with a mean subnet and a bias subnet to better utilize every judge score in MOS datasets, where the mean subnets is used to predict the mean score of each utterance similar to that in previous works, and the bias subnets to predictThe bias score and capture the personal preference of individual judges.
- A Survey on Non-Autoregressive Generation for Neural Machine Translation and Beyond135
A systematic survey with comparisons and discussions of various non-autoregressive translation (NAT) models from different aspects to help researchers capture the latest progress in NAR generation, inspire the design of advanced NAR models and algorithms, and enable industry practitioners to choose appropriate solutions for their applications.
- Non-Autoregressive Neural Machine Translation with Enhanced Decoder Input135
This paper proposes two methods to enhance the decoder inputs so as to improve NAT models, one directly leverages a phrase table generated by conventional SMT approaches to translate source tokens to target tokens, and the other transforms source-side word embeddings to target-side words through sentence-level alignment and word-level adversary learning.
- Adaptive Memory Networks With Self-Supervised Learning for Unsupervised Anomaly Detection129
A novel approach called Adaptive Memory Network with Self-supervised Learning (AMSL) is proposed to address challenges and enhance the generalization ability in unsupervised anomaly detection and is significantly improves the performance compared to other state-of-the-art methods.
- Layer-Wise Coordination between Encoder and Decoder for Neural Machine Translation127
The concept of layer-wise coordination for NMT is proposed, which explicitly coordinates the learning of hidden representations of the encoder and decoder together layer by layer, gradually from low level to high level.
- Multilingual Neural Machine Translation with Language Clustering126
This work develops a framework that clusters languages into different groups and trains one multilingual model for each cluster and obtains the embedding vectors of all the languages by training a universal neural machine translation model.
- MultiSpeech: Multi-Speaker Text to Speech with Transformer123
A robust and high-quality Transformer TTS system called MultiSpeech, with several specially designed components/techniques to improve text-to-speech alignment, and synthesizes more robust and better quality multi-speaker voice than naive Transformer based TTS.
- Towards a trust aware cognitive radio architecture117
A novel trustaware hybrid spectrum sensing scheme which can detect misbehaving secondary users and filter out their reported spectrum sensing results from the decision making process is proposed.
- SimulSpeech: End-to-End Simultaneous Speech to Text Translation115
Experiments on MuST-C English-Spanish and English-German spoken language translation datasets show that SimulSpeech achieves reasonable BLEU scores and lower delay compared to full-sentence end-to-end speech to text translation (without simultaneous translation), and better performance than the two-stage cascaded simultaneous translation model in terms of BLEu scores and translation delay.
- LRSpeech109
LRSpeech is developed, a TTS and ASR system under the extremely low-resource setting, which can support rare languages with low data cost and achieves promising recognition accuracy for ASR, and last but not least, uses extremelyLow resource training data.
- 108
- Ranking with multiple hyperplanes107
This paper looks at an alternative approach to Ranking SVM, which it is called "Multiple Hyperplane Ranker" (MHR), and makes comparisons between the two approaches, which takes the divide-and-conquer strategy.
- Personalized Federated Learning With Adaptive Batchnorm for Healthcare102
FedAP is proposed to tackle domain shifts and then obtain personalized models for local clients by learning the similarity between clients based on the statistics of the batch normalization layers while preserving the specificity of each client with different local batch normalization.
- Sponsored Search Auctions102
A comprehensive review of sponsored search auctions is provided in hopes of helping both industry practitioners and academic researchers to become familiar with this field, to know the state of the art, and to identify future research topics.
- A study of relevance propagation for web search100
Drawing many conclusions that are useful for selecting a propagation model in real-world search applications are come to, including 1) sitemap-based propagation models outperform hyperlink-based models in sense of both effectiveness and efficiency, and 2) sitEMap- based term propagation is easier to be integrated into real- world search engines because of its parallel offline implementation and acceptable complexity.
- 99
- A New Probabilistic Model for Rank Aggregation99
Experiments on public datasets show that the derived algorithms based on the CPS model can achieve state-of-the-art ranking accuracy, and are much more efficient than previous algorithms.
- TamGen: drug design with target-aware molecule generation through a chemical language model98
TamGen is introduced, a GPT-like model designed to generate molecules tailored to specific target proteins that enables target-aware molecule generation and compound refinement and identified 14 potent compounds against the Tuberculosis ClpP protease, showing its potential for drug discovery.
- NAS-BERT: Task-Agnostic and Adaptive-Size BERT Compression with Neural Architecture Search98
This work uses techniques in neural architecture search (NAS) and proposes NAS-BERT, an efficient method for BERT compression that can find lightweight models with better accuracy than previous approaches, and can be directly applied to different downstream tasks with adaptive model sizes for different requirements of memory or latency.
- 98
- Unified 2D and 3D Pre-Training of Molecular Representations96
This work proposes a new representation learning method based on a unified 2D and 3D pre-training that achieves state-of-the-art results on 10 tasks, and the average improvement on 2D-only tasks is 8.3%.
- MixSpeech: Data Augmentation for Low-Resource Automatic Speech Recognition93
Experimental results show that MixSpeech achieves better accuracy than the baseline models without data augmentation, and outperforms a strongData augmentation method SpecAugment on these recognition tasks.
- Fine-Tuning by Curriculum Learning for Non-Autoregressive Neural Machine Translation93
This work designs a curriculum in the fine-tuning process to progressively switch the training from autoregressive generation to non-autoregressive Generation for NAT, and achieves good improvement over previous NAT baselines in terms of translation accuracy.
- Masked Contrastive Representation Learning for Reinforcement Learning89
This work proposes a new algorithm, i.e., masked contrastive representation learning for RL (M-CURL), which takes the correlation among consecutive inputs into consideration and achieves consistent improvements over CURL on 14 out of 16 environments from DMControl suite and 23 out of 26 environments from Atari 2600 Games.
- DeepSinger: Singing Voice Synthesis with Data Mined From the Web89
DeepSinger, a multi-lingual multi-singer singing voice synthesis (SVS) system built from scratch using singing training data mined from music websites, is developed, which is the first SVS system that directly mines training data fromMusic websites.
- SongMASS: Automatic Song Writing with Pre-training and Alignment Constraint87
SongMASS is proposed, which leverages masked sequence to sequence (MASS) pre-training and attention based alignment modeling for lyric-to-melody and melody- to-lyric generation and uses a dynamic programming strategy to obtain the alignment between each word/syllable in lyric and note in melody.
- A Survey on Low-Resource Neural Machine Translation86
A survey for low-resource NMT is provided and related works are classified into three categories according to the auxiliary data they used: (1) exploiting monolingual data of source and/or target languages, (2) exploiting data from auxiliary languages, and (3) exploiting multi-modal data.
- Multi-Task Learning for Conversational Question Answering over a Large-Scale Knowledge Base85
This work proposes an innovative multi-task learning framework where a pointer-equipped semantic parsing model is designed to resolve coreference in conversations, and naturally empower joint learning with a novel type-aware entity detection model.
- Lightspeech: Lightweight and Fast Text to Speech with Neural Architecture Search84
LightSpeech is proposed, which leverages neural architecture search (NAS) to automatically design more lightweight and efficient models based on FastSpeech, and achieves 15x model compression ratio and 6.5x inference speedup on CPU with on par voice quality.
- Exploiting Adapters for Cross-Lingual Low-Resource Speech Recognition83
This paper proposes a novel algorithm called SimAdapter for explicitly learning knowledge from adapters for parameter-efficient cross-lingual speech adaptation and shows that these two novel algorithms can be integrated for better performance with up to 3.55% relative WER reduction.
- Revisiting Over-Smoothness in Text to Speech81
This work conducts a comprehensive study on NAR-TTS models that use some advanced modeling methods and finds that Laplacian mixture loss performs well at modeling multimodal distributions and enjoys its simplicity, while GAN and Glow achieve the best voice quality while suffering from increased training or model complexity.
- 81
- Learning to rank relational objects and its application to web search79
Experimental results show that the proposed method outperforms the baseline methods for two ranking tasks (Pseudo Relevance Feedback and Topic Distillation) in web search, indicating that the suggested method can indeed make effective use of relation information and content information in ranking.
- Hint-Based Training for Non-Autoregressive Machine Translation77
A novel approach to leveraging the hints from hidden states and word alignments to help the training of N ART models achieves significant improvement over previous NART models for the WMT14 En-De and De-En datasets and are even comparable to a strong LSTM-based ART baseline but one order of magnitude faster in inference.
- 76
- Exploiting Monolingual Data at Scale for Neural Machine Translation72
This work studies how to use both the source-side and target-side monolingual data for NMT, and proposes an effective strategy leveraging both of them.
- Tied Transformers: Neural Machine Translation with Shared Encoder and Decoder72
This work considers model-level sharing and ties the whole parts of the encoder and decoder of an NMT model, and obtains a compact model named Tied Transformer, which demonstrates that such a simple method works well for both similar and dissimilar language pairs.
- UWSpeech: Speech to Speech Translation for Unwritten Languages71
A translation system for unwritten languages, named as UWSpeech, is developed, which converts target unwritten speech into discrete tokens with a converter, and then translates source-language speech into target discrete token with a translator, and finally synthesizes target speech from target discrete Tokens with an inverter.
- 71
- 71
- Individualized Indicator for All69
A Technical Trading Indicator Optimization framework that manages to optimize the original technical indicator by leveraging stock-wise properties and a Skip-gram architecture to learn stock embedding inspired by a valuable knowledge repository formed by fund manager's collective investment behaviors is proposed.
- 69
- 69
- Token-Level Ensemble Distillation for Grapheme-to-Phoneme Conversion68
This work proposes token-level ensemble distillation for G2P conversion, which can boost the accuracy by distilling the knowledge from additional unlabeled data, and reduce the model size but maintain the high accuracy, both of which are very practical and helpful in the online production system.
- Dual Transfer Learning for Neural Machine Translation with Marginal Distribution Regularization68
This work leverages the dual translation model (target-to-source translation) to sample several mostly likely source-side sentences and avoid enumerating all possible candidate source sentences, and transfers the knowledge contained in the dual model to boost the training of the primal model (source- to-target translation), and calls this approach dual transfer learning.
- Dual-view Molecular Pre-training67
This work proposes to leverage both representations and design a new pre-training algorithm, dual-view molecule pre- training (briefly, DVMP), that can effectively combine the strengths of both types of molecule representations.
- 62
- Towards Generating Real-World Time Series Data59
A novel generative framework for RTS data – RTSGAN is proposed to tackle the aforementioned challenges and outperforms the previous generation methods in terms of synthetic data utility for downstream classification and prediction tasks.
- ERNIE-RNA: an RNA language model with structure-enhanced representations58
It is found that the attention maps from ERNIE-RNA with no fine-tuning are able to capture RNA structure in the zero-shot experiment more precisely than conventional methods such as fine-tuned RNAfold and RNAstructure, suggesting that the ERNIE-RNA can provide comprehensive RNA structural representations.
- 58
- Beyond Error Propagation in Neural Machine Translation: Characteristics of Language Also Matter57
The role of error propagation on accuracy drop is overstated in the literature, although it indeed contributes to the accuracy drop problem.
- Exploring Explicit Domain Supervision for Latent Space Disentanglement in Unpaired Image-to-Image Translation55
An unpaired image-to-image translation framework, Domain-supervised GAN (DosGAN), which takes the first step towards the exploration of explicit domain supervision and pre-train a classification network to explicitly classify the domain of an image.
- Decoding with Value Networks for Neural Machine Translation54
Experiments show that a prediction network using a recurrent structure for the value network can significantly improve the translation accuracy on several translation tasks.
- 50
- Adaspeech 2: Adaptive Text to Speech with Untranscribed Data50
This paper develops AdaSpeech 2, an adaptive TTS system that only leverages untranscribed speech data for adaptation, and introduces a mel-spectrogram encoder to a well-trained TTS model to conduct speech reconstruction, and constrain the output sequence of the mel-Spectrogram encoding to be close to that of the original phoneme encoder.
- Denoispeech: Denoising Text to Speech with Frame-Level Noise Modeling50
DenoiSpeech is developed, a TTS system that can synthesize clean speech for a speaker with noisy speech data by modeling the fine-grained frame-level noise with a noise condition module, which is jointly trained with the TTS model.
- FastCorrect 2: Fast Error Correction on Multiple Candidates for Automatic Speech Recognition47
This work proposes FastCorrect 2, an error correction model that takes multiple ASR candidates as input for better correction accuracy, and develops a candidate predictor to detect the most suitable candidate for the decoder.
- 47
- Normalization Helps Training of Quantized LSTM47
It is shown theoretically that training a quantized LSTM is difficult because quantization makes the exploding gradient problem more severe, and that the popularly used weight/layer/batch normalization schemes can help stabilize the gradient magnitude in training quantization LSTMs.
- Efficient Sequence Learning with Group Recurrent Networks47
An efficient architecture is proposed to improve the efficiency of such RNN model training, which adopts the group strategy for recurrent layers, while exploiting the representation rearrangement strategy between layers as well as time steps.
- 47
- A study of BERT for context-aware neural machine translation46
This work conducts experiments on five translation tasks and finds that concatenating all contextual sequences as a longer one and then encoding it by BERT obtains the best translation results.
- Depth Growing for Neural Machine Translation46
This work proposes an effective two-stage approach with three specially designed components to construct deeper NMT models, which result in significant improvements over the strong Transformer baselines on WMT14 English\toGerman and English-toFrench translation tasks.
- 45
- 42
- 42
- 42
- 42
- 41
- Breaking the barriers of data scarcity in drug–target affinity prediction40
The proposed Semi-Supervised Multi-task training (SSM)-DTA framework addresses the data limitation challenge in DTA prediction and yields promising results, paving the way for more efficient and accurate drug discovery processes.
- 40
- 40
- Robust sparse rank learning for non-smooth ranking measures40
This paper considers the sparse algorithms to directly optimize the Normalized Discounted Cumulative Gain (NDCG) which is a widely-used ranking measure, and establishes a reduction framework under which to reduce ranking, as measured by NDCG, to the importance weighted pairwise classification.
- Predicting Groundwater Level Based on Machine Learning: A Case Study of the Hebei Plain38
The GRU model performed best and could support dynamic predictions of GWL in the Hebei Plain and was quantitatively evaluated using the root mean square error (RMSE), coefficient of determination (R2), and Nash–Sutcliffe efficiency (NSE).
- End-to-end entity-aware neural machine translation38
This work proposes an end-to-end algorithm that carefully handles the translation of entities and improves 1.7 BLEU scores on Japanese to English translation and 4.6 entity F1 scores on English to Chinese translation, without additional inference cost.
- Query-level stability and generalization in learning to rank38
The proposed theory of generalization ability of learning to rank algorithms for information retrieval (IR) is applied to the existing algorithms of Ranking SVM and IRSVM, and a number of new concepts are defined, including query-level loss, query- level risk, andquery-level stability are defined.
- Task-Level Curriculum Learning for Non-Autoregressive Neural Machine Translation37
An error correction mechanism is introduced into NMT, which corrects the error information in the previous generated tokens to better predict the next token, and introduces two-stream self-attention from XLNet into NMT decoder.
- 37
- 37
- Dual Inference for Machine Learning36
This paper proposes a general framework of dual inference which can take advantage of both existing models from two dual tasks, without re-training, to conduct inference for one individual task.
- 34
- 34
- R2-DDI: relation-aware feature refinement for drug–drug interaction prediction33
This work proposes a novel R$^2-DDI framework, which introduces a relation-aware feature refinement module for drug representation learning and incorporates the consistency training method to regularize the multi-branch predictions for better generalization.
- 32
- 30
- 27
- 27
- 27
- 27
- 26
- 26
- 25
- 25
- 24
- 23
- 23
- 22
- 22
- 22
- 22
- 22
- 22
- 22
- 21
- 21
- 21
- 21
- 20
- Image-to-Image Translation with Multi-Path Consistency Regularization19
A new kind of loss, multi-path consistency loss, is introduced, which evaluates the differences between direct translation $\mathcal{D}_s\to\mathcal(D)_t$ and indirect translation, to regularize training.
- 19
- 19
- 19
- 18
- 18
- 17
- 17
- 16
- 16
- 16
- 15
- 15
- 15
- 15
- 14
- 14
- 14
- 13
- 13
- 13
- 12
- 12
- 12
- 12
- 12
- 12
- 12
- 12
- 12
- 12
- 12
- 11
- 11
- 11
- 11
- 11
- 11
- 11
- 11
- 11
- 11
- 11
- 10
- 10
- 10
- 10
- 9
- 9
- 9
- 9
- 9
- 9
- 9
- 9
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- 8
- Deliberation Learning for Image-to-Image Translation8
Inspired by the success of deliberation network in natural language processing, deliberation process is extended to the field of image translation by considering an additional polishing step on the output image.
- 8
- 8
- 8
- 8
- 7
- 7
- 7
- 7
- 7
- 7
- 7
- 7
- 7
- 7
- 7
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 6
- 5
- 5
- 5
- 5
- 5
- Towards Interpretable Reasoning over Paragraph Effects in Situation5
A sequential approach for the task of reasoning over paragraph effects in situation which explicitly models each step of the reasoning process with neural network modules, which leads to a more interpretable model.
- 5
- 5
- 5
- 5
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 4
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- Independence-aware Advantage Estimation3
The recognized independence property can be naturally utilized to construct a novel importance sampling advantage estimator with close-to-zero variance even when the Monte-Carlo return signal yields a large variance.
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- Unpaired Image-to-Image Translation with Domain Supervision.2
This work designs an unpaired image-to-image translation framework, Domain-supervised GAN (briefly, DosGAN), that takes the first step towards exploration of domain supervision, and transfers the domain feature extractor obtained on the Facescrub dataset to the CelebA dataset without domain supervision information.
- 2
- Efficient Algorithms with performance guarantees for the stochastic multiple-choice knapsack problem2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-13. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.