thu-ml / thu-ml/TurboDiffusion

Strategy for applying rCM with Autoregressive Student

Open
#126 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
3.7k
Forks
277
Avg merge
2h 57m
Merged PRs (30d)
2

Description

Hi, thanks for your great work!

I saw causal models were on the roadmap.

For rCM, the DMD component can easily be replaced with Causal Forcing or Self Forcing to keep the rollout on manifold. However I was wondering what the planned approach was for adjusting the sCM loss component so the student can be conditioned on its own rollout? Will using a bidirectional teacher and a causal student conditioned on its own generated sequence be stable for sCM?

Is sCM too sensitive to conditioning mismatch, requiring one of these?

  • a different method (sCT?)
  • a teacher initialized on self-generated sequences (e.g. resampling forcing)
  • perhaps for the sCM path the student and teacher don't need a self-generated sequence (teacher forced ground truth for both models)?
  • something else?

Appreciate any insights.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

No file, test, or entry point is named. Start by locating the repository's roadmap and the existing rCM, sCM, DMD, Causal Forcing, and Self Forcing implementations, then compare how teacher and student conditioning are handled; done requires an agreed, stable strategy for autoregressive Student Conditioning Matching.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.