In short: arXiv is a repository for scholarly research articles (in certain scientific disciplines) that allows them to be recorded and distributed widely without paywalls. The articles should be genuine attempts to produce scientific contributions, and be "refereeable" - which means that it should be in a state that it would be suitable for submission to a peer-reviewed venue, but it need not actually have been submitted to one, nor does it need to have been accepted by one.
Papers on arXiv are not required to be published anywhere (conference, journal, or otherwise) - some are submitted before even being sent to a venue for review at all, some are in review progress, some may have been rejected by peer review and may or may not be in the process of revision for sending to a new venue, etc.
You may want to read about arXiv a bit more, such as here https://arxiv.org/help/general (see the disclaimer):
Disclaimer: Papers will be entered in the listings in order of receipt
on an impartial basis and appearance of a paper is not intended in any
way to convey tacit approval of its assumptions, methods, or
conclusions by any agent (electronic, mechanical, or other). We
reserve the right to reject any inappropriate submissions.
And you may wish to read the primer: https://arxiv.org/help/primer
In practice, papers found on the arXiv are in every state of the process, from before-submission, to accepted-but-not-yet-appearing, to being the home for the pre-print or even final version of the article appearing in some peer-reviewed venue (see How many arXiv papers are uploaded in their final (refereed) versions? for more on that). And you'll even find some "submitted but rejected" papers, too - it's not a bug, it's a feature.
It is ultimately up to you as a scholar to decide how to interpret the credibility or usefulness of any given paper. Appearing on arXiv is proof of availability of a paper, which allows you to read it and judge for yourself.