English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In the quest to advance human-centric natural language generation (NLG)systems, ensuring alignment between NLG models and human preferences iscrucial. For this alignment, current popular methods leverage a reinforcementlearning (RL) approach with a reward model trained on feedback from humans.However, inherent disagreements due to the subjective nature of humanpreferences pose a significant challenge for training the reward model,resulting in a deterioration of the NLG performance. To tackle this issue,previous approaches typically rely on majority voting or averaging toconsolidate multiple inconsistent preferences into a merged one. Althoughstraightforward to understand and execute, such methods suffer from aninability to capture the nuanced degrees of disaggregation among humans and mayonly represent a specialized subset of individuals, thereby lacking the abilityto quantitatively disclose the universality of human preferences. To addressthis challenge, this paper proposes a novel approach, which employs a Bayesianframework to account for the distribution of disagreements among humanpreferences as training a preference model, and names it as d-PM. Besides,considering the RL strategy's inefficient and complex training process over thetraining efficiency, we further propose utilizing the contrastive learningstrategy to train the NLG model with the preference scores derived from thed-PM model. Extensive experiments on two human-centric NLG tasks, i.e.,emotional support conversation and integrity "Rule-of-Thumb" generation, showthat our method consistently exceeds previous SOTA models in both automatic andhuman evaluations.

Aligning Language Models with Human Preferences via a Bayesian Approach