huggingface / huggingface/carbon
Verify evals on Papers with Code
- Dominant language
- Python
- Stars
- 214
- Forks
- 27
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
Niels here from the open-source team at Hugging Face. Congratulations on your work!
I've made the [paper](https://paperswithcode.co/paper/83340) and [8 paper-native evaluations](https://paperswithcode.co/paper/83340#results) available on Papers with Code.
The paper is part of the [Biology](https://paperswithcode.co/tasks/biology) task page.
The Carbon-8B results currently rank first on [Genomic-NIAH](https://paperswithcode.co/benchmark/genomic-niah?task=biology&eval=5760) and [Sequence Recovery](https://paperswithcode.co/benchmark/sequence-recovery?task=biology&eval=5758).
The Carbon-8B (synonymous codon) result currently ranks first on [Perturbation Bench](https://paperswithcode.co/benchmark/perturbation-bench?task=biology&eval=5764).
The Carbon-3B result currently ranks second on [Genomic-NIAH](https://paperswithcode.co/benchmark/genomic-niah?task=biology&eval=5759).
Would it be possible to verify these results and let me know if any score, model name, benchmark protocol, or openness metadata should be corrected? The imported rows are tied to the paper or its official release artifacts; comparison-table baselines were not added.
You can also edit the task, methods, project page, and GitHub URL directly from the paper page using your Hugging Face account.
If you'd like to showcase the results in your repository README, you can copy these live leaderboard badges (or use the “Copy PwC badge” button in the Results section):
[](https://paperswithcode.co/benchmark/genomic-niah?task=biology&eval=5760)
[](https://paperswithcode.co/benchmark/perturbation-bench?task=biology&eval=5764)
[](https://paperswithcode.co/benchmark/sequence-recovery?task=biology&eval=5758)
Kind regards,
Niels
Contributor guide
No contributing guide indexed for this repository
Research direction
Open the linked Papers with Code paper and the four benchmark result pages, then compare their scores, model names, protocols, and openness metadata with the paper and official release artifacts. Done means reporting any corrections needed, or confirming that the imported evaluations are accurate; README badges are optional.
Written by the indexing model from the issue text.
Assessment
- Domain
- documentation
- Issue type
- Documentation
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 55/100