add new inference paradiam

[tangming1996]

What type of PR is this?
/kind documentation

What this PR does / why we need it:
The current joint inference paradigm can only support the traditional discriminative model and cannot support the large language model (LLM). This proposal aims to introduce a new inference paradigm to support the joint inference of LLM
Which issue(s) this PR fixes:

Fixes #

[MooreZheng]

In this Era, it is critical and urgent to introduce LLM in Sedna. The new paradigm shown in this proposal is thus important.

As discussed, since the Sedna lib is removed from the paradigm, we need several serious discussions on this proposal.

0. Sedna and Ianvs paradigms for LLM joint inference should be consistent - might discussed with @FuryMartin and @hsj576

1. How will this new paradigm improve the integration of LLM models by removing Sedna lib in cloud and edge inference?

• Will that bring concerns for the data interfaces? When switching to a new model, people need to switch the data format to the new, leading to an additional cost.
• Will this paradigm support a non-LLM model on the edge? It seems that this paradigm only works in the case where both cloud and edge deploy LLMs, instead of cloud-only LLM and edge-only LLM, as claimed in the proposal.

2. How will this new paradigm improve the integration of hard example mining by removing Sedna lib in hard example mining (sider car)?

• What is the hard example mining algorithm used now in the new paradigm? How to switch hard example mining algorithms?

3. With the above consideration, we wonder whether this paradigm fully supports LLM joint inference. The new paradigm seems more like a new paradigm of LLM single-task learning, doesn't it? That is, a trained LLM is deployed either on the edge or on the cloud.

[kubeedge-bot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
To complete the pull request process, please assign moorezheng after the PR has been reviewed.
You can assign the PR to them by writing /assign @moorezheng in a comment when ready.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• docs/OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment