| Challenge: | Evidence has shown that multi-head attentive neural architectures are overparameterized. |
| Approach: | They propose a multi-head attentive neural architecture that “reallocates” attention heads to different inputs. |
| Outcome: | The proposed model outperforms baselines on machine translation and language modeling tasks. |
Similar Papers
Mixed Multi-Head Self-Attention for Neural Machine Translation (D19-56)
Copied to clipboard
| Challenge: | Recent advances in neural machine translation have been made in the field of multi-head self-attention and there is no explicit mechanism to ensure that different attention heads capture different features. |
| Approach: | They propose a novel multi-head self-attention model which models not only global and local attention but also forward and backward attention in different attention heads. |
| Outcome: | The proposed model improves on WAT17 English-Japanese and IWSLT14 German-English translation tasks without increasing the number of parameters. |
Mixture of Attention Heads: Selecting Attention Heads Per Token (2022.emnlp-main)
Copied to clipboard
| Challenge: | Mixture-of-Experts (MoE) networks have been proposed as an efficient way to scale up model capacity and implement conditional computing. |
| Approach: | They propose a new architecture that combines multi-head attention with the MoE mechanism and a sparsely gated architecture that allows for faster computations. |
| Outcome: | The proposed architecture can scale up the number of attention heads and the number parameters while preserving computational efficiency. |
A Closer Look into Mixture-of-Experts in Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Mixture-of-experts (MoE) architectures are gaining increasing attention for their unique properties and remarkable performance. |
| Approach: | They propose a mixture-of-experts architecture that allows for model scaling without sacrificing computational efficiency. |
| Outcome: | The proposed model increases model size without sacrificing computational efficiency . the proposed model is modular and can be used by a broad spectrum of practitioners . |
Mixture-of-Linguistic-Experts Adapters for Improving and Interpreting Pre-trained Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | In recent years, pre-trained language models have become the de facto instrument for the field of natural language processing (NLP). |
| Approach: | They propose a method that injects linguistic structures into pre-trained language models in the parameter-efficient fine-tuning setting. |
| Outcome: | The proposed approach outperforms state-of-the-art methods with a comparable number of parameters. |
MoCE: Adaptive Mixture of Contextualization Experts for Byte-based Neural Machine Translation (2025.naacl-long)
Copied to clipboard
| Challenge: | Byte-based machine translation systems can be used in multilingual settings. |
| Approach: | They propose a method that maps each character to specific byte(s) they propose byte-level tokenization that eliminates unknown words . |
| Outcome: | The proposed method outperforms existing methods without manual adjustment of hyper-parameters and surpasses subword-based models with fewer parameters in Ted-59 dataset. |
Multi-Source Domain Adaptation with Mixture of Experts (D18-1)
Copied to clipboard
| Challenge: | Existing methods for domain adaptation from multiple sources are designed to transfer supervision from a single source domain. |
| Approach: | They propose to capture the relationship between a target example and different source domains by a point-to-set metric. |
| Outcome: | The proposed method outperforms baselines and can handle negative transfer. |
HMoE: Heterogeneous Mixture of Experts for Language Modeling (2025.emnlp-main)
Copied to clipboard
An Wang, Xingwu Sun, Ruobing Xie, Shuaipeng Li, Jiaqi Zhu, Zhen Yang, Pinxue Zhao, Weidong Han, Zhanhui Kang, Di Wang, Naoaki Okazaki, Cheng-zhong Xu
| Challenge: | Mixture of Experts (MoE) models use homogeneous experts with diverse capacities, resulting in a lack of expert specialization and parameter utilization. |
| Approach: | They propose a framework where experts differ in size and possess diverse capacities . they propose HMoE to encourage frequent activation of smaller experts . |
| Outcome: | The proposed framework outperforms homogeneous homogenous MoE models on evaluation benchmarks and achieves lower loss rate with fewer activated parameters. |
Alleviating the Inequality of Attention Heads for Neural Machine Translation (2022.coling-1)
Copied to clipboard
| Challenge: | Recent studies show that the attention heads in Transformer are not equal. |
| Approach: | They propose a masking method to mask attention heads in Transformer . they empirically validate the inequality and propose 'head mask' method to avoid bottleneck . |
| Outcome: | The proposed masking method improves translation performance on multiple languages . it can be used to remove a small subset of heads without affecting performance . |
Fixed Encoder Self-Attention Patterns in Transformer-Based Machine Translation (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies have shown that attention heads learn simple positional patterns . |
| Approach: | They propose to replace all but one attention head of each encoder layer with simple fixed – non-learnable – attentive patterns that are solely based on position and do not require external knowledge. |
| Outcome: | The proposed model improves translation quality and improves BLEU scores by up to 3 points in low-resource scenarios. |
Learning Slice-Aware Representations with Mixture of Attentions (2021.findings-acl)
Copied to clipboard
| Challenge: | Real-world machine learning systems are achieving excellent performance in terms of coarse-grained metrics like overall accuracy and F-1 score. |
| Approach: | They extend slice-based learning (SBL) with a mixture of attentions to learn slice-aware dual attentive representations. |
| Outcome: | The proposed approach outperforms the baseline method and the original SBL approach on monitored slices with two natural language understanding tasks. |