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

One of the essential missions in the AI research community is to build anautonomous embodied agent that can attain high-level performance across a widespectrum of tasks. However, acquiring reward/penalty in all open-ended tasks isunrealistic, making the Reinforcement Learning (RL) training procedureimpossible. In this paper, we propose a novel cross-modal contrastive learningframework architecture, CLIP4MC, aiming to learn an RL-friendly vision-languagemodel that serves as a reward function for open-ended tasks. Therefore, nofurther task-specific reward design is needed. Intuitively, it is morereasonable for the model to address the similarity between the video snippetand the language prompt at both the action and entity levels. To this end, amotion encoder is proposed to capture the motion embeddings across differentintervals. The correlation scores are then used to construct the auxiliaryreward signal for RL agents. Moreover, we construct a neat YouTube datasetbased on the large-scale YouTube database provided by MineDojo. Specifically,two rounds of filtering operations guarantee that the dataset covers enoughessential information and that the video-text pair is highly correlated.Empirically, we show that the proposed method achieves better performance on RLtasks compared with baselines.

CLIP4MC: An RL-Friendly Vision-Language Model for Minecraft