Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Jiawei HanSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works99 from public data
- 2,450
- 1,717
- RESA: Recurrent Feature-Shift Aggregator for Lane Detection380
A novel module named REcurrent Feature-Shift Aggregator (RESA) to enrich lane feature after preliminary feature extraction with an ordinary CNN and achieves state-of-the-art results on two popular lane detection benchmarks (CULane and Tusimple).
- 378
- 347
- Sparse Fuse Dense: Towards High Quality 3D Detection with Depth Completion284
A novel multi-modal framework SFD (Sparse Fuse Dense), which utilizes pseudo point clouds generated from depth completion to tackle the issues mentioned above and a new RoI fusion strategy 3D-GAF (3D Grid-wise Attentive Fusion) to make fuller use of information from different types of point clouds.
- Graph Transformer for Graph-to-Sequence Learning257
A new model, known as Graph Transformer, is proposed that uses explicit relation encoding and allows direct communication between two distant nodes and provides a more efficient way for global graph structure modeling.
- Multi-Task Pre-Training for Plug-and-Play Task-Oriented Dialogue System223
This study presents PPTOD, a unified plug-and-play model for task-oriented dialogue, and introduces a new dialogue multi-task pre-training strategy that allows the model to learn the primary TOD task completion skills from heterogeneous dialog corpora.
- 219
- 212
- Community Mining from Multi-relational Networks178
This paper systematically analyzes the problem of mining hidden communities on heterogeneous social networks and proposes a new method for learning an optimal linear combination of these relations which can best meet the user's expectation.
- Neural Word Segmentation Learning for Chinese162
A novel neural framework is proposed which thoroughly eliminates context windows and can utilize complete segmentation history and employs a gated combination neural network over characters to produce distributed representations of word candidates, which are then given to a long short-term memory (LSTM) language scoring model.
- AMR Parsing via Graph-Sequence Iterative Inference124
A new end-to-end model that treats AMR parsing as a series of dual decisions on the input sequence and the incrementally constructed graph based on iterative inference that helps achieve better answers in both perspectives, leading to greatly improved parsing accuracy.
- Neural Machine Translation with Monolingual Translation Memory121
A new framework that uses monolingual memory and performs learnable memory retrieval in a cross-lingual manner and outperforms strong TM-augmented NMT baselines using bilingual TM.
- Recent Advances in Retrieval-Augmented Text Generation116
The generic paradigm of retrieval-augmented text generation is highlighted, notable works for different text generation tasks including dialogue generation, machine translation, and other generation tasks are reviewed, and some limitations and shortcomings are pointed out to facilitate future research.
- Skeleton-to-Response: Dialogue Generation Guided by Retrieval Memory99
This paper proposes a new framework which exploits retrieval results via a skeleton-to-response paradigm and shows that this approach significantly improves the informativeness of the generated responses.
- Whom to mention99
An in-depth study of mention mechanism and a recommendation scheme to solve the essential question of whom to mention in a tweet are presented and the advantage of the proposed algorithm compared with the traditional recommendation methods is demonstrated.
- Fast and Accurate Neural Word Segmentation for Chinese96
This paper proposes a greedy neural word segmenter with balanced word and character embedding inputs to alleviate the existing drawbacks of current neural models, capable of performing segmentation much faster and even more accurate than state-of-the-art neural models on Chinese benchmark datasets.
- Charge-Based Prison Term Prediction with Deep Gating Network93
This paper argues that charge-based prison term prediction (CPTP) not only better fits realistic needs, but also makes the total jail term prediction more accurate and interpretable.
- Retrieval-guided Dialogue Response Generation via a Matching-to-Generation Framework91
A novel framework in which the skeleton extraction is made by an interpretable matching model and the following skeleton-guided response generation is accomplished by a separately trained generator is presented.
- COP: Customized Deep Model Compression via Regularized Correlation-Based Filter-Level Pruning84
A novel algorithm named COP (correlation-based pruning) is developed, which can detect the redundant filters efficiently and enable the cross-layer filter comparison through global normalization, and adds parameter-quantity and computational-cost regularization terms to the importance.
- 81
- Tracking people in RGBD videos using deep learning and motion clues72
This work proposes to train a deep convolutional neural network, which improves tracking performance, to classify people and a motion model based on spatial and kinetic clues is combined with the network to track people in the scene.
- 70
- Describe What to Change62
This work proposes a novel unsupervised approach, based on image-to-image translation, that alters the attributes of a given image through a command-like sentence such as "change the hair color to black".
- Deep Rotation Equivariant Network62
This work proposes Deep Rotation Equivariant Network consisting of cycle layers, isotonic layers and decycle layers, and evaluates DRENs on Rotated MNIST and CIFAR-10 datasets and demonstrates that it can improve the performance of state-of-the-art architectures.
- 59
- 59
- Core Semantic First: A Top-down Approach for AMR Parsing56
A novel scheme for parsing a piece of text into its Abstract Meaning Representation (AMR): Graph Spanning based Parsing (GSP), which achieves the state-of-the-art performance in the sense that no heuristic graph re-categorization is adopted.
- Dialogue Response Selection with Hierarchical Curriculum Learning51
A hierarchical curriculum learning framework that trains the matching model in an “easy-to-difficult” scheme for dialogue response selection significantly improves the model performance across various evaluation metrics.
- Frame-wise Action Representations for Long Videos via Sequence Contrastive Learning48
This paper introduces a novel contrastive action representation learning (CARL) framework to learn frame-wise action representations, especially for long videos, in a self-supervised manner and shows outstanding performance on video alignment and fine-grained frame retrieval tasks.
- Reinforced Dynamic Reasoning for Conversational Question Generation48
A new approach named Reinforced Dynamic Reasoning network is proposed, which is based on the general encoder-decoder framework but incorporates a reasoning procedure in a dynamic manner to better understand what has been asked and what to ask next about the passage into the general Encoder-Decoder framework.
- 45
- Dynamic Semantic Graph Construction and Reasoning for Explainable Multi-hop Science Question Answering41
Results on two scientific multi-hop QA datasets show that the proposed framework can surpass recent approaches including those using additional knowledge graphs while maintaining high explainability on OpenBookQA and achieve a new state-of-the-art result on ARC-Challenge in a computationally practicable setting.
- The World is Not Binary: Learning to Rank with Grayscale Data for Dialogue Response Selection41
This work shows that grayscale data can be automatically constructed without human effort, and proposes multi-level ranking objectives for training, which can teach a matching model to capture more fine-grained context-response relevance difference and reduce the train-test discrepancy in terms of distractor strength.
- Copy Is All You Need37
This paper formulate text generation as progressively copying text segments from an existing text collection, and compute the contextualized representations of meaningful text segments and index them using efficient vector search toolkits.
- 35
- A Hybrid Model for Chinese Spelling Check33
The proposed model is evaluated on the standard datasets released by the SIGHAN Bake-off shared tasks, giving state-of-the-art results.
- Exploiting Reasoning Chains for Multi-hop Science Question Answering32
The proposed Chain Guided Retriever-reader framework allows the retriever to capture step-by-step clues of the entire reasoning process, which is not only shown to be effective on two challenging multi-hop Science QA tasks, namely OpenBookQA and ARC-Challenge, but also favors explainability.
- 32
- 28
- 28
- 28
- 27
- 25
- 24
- Exploring Dense Retrieval for Dialogue Response Selection21
This work presents a novel and efficient response selection model and a set of tailor-designed learning strategies to train it effectively, which notably outperforms the state-of-the-art baselines on both re-rank and full-rank evaluations.
- Multilingual AMR Parsing with Noisy Knowledge Distillation21
This work obtains an AMR parser whose performances surpass all previously published results on four different foreign languages, including German, Spanish, Italian, and Chinese, by large margins.
- 20
- 18
- 18
- 17
- 16
- 16
- Unsupervised Learning Helps Supervised Neural Word Segmentation16
This work makes the first attempt at exploring adding unsupervised segmentation information into neural supervised segmenter, and verifies the proposed approach outperforms the widelyused pre-training approach in terms of effectively making use of freely abundant unlabeled data.
- SCALoss: Side and Corner Aligned Loss for Bounding Box Regression15
Side Overlap (SO) loss is proposed by maximizing the side overlap of two bounding boxes, which puts more penalty for low overlapping bounding box cases, and the Corner Distance is added into the objective function to speed up the convergence.
- 15
- 14
- 14
- 14
- 14
- 12
- 11
- 11
- 9
- 8
- 8
- 8
- 7
- 6
- 6
- Bi-Decoder Augmented Network for Neural Machine Translation6
This paper proposes a novel Bi-Decoder Augmented Network (BiDAN) for the neural machine translation task, and adds an auxiliary decoder to generate back the source language sequence at the training time.
- 6
- 5
- 5
- 5
- 5
- 5
- 5
- SeqPE: Transformer With Sequential Position Encoding4
Experiments across language modeling, longcontext question answering, and 2D image classification demonstrate that SeqPE not only surpasses strong baselines in perplexity, exact match (EM), and accuracy—particularly under context length extrapolation—but also enables seamless generalization to multi-dimensional inputs without requiring manual architectural redesign.
- 4
- 4
- 4
- 3
- 3
- 3
- 2
- 2
- 2
- 2
- Spatiotemporal Multi-Task Network for Human Activity Understanding2
A spatiotemporal, multi-task, 3D deep convolutional neural network to detect actions in untrimmed videos and introduces a novel video representation, interlaced images, as an additional network input stream to better utilize the rich motion information in videos.
- 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.