dongguanting/Agent-Reflex-RL-2K
Agent-Reflex-RL-2K Paper: arXiv:2610.17953 · Hugging Face Papers The ORPO training tasks of Agent-Reflex: 2,550 executable tool-use tasks (152 MB). Each task carries its executable environment, a checklist verifier and the RSD reflection skills of that environment, which ORPO uses for skill-guided on-policy distillation. Format verl RL format, one task per line: Field Content prompt system + user messages agent_name envscaler_openai_fc… See the full description on the dataset page: https://huggingface.co/datasets/dongguanting/Agent-Reflex-RL-2K.
| Author / organization | dongguanting |
|---|---|
| License (as declared) | mit |
| Downloads | 173 (as reported by Hugging Face) |
| Likes | 2 |
| Created | 9 October 2026 |
| Last updated at source | 14 October 2026 |
Popularity, last 2 days
173 now (+11 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
- mradermacher/Agent-Reflex-8B-i1-GGUFModel · used to train
- Agent-Reflex: Teaching Language Agents to Act ReflectivelyPaper · cites
- dongguanting/Agent-Reflex-8BModel · used to train
- mradermacher/Agent-Reflex-8B-GGUFModel · used to train
Related papers
More from dongguanting
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 Information may be incomplete or out of date.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.