Sign inGet started
Dataset · Hugging Face

hkust-nlp/drkernel-rl-data

DR.Kernel RL Dataset Paper | Code | Dataset This directory documents the format of hkust-nlp/drkernel-rl-data. Unlike the cold-start SFT set, this RL dataset is primarily a single-turn query pool (plus reference metadata) used to launch multi-turn rollouts online in KernelGYM. Overview Purpose: provide RL training queries + reference code metadata for reward evaluation. Data form: one row per optimization task. Current local Parquet… See the full description on the dataset page: https://huggingface.co/datasets/hkust-nlp/drkernel-rl-data.

Author / organizationhkust-nlp
License (as declared)mit
Downloads145 (as reported by Hugging Face)
Likes1
Created5 February 2026
Last updated at source6 February 2026
reinforcement-learningtritonkernel-generation
Embed this entry
<iframe src="https://silytics.si/embed/item/771c62db-d110-44ea-ada7-8f676b9d10d8" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 2 days

145 now (-1 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

Connected in the knowledge graph

Explore this entry in the graph

Recent changes

  • 16 Oct 2026 Description: DR.Kernel RL Dataset Paper | Code | Dataset This directory documents the format of hkust-nlp/drkernel-rl-data. Unlike the cold-start SFT set, this RL dataset is primarily a single-turn query pool (plus reference metadata) used to launch mul → DR.Kernel RL Dataset Paper | Code | Dataset This directory documents the format of hkust-nlp/drkernel-rl-data. Unlike the cold-start SFT set, this RL dataset is primarily a single-turn query pool (plus reference metadata) used to launch mul

Changes detected by Silytics between refreshes of the source.

More from hkust-nlp

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.