Verb Conjugation in Transformers Is Determined by Linear Encodings of Subject Number (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work on LMs' internal vector representations has demonstrated the existence of syntactic information in those vectors, but has not described how LM models use this information. |
| Approach: | They propose a causal intervention paradigm to show that some linguistic features are represented in a linear format. |
| Outcome: | The proposed model can generate verb conjugations 91% correct before and 85% incorrect after the intervention. |
Similar Papers
Representation biases in sentence transformers (2023.eacl-main)
Copied to clipboard
| Challenge: | argued that transformer-based models are not well suited for sentence-level downstream tasks. |
| Approach: | They propose to use sentence transformers to produce full-sentence representations . they propose to combine transformers with a training regime that embeds tokens into the model . |
| Outcome: | The proposed model performs better on downstream tasks than the vanilla model and its variants. |
Are Transformers a Modern Version of ELIZA? Observations on French Object Verb Agreement (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies have shown that unsupervised sentence representations of neural networks encode syntactic information by observing that neural language models are able to predict the agreement between a verb and its subject. |
| Approach: | They propose to take an alternative look at these results by studying whether neural networks are able to build an abstract sentence representation rather than capture surface statistical regularities. |
| Outcome: | The proposed model can achieve high accuracy on the long-range French object-verb agreement, indicating a possible flaw in the model's syntactic ability. |
The Functional Relevance of Probed Information: A Case Study (2023.eacl-main)
Copied to clipboard
| Challenge: | Recent studies have shown that transformer models like BERT rely on number information encoded in their representations of sentences’ subjects and head verbs when performing subject-verb agreement. |
| Approach: | They propose to use probing to find out which words contain functionally relevant information encoded in the representations of subject plurality and words that agree with it in number in BERT. |
| Outcome: | The proposed model only uses the subject plurality information encoded in its representations of the subject and words that agree with it in number. |
Learning Syntax Without Planting Trees: Understanding Hierarchical Generalization in Transformers (2025.tacl-1)
Copied to clipboard
Kabir Ahuja, Vidhisha Balachandran, Madhur Panwar, Tianxing He, Noah A. Smith, Navin Goyal, Yulia Tsvetkov
| Challenge: | Inductive biases in transformers can cause hierarchical generalization without explicitly encoding structural bias. |
| Approach: | They investigate sources of inductive bias in transformer models and their training that could cause such preference for hierarchical generalization. |
| Outcome: | The proposed model can generalize to novel syntactic forms without explicit bias . the proposed model is able to generalize on a dataset with a hierarchical grammar . |
Jump to Conclusions: Short-Cutting Transformers with Linear Transformations (2024.lrec-main)
Copied to clipboard
| Challenge: | Transformer-based language models create hidden representations of inputs at every layer, but only use final-layer representations for prediction. |
| Approach: | They propose a method for casting hidden representations as final representations, bypassing transformer computation in-between. |
| Outcome: | The proposed method produces more accurate predictions from hidden layers across various model scales, architectures, and data distributions. |
Theoretical Analysis of Hierarchical Language Recognition and Generation by Transformers without Positional Encoding (2025.acl-long)
Copied to clipboard
| Challenge: | Existing studies show that Transformers can recognize hierarchical structures without a specific positional encoding. |
| Approach: | They show that Transformers can generate hierarchical languages without a positional encoding . they also suggest that explicit positional encoders might have a detrimental effect on generalization . |
| Outcome: | The proposed model can generate hierarchical languages with respect to model size without encoding . |
Does BERT really agree ? Fine-grained Analysis of Lexical Dependence on a Syntactic Task (2022.findings-acl)
Copied to clipboard
| Challenge: | lexically-independent subject-verb number agreement (NA) is performed by transformer-based neural language models (NLMs) . but when as little as one attractor is present, the model fails to perform lexical generalization . |
| Approach: | They propose to disrupt lexical patterns found in naturally occurring stimuli for each targeted structure in a novel fine-grained analysis of BERT's behavior. |
| Outcome: | The proposed model generalizes well for simple templates, but fails to perform lexically-independent generalization when as little as one attractor is present. |
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (N19-1)
Copied to clipboard
| Challenge: | Existing language representation models pre-train deep bidirectional representations from unlabeled text without significant task-specific architecture modifications. |
| Approach: | They propose a language representation model that pre-trains bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers. |
| Outcome: | The proposed model achieves state-of-the-art results on eleven natural language processing tasks, pushing the GLUE score to 80.5 (7.7 point absolute improvement), MultiNLI accuracy to 86.7% (4.6% absolute improvement) |
Probing Image-Language Transformers for Verb Understanding (2021.findings-acl)
Copied to clipboard
| Challenge: | Multimodal image-language transformers have achieved impressive results on a variety of tasks that rely on fine-tuning. |
| Approach: | They collect a dataset of image-sentence pairs consisting of 421 verbs . they evaluate pretrained image-language transformers and find they fail more in situations that require verb understanding compared to other parts of speech. |
| Outcome: | The proposed model trains on a manually-annotated and smaller dataset does better on the task. |
Do Transformers Parse while Predicting the Masked Word? (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies show that pre-trained language models encode linguistic structures like parse trees while being trained unsupervised. |
| Approach: | They propose to train pre-trained language models to encode linguistic structures like parse trees while unsupervised. |
| Outcome: | The proposed model performs optimally for masked language modeling loss on the English PCFG. |