THUDM / THUDM/slime

[feature] Implementing ChunkFlow for really long sequence

Open
#697 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
8.5k
Forks
1.3k
Avg merge
5h 36m
Merged PRs (30d)
22

Description

chunkflow (https://arxiv.org/abs/2503.02356) could be a really good help for long sequence RL as it allows for almost infinite sequence length with kv cache offload without adding GPUs. For some context, please refer to https://zhuanlan.zhihu.com/p/1969847127564871140

Contributor guide

Open the contributing guide

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

The issue names no files, tests, or entry points. Start by reading the linked ChunkFlow paper and the referenced context, then locate how slime handles long-sequence RL and KV-cache offload. Done should mean ChunkFlow is integrated for the described long-sequence use case, with project validation covering the behavior.

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
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.