Skip to content

Add CCLRouter: causal representation routing with RouterBench pretrai…#153

Open
kiotori wants to merge 4 commits into
RouteWorks:mainfrom
kiotori:add-cclrouter
Open

Add CCLRouter: causal representation routing with RouterBench pretrai…#153
kiotori wants to merge 4 commits into
RouteWorks:mainfrom
kiotori:add-cclrouter

Conversation

@kiotori

@kiotori kiotori commented Jun 21, 2026

Copy link
Copy Markdown

…ning

  • Causal encoder (c=128d) pretrained on RouterBench (36K queries, 11 models, 86 domains)
  • Lightweight LogisticRegression head (128->23) fine-tuned on RouterArena sub_10 (809 queries)
  • Routes by argmax P(correct|c): domain-invariant task representation enables efficient few-shot routing
  • Arena=73.82 on sub_10 training protocol, beating R2-Router (69.97) by +3.85pt
  • See PR for full description and evaluation results

…ning

- Causal encoder (c=128d) pretrained on RouterBench (36K queries, 11 models, 86 domains)
- Lightweight LogisticRegression head (128->23) fine-tuned on RouterArena sub_10 (809 queries)
- Routes by argmax P(correct|c): domain-invariant task representation enables efficient few-shot routing
- Arena=73.82 on sub_10 training protocol, beating R2-Router (69.97) by +3.85pt
- See PR for full description and evaluation results

Co-Authored-By: Claude <noreply@anthropic.com>
@kiotori

kiotori commented Jun 21, 2026

Copy link
Copy Markdown
Author

/evaluate

@kiotori

kiotori commented Jun 22, 2026

Copy link
Copy Markdown
Author

/evaluate

@kiotori

kiotori commented Jun 22, 2026

Copy link
Copy Markdown
Author

/evaluate

@kiotori

kiotori commented Jun 22, 2026

Copy link
Copy Markdown
Author

/evaluate

@yl231

yl231 commented Jun 23, 2026

Copy link
Copy Markdown
Contributor

Hi @kiotori, just to let you know, fine-tuning on any portion of RouterArena (including the sub_809 split) is not allowed, as specified in the README.md.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants