DiNADO: Norm-Disentangled Neurally-Decomposed Oracles for Controlling Language Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lu, Sidi, Zhao, Wenbo, Tao, Chenyang, Gupta, Arpit, Wu, Shanchan, Chung, Tagyoung, Peng, Nanyun
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916278214590464
author Lu, Sidi
Zhao, Wenbo
Tao, Chenyang
Gupta, Arpit
Wu, Shanchan
Chung, Tagyoung
Peng, Nanyun
author_facet Lu, Sidi
Zhao, Wenbo
Tao, Chenyang
Gupta, Arpit
Wu, Shanchan
Chung, Tagyoung
Peng, Nanyun
contents NeurAlly-Decomposed Oracle (NADO) is a powerful approach for controllable generation with large language models. It is designed to avoid catastrophic forgetting while achieving guaranteed convergence to an entropy-maximized closed-form optimal solution with reasonable modeling capacity. Despite the success, several challenges arise when apply NADO to a wide range of scenarios. Vanilla NADO suffers from gradient vanishing for low-probability control signals and is highly reliant on a regularization to satisfy the stochastic version of Bellman equation. In addition, the vanilla implementation of NADO introduces a few additional transformer layers, suffering from a limited capacity especially compared to other finetune-based model adaptation methods like LoRA. In this paper, we propose a improved version of the NADO algorithm, namely DiNADO (norm-Disentangled NeurAlly-Decomposed Oracles), which improves the performance of the NADO algorithm through disentangling the step-wise global norm over the approximated oracle $R$-value for all potential next-tokens, allowing DiNADO to be combined with finetuning methods like LoRA. We discuss in depth how DiNADO achieves better capacity, stability and flexibility with both empirical and theoretical results. Experiments on formality control in machine translation and the lexically constrained generation task CommonGen demonstrates the significance of the improvements.
format Preprint
id arxiv_https___arxiv_org_abs_2306_11825
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle DiNADO: Norm-Disentangled Neurally-Decomposed Oracles for Controlling Language Models
Lu, Sidi
Zhao, Wenbo
Tao, Chenyang
Gupta, Arpit
Wu, Shanchan
Chung, Tagyoung
Peng, Nanyun
Computation and Language
NeurAlly-Decomposed Oracle (NADO) is a powerful approach for controllable generation with large language models. It is designed to avoid catastrophic forgetting while achieving guaranteed convergence to an entropy-maximized closed-form optimal solution with reasonable modeling capacity. Despite the success, several challenges arise when apply NADO to a wide range of scenarios. Vanilla NADO suffers from gradient vanishing for low-probability control signals and is highly reliant on a regularization to satisfy the stochastic version of Bellman equation. In addition, the vanilla implementation of NADO introduces a few additional transformer layers, suffering from a limited capacity especially compared to other finetune-based model adaptation methods like LoRA. In this paper, we propose a improved version of the NADO algorithm, namely DiNADO (norm-Disentangled NeurAlly-Decomposed Oracles), which improves the performance of the NADO algorithm through disentangling the step-wise global norm over the approximated oracle $R$-value for all potential next-tokens, allowing DiNADO to be combined with finetuning methods like LoRA. We discuss in depth how DiNADO achieves better capacity, stability and flexibility with both empirical and theoretical results. Experiments on formality control in machine translation and the lexically constrained generation task CommonGen demonstrates the significance of the improvements.
title DiNADO: Norm-Disentangled Neurally-Decomposed Oracles for Controlling Language Models
topic Computation and Language
url https://arxiv.org/abs/2306.11825