Recurrent Neural Networks Learn to Store and Generate Sequences using Non-Linear Representations
Open Access
- 1 January 2024
- conference paper
- Published by Association for Computational Linguistics (ACL)
- p. 248-262
Abstract
Róbert Csordás, Christopher Potts, Christopher D Manning, Atticus Geiger. Proceedings of the 7th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP. 2024.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- Emergent Linear Representations in World Models of Self-Supervised Sequence Models2023
- Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models2021
- How do Decisions Emerge across Layers in Neural Models? Interpretation with Differentiable Masking2020
- Neural Natural Language Inference Models Partially Embed Theories of Lexical Entailment and Negation2020
- Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection2020
- Under the Hood: Using Diagnostic Classifiers to Investigate and Improve how Language Models Track Agreement Information2018
- Deep Contextualized Word Representations2018
- A Latent Variable Model Approach to PMI-based Word EmbeddingsTransactions of the Association for Computational Linguistics, 2016
- Distributed representations, simple recurrent networks, and grammatical structureMachine Learning, 1991
- Finding Structure in TimeCognitive Science, 1990