| Challenge: | Anisotropic Embeddings Large Language Models exhibit undesirable yet poorly understood feature of anisotropy. |
| Approach: | They propose an algorithm that uses the second moment in Adam to mitigate anisotropic embeddings . they propose an embeddable matrix and unembedding matrix to map the input and output tokens based on weight tying . |
| Outcome: | The proposed model improves quality and performance on large datasets. |
Similar Papers
More Embeddings, Better Sequence Labelers? (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work suggests contextual embeddings improve sequence labeling accuracy . but, there is no definite conclusion on whether concatenating different kinds of embeddables is effective . |
| Approach: | They propose a family of contextual embeddings that improves sequence labeling accuracy . they conduct extensive experiments on 3 tasks over 18 datasets and 8 languages . |
| Outcome: | The proposed family of contextual embeddings improves the accuracy of sequence labelers over non-contextual embedders. |
How to represent a word and predict it, too: Improving tied architectures for language modelling (D18-1)
Copied to clipboard
| Challenge: | Recent state-of-the-art models use word embeddings as input and output mappings instead of tied models. |
| Approach: | They propose to decouple hidden state from word embedding prediction . they extend their proposed modification to word2vec models . |
| Outcome: | The proposed architectures achieve comparable or better results compared to previous models without tying . the proposed architecture reduces parameters, enabling more compact models and faster learning. |
Dynamic Meta-Embeddings for Improved Sentence Representations (D18-1)
Copied to clipboard
| Challenge: | A sprawling literature has emerged about what word embeddings are most useful for which tasks . word embed-ding is a technique that can be used to learn word-level meaning representations for a variety of tasks. |
| Approach: | They propose a method for supervised learning of embedding ensembles that leads to state-of-the-art performance on a variety of tasks. |
| Outcome: | The proposed method leads to state-of-the-art performance on a variety of tasks. |
Unlike “Likely”, “Unlike” is Unlikely: BPE-based Segmentation hurts Morphological Derivations in LLMs (2025.coling-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) use subword vocabularies to process and generate text. |
| Approach: | They find that Large Language Models (LLMs) perform poorly at handling some types of affixations because subwords are marked as initial- or intra-word . |
| Outcome: | The largest models trained on enough data can mitigate this tendency because initial- and intra-word embeddings are aligned; in-context learning also helps when all examples are selected in a consistent way; but only morphological segmentation can achieve a near-perfect accuracy. |
Too Much in Common: Shifting of Embeddings in Transformer Language Models and its Implications (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing studies have shown that word embeddings do not occupy a narrow cone, but rather drift in common directions. |
| Approach: | They show that anisotropy can be restored using a simple transformation of word embeddings. |
| Outcome: | The proposed model can restore anisotropy using a simple transformation. |
Is Anisotropy Truly Harmful? A Case Study on Text Clustering (2023.acl-short)
Copied to clipboard
| Challenge: | Contextualized pre-trained representations are widely used as input to various tasks such as information retrieval, anomaly detection and document clustering. |
| Approach: | They propose to examine the impact of different transformations on isotropy and performance to assess the true impact of anisotropi. |
| Outcome: | The proposed model is based on a clustering task and shows that it has limited impact on expressiveness and closeness. |
Autoencoding Improves Pre-trained Word Embeddings (2020.coling-main)
Copied to clipboard
| Challenge: | Existing work has shown that word embeddings are distributed in a narrow cone and that centering and projection can improve the accuracy of pre-trained word embeds without requiring additional training data. |
| Approach: | They propose to remove the top principal components from pre-trained word embeddings and center and project them onto principal component vectors to reinstate isotropy in the embeddable space. |
| Outcome: | The proposed method is equivalent to applying a linear autoencoder to minimize the squared L2 reconstruction error. |
Exploring Anisotropy and Outliers in Multilingual Language Models for Cross-Lingual Semantic Sentence Similarity (2023.findings-acl)
Copied to clipboard
| Challenge: | Recent studies have shown that contextual language models display outlier dimensions . this is true for monolingual and multilingual models, but little work has been done on multilingual contexts . |
| Approach: | They investigate outlier dimensions and their relationship to anisotropy in multilingual contexts . they focus on cross-lingual semantic similarity tasks . |
| Outcome: | The proposed model improves on cross-lingual semantic similarity tasks. |
Frustratingly Easy Performance Improvements for Low-resource Setups: A Tale on BERT and Segment Embeddings (2022.lrec-1)
Copied to clipboard
| Challenge: | Understanding why contextualized embeddings work is still an active area of research. |
| Approach: | They propose to use a BERT architecture to encode a sub-word, position and a segment embedding as input representations for each sub- word. |
| Outcome: | The proposed model performs well on single-sentence prediction tasks while swapping segment IDs in paired-sentent tasks. |
Contextual Embeddings: When Are They Worth It? (2020.acl-main)
Copied to clipboard
| Challenge: | In recent years, rich contextual embeddings have enabled rapid progress on benchmarks like GLUE, but require significant computational resources during pretraining and during downstream task training and inference. |
| Approach: | They empirically compare contextual embeddings with classic pretrained embedders and a random word embeddable with a simple baseline. |
| Outcome: | The proposed models perform within 5 to 10% accuracy on industry-scale data. |