Picture for Yair Lakretz

Yair Lakretz

UNICOG-U992

A polar coordinate system represents syntax in large language models

Add code
Dec 07, 2024
Viaarxiv icon

What Makes Two Language Models Think Alike?

Add code
Jun 24, 2024
Figure 1 for What Makes Two Language Models Think Alike?
Figure 2 for What Makes Two Language Models Think Alike?
Figure 3 for What Makes Two Language Models Think Alike?
Figure 4 for What Makes Two Language Models Think Alike?
Viaarxiv icon

What makes two models think alike?

Add code
Jun 18, 2024
Figure 1 for What makes two models think alike?
Figure 2 for What makes two models think alike?
Figure 3 for What makes two models think alike?
Figure 4 for What makes two models think alike?
Viaarxiv icon

Metric-Learning Encoding Models Identify Processing Profiles of Linguistic Features in BERT's Representations

Add code
Feb 18, 2024
Figure 1 for Metric-Learning Encoding Models Identify Processing Profiles of Linguistic Features in BERT's Representations
Figure 2 for Metric-Learning Encoding Models Identify Processing Profiles of Linguistic Features in BERT's Representations
Figure 3 for Metric-Learning Encoding Models Identify Processing Profiles of Linguistic Features in BERT's Representations
Figure 4 for Metric-Learning Encoding Models Identify Processing Profiles of Linguistic Features in BERT's Representations
Viaarxiv icon

Language acquisition: do children and language models follow similar learning stages?

Add code
Jun 06, 2023
Viaarxiv icon

Probing Brain Context-Sensitivity with Masked-Attention Generation

Add code
May 23, 2023
Viaarxiv icon

Information-Restricted Neural Language Models Reveal Different Brain Regions' Sensitivity to Semantics, Syntax and Context

Add code
Feb 28, 2023
Viaarxiv icon

Neural Language Models are not Born Equal to Fit Brain Data, but Training Helps

Add code
Jul 07, 2022
Figure 1 for Neural Language Models are not Born Equal to Fit Brain Data, but Training Helps
Figure 2 for Neural Language Models are not Born Equal to Fit Brain Data, but Training Helps
Figure 3 for Neural Language Models are not Born Equal to Fit Brain Data, but Training Helps
Figure 4 for Neural Language Models are not Born Equal to Fit Brain Data, but Training Helps
Viaarxiv icon

Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models

Add code
Jun 10, 2022
Viaarxiv icon

Causal Transformers Perform Below Chance on Recursive Nested Constructions, Unlike Humans

Add code
Oct 14, 2021
Figure 1 for Causal Transformers Perform Below Chance on Recursive Nested Constructions, Unlike Humans
Figure 2 for Causal Transformers Perform Below Chance on Recursive Nested Constructions, Unlike Humans
Figure 3 for Causal Transformers Perform Below Chance on Recursive Nested Constructions, Unlike Humans
Viaarxiv icon