Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Length cutoff: SpladePlusPlusEnsembleDistilEncoder vs. SpladePlusPlusSelfDistilEncoder #2760

Closed
lintool opened this issue Mar 17, 2025 · 2 comments
Assignees

Comments

@lintool
Copy link
Member

lintool commented Mar 17, 2025

Why does SpladePlusPlusEnsembleDistilEncoder have a max seq length cutoff?
https://github.com/castorini/anserini/blob/master/src/main/java/io/anserini/encoder/sparse/SpladePlusPlusEnsembleDistilEncoder.java#L69

And SpladePlusPlusSelfDistilEncoder doesn't?
https://github.com/castorini/anserini/blob/master/src/main/java/io/anserini/encoder/sparse/SpladePlusPlusSelfDistilEncoder.java#L67

@ArthurChen189 this is your impl right? Was this just an oversight?

@cadurosar wanted to check with you also... both models should have a 512 length cutoff? So it's a bug in our impl?

@lintool lintool self-assigned this Mar 17, 2025
@ArthurChen189
Copy link
Member

Should be an oversight.

iirc, I added one MAX_SEQ_LEN because only Splade++ ED failed the regression issue (#2368). According to HF config, both models should have MAX_SEQ_LEN=512.

@lintool
Copy link
Member Author

lintool commented Mar 27, 2025

Closed by #2761

@lintool lintool closed this as completed Mar 27, 2025
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

No branches or pull requests

2 participants