# Tying Word Vectors and Word Classifiers: A Loss Framework for Language Modeling

arXiv: https://arxiv.org/abs/1611.01462 · HTML: https://mlsanity.com/paper/1611.01462/

## Key facts

- Cited by 92 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Nov 2016.

## Built on by (most cited first, 30 of 92)

- [Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation](https://mlsanity.com/paper/2108.12409/) (2021)
- [The Falcon Series of Open Language Models](https://mlsanity.com/paper/2311.16867/) (2023)
- [Training Graph Neural Networks with 1000 Layers](https://mlsanity.com/paper/2106.07476/) (2021)
- [The mechanistic basis of data dependence and abrupt learning in an in-context classification task](https://mlsanity.com/paper/2312.03002/) (2023)
- [Language models scale reliably with over-training and on downstream tasks](https://mlsanity.com/paper/2403.08540/) (2024)
- [Finetuning Pretrained Transformers into RNNs](https://mlsanity.com/paper/2103.13076/) (2021)
- [Adaptive Semiparametric Language Models](https://mlsanity.com/paper/2102.02557/) (2021)
- [Tied-Lora: Enhancing parameter efficiency of LoRA with weight tying](https://mlsanity.com/paper/2311.09578/) (2023)
- [Spike No More: Stabilizing the Pre-training of Large Language Models](https://mlsanity.com/paper/2312.16903/) (2023)
- [Fishing for Magikarp: Automatically Detecting Under-trained Tokens in Large Language Models](https://mlsanity.com/paper/2405.05417/) (2024)
- [Tied & Reduced RNN-T Decoder](https://mlsanity.com/paper/2109.07513/) (2021)
- [FLCert: Provably Secure Federated Learning against Poisoning Attacks](https://mlsanity.com/paper/2210.00584/) (2022)
- [Selfish Sparse RNN Training](https://mlsanity.com/paper/2101.09048/) (2021)
- [Wav2Seq: Pre-training Speech-to-Text Encoder-Decoder Models Using Pseudo Languages](https://mlsanity.com/paper/2205.01086/) (2022)
- [Efficient recurrent architectures through activity sparsity and sparse back-propagation through time](https://mlsanity.com/paper/2206.06178/) (2022)
- [Hybrid Random Features](https://mlsanity.com/paper/2110.04367/) (2021)
- [Relational Memory Augmented Language Models](https://mlsanity.com/paper/2201.09680/) (2022)
- [How Does Fine-tuning Affect the Geometry of Embedding Space: A Case Study on Isotropy](https://mlsanity.com/paper/2109.04740/) (2021)
- [Convex Aggregation for Opinion Summarization](https://mlsanity.com/paper/2104.01371/) (2021)
- [Multilingual Machine Translation with Hyper-Adapters](https://mlsanity.com/paper/2205.10835/) (2022)
- [Tensor Decomposition for Model Reduction in Neural Networks: A Review](https://mlsanity.com/paper/2304.13539/) (2023)
- [Solve the Loop: Attractor Models for Language and Reasoning](https://mlsanity.com/paper/2605.12466/) (2026)
- [InvBERT: Reconstructing Text from Contextualized Word Embeddings by inverting the BERT pipeline](https://mlsanity.com/paper/2109.10104/) (2021)
- [Frequency-Aware Contrastive Learning for Neural Machine Translation](https://mlsanity.com/paper/2112.14484/) (2021)
- [Generalization in Neural Networks: A Broad Survey](https://mlsanity.com/paper/2209.01610/) (2022)
- [What makes math problems hard for reinforcement learning: a case study](https://mlsanity.com/paper/2408.15332/) (2024)
- [The Rediscovery Hypothesis: Language Models Need to Meet Linguistics](https://mlsanity.com/paper/2103.01819/) (2021)
- [Transformers with multi-modal features and post-fusion context for e-commerce session-based recommendation](https://mlsanity.com/paper/2107.05124/) (2021)
- [Coordinating Distributed Example Orders for Provably Accelerated Training](https://mlsanity.com/paper/2302.00845/) (2023)
- [Masked Generative Priors Improve World Models Sequence Modelling Capabilities](https://mlsanity.com/paper/2410.07836/) (2024)

## Abstract (arXiv)

Recurrent neural networks have been very successful at predicting sequences of words in tasks such as language modeling. However, all such models are based on the conventional classification framework, where the model is trained against one-hot targets, and each word is represented both as an input and as an output in isolation. This causes inefficiencies in learning both in terms of utilizing all of the information and in terms of the number of parameters needed to train. We introduce a novel theoretical framework that facilitates better learning in language modeling, and show that our framework leads to tying together the input embedding and the output projection matrices, greatly reducing the number of trainable variables. Our framework leads to state of the art performance on the Penn Treebank with a variety of network models.