Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks277 from public data
- An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-Tuning835
It is observed that general instruction tuning can help alleviate the forgetting phenomenon in LLMs during subsequent fine-tuning and also observes that LLMs can mitigate language biases, such as gender bias, during continual fine-tuning.
- Large Language Models Are Not Robust Multiple Choice Selectors620
This work proposes a label-free, inference-time debiasing method, called PriDe, which separates the model's prior bias for option IDs from the overall prediction distribution, and demonstrates that it achieves interpretable and transferable debiasing with high computational efficiency.
- Bridging the Gap between Training and Inference for Neural Machine Translation280
This paper addresses the issues of overcorrection over different but reasonable translations by sampling context words not only from the ground truth sequence but also from the predicted sequence by the model during training, where the predicted sequences is selected with a sentence-level optimum.
- 231
- Modeling Localness for Self-Attention Networks196
This work cast localness modeling as a learnable Gaussian bias, which indicates the central and scope of the local region to be paid more attention in self-attention networks, to maintain the strength of capturing long distance dependencies while enhance the ability of capturing short-range dependencies.
- A Novel Graph-based Multi-modal Fusion Encoder for Neural Machine Translation177
A novel graph-based multi-modal fusion encoder for NMT that captures various semantic relationships between multi- modal semantic units (words and visual objects) and provides an attention-based context vector for the decoder.
- Towards Robust Neural Machine Translation169
Experimental results on Chinese-English, English-German and English-French translation tasks show that the proposed approaches can not only achieve significant improvements over strong NMT systems but also improve the robustness of NMT models.
- Incremental Transformer with Deliberation Decoder for Document Grounded Conversations124
This paper designs an Incremental Transformer to encode multi-turn utterances along with knowledge in related documents and designs a two-pass decoder (Deliberation Decoder) to improve context coherence and knowledge correctness.
- Encoding Source Language with Convolutional Neural Network for Machine Translation109
A more systematic treatment by summarizing the relevant source information through a convolutional architecture guided by the target information, which can achieve significant improvements over the previous NNJM.
- GCDT: A Global Context Enhanced Deep Transition Architecture for Sequence Labeling99
A Global Context enhanced Deep Transition architecture for sequence labeling named GCDT is proposed, which deepen the state transition path at each position in a sentence, and further assign every token with a global representation learned from the entire sentence.
- Minimizing the Bag-of-Ngrams Difference for Non-Autoregressive Neural Machine Translation95
This paper proposes to train NAT to minimize the Bag-of-Ngrams (BoN) difference between the model output and the reference sentence, and shows that this approach largely outperforms the NAT baseline on three translation tasks.
- A dependency syntactic knowledge augmented interactive architecture for end-to-end aspect-based sentiment analysis91
This paper proposes a novel dependency syntactic knowledge augmented interactive architecture with multi-task learning for end-to-end ABSA that significantly outperforms existing state-of-the-art methods.
- CM-Net: A Novel Collaborative Memory Network for Spoken Language Understanding90
A novel Collaborative Memory Network (CM-Net) based on the well-designed block, named CM-block, which achieves the state-of-the-art results on the ATIS and SNIPS in most of criteria, and significantly outperforms the baseline models on the CAIS.
- A Survey on Cross-Lingual Summarization84
This is the first systematic critical review on the datasets, approaches, and challenges in cross-lingual summarization according to different construction methods and solution paradigms, respectively and discusses promising directions.
- Unsupervised Paraphrasing by Simulated Annealing79
Results show that UPSA achieves the state-of-the-art performance compared with previous unsupervised methods in terms of both automatic and human evaluations, and outperforms most existing domain-adapted supervised models, showing the generalizability of UPSA.
- Selective Knowledge Distillation for Neural Machine Translation77
A novel protocol is designed that can effectively analyze the different impacts of samples by comparing various samples’ partitions and find that the teacher’s knowledge is not the more, the better.
- Bridging the Gap between Prior and Posterior Knowledge Selection for Knowledge-Grounded Dialogue Generation77
A prior selection module is enhanced with the necessary posterior information obtained from the specially designed Posterior Information Prediction Module (PIPM) and a Knowledge Distillation Based Training Strategy (KDBTS) is proposed to train the decoder with the knowledge selected from the prior distribution, removing the exposure bias of knowledge selection.
- On Prompt-Driven Safeguarding for Large Language Models69
It is demonstrated that DRO remarkably improves the safeguarding performance of human-crafted safety prompts and outperforms strong baselines, as evaluated on out-of-domain benchmarks, without compromising the general model capability.
- Infusing Multi-Source Knowledge with Heterogeneous Graph Neural Network for Emotional Conversation Generation62
A heterogeneous graph-based model that can effectively perceive emotions from multi-source knowledge and generate a satisfactory response, which significantly outperforms previous state-of-the-art models.
- A Contextual Hierarchical Attention Network with Adaptive Objective for Dialogue State Tracking60
This paper proposes to enhance the DST through employing a contextual hierarchical attention network to not only discern relevant information at both word level and turn level but also learn contextual representations, and proposes an adaptive objective to alleviate the slot imbalance problem by dynamically adjust weights of different slots during training.
- 57
- A Novel Aspect-Guided Deep Transition Model for Aspect Based Sentiment Analysis49
A novel Aspect-Guided Deep Transition model, named AGDT, is proposed, which utilizes the given aspect to guide the sentence encoding from scratch with the specially-designed deep transition architecture, and an aspect-oriented objective is designed to enforce AGDT to reconstruct thegiven aspect with the generated sentence representation.
- DTMT: A Novel Deep Transition Architecture for Neural Machine Translation49
A novel Deep Transition RNN-based Architecture for Neural Machine Translation, named DTMT, which enhances the hidden-to-hidden transition with multiple non-linear transformations, as well as maintains a linear transformation path throughout this deep transition by the well-designed linear transformation mechanism to alleviate the gradient vanishing problem.
- Retrieving Sequential Information for Non-Autoregressive Neural Machine Translation44
Experimental results on three translation tasks show that the Reinforce-NAT surpasses the baseline NAT system by a significant margin on BLEU without decelerating the decoding speed and the FS-decoder achieves comparable translation performance to the autoregressive Transformer with considerable speedup.
- Zero-Shot Cross-Lingual Summarization via Large Language Models43
It is found that ChatGPT and GPT-4 originally prefer to produce lengthy summaries with detailed information, and these two LLMs can further balance informativeness and conciseness with the help of an interactive prompt, significantly improving their CLS performance.
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.