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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Large Language Models (LLMs) have demonstrated superior performance inlanguage understanding benchmarks. CALM, a popular approach, leverageslinguistic priors of LLMs -- GPT-2 -- for action candidate recommendations toimprove the performance in text games in Jericho without environment-providedactions. However, CALM adapts GPT-2 with annotated human gameplays and keepsthe LLM fixed during the learning of the text based games. In this work, weexplore and evaluate updating LLM used for candidate recommendation during thelearning of the text based game as well to mitigate the reliance on the humanannotated gameplays, which are costly to acquire. We observe that by updatingthe LLM during learning using carefully selected in-game transitions, we canreduce the dependency on using human annotated game plays for fine-tuning theLLMs. We conducted further analysis to study the transferability of the updatedLLMs and observed that transferring in-game trained models to other games didnot result in a consistent transfer.

Language Model-In-The-Loop: Data Optimal Approach to Learn-To-Recommend Actions in Text Games