Datasets:
Formats:
json
Languages:
English
Size:
< 1K
Tags:
reinforcement-learning
scientific-discovery
multi-turn
procedural-generation
tool-use
verifiable-rewards
License:
|
Download README.md from FlaneAI/Scientific-Research-RL: direct link, hf CLI and curl.
- Browser
- Download file 2.77 kB
-
https://huggingface.co/datasets/FlaneAI/Scientific-Research-RL/resolve/main/README.md
- Command line
-
hf download hf://datasets/FlaneAI/Scientific-Research-RL/README.md
-
curl -L -o README.md https://huggingface.co/datasets/FlaneAI/Scientific-Research-RL/resolve/main/README.md
2.77 kB
metadata
pretty_name: ResearchForge RL — Scientific Discovery Environments
language:
- en
license: mit
tags:
- rl-environment
- reinforcement-learning
- scientific-discovery
- multi-turn
- procedural-generation
- tool-use
- verifiable-rewards
size_categories:
- n<1K
ResearchForge RL v0.5.0 — Hugging Face RL Environment Repository
A multi-turn, verifiable scientific-discovery taskset and Python runtime for LLM agents.
Compatibility: This repository uses the Hub's
rl-environmenttag. It ships a working native Python runtime, not a verified OpenEnv, Verifiers, Harbor or NeMo Gym integration. We intentionally do not claim those framework tags. You can build adapters on top ofResearchSession.
Install and run locally
pip install -e '.[dev]'
python scripts/smoke_test.py
pytest -q
from researchforge.hub import load_tasks, ResearchSession
task = load_tasks('train')[0]
session = ResearchSession(task)
state = session.reset()
print(state['instructions'])
print(state['action_schema'])
result = session.step({'action': 'observe', 'n': 2})
print(result)
The model receives observations and JSON action descriptions. Repeat session.step(action) until terminated or truncated; terminal scores are computed by the simulator, not a language-model judge.
Train / evaluate correctly
- Use
traintasks for SFT, preference optimization and RL rollouts. Usevalfor selection; reservetestfor final reporting. - Public seeds and source code are not a secure blind benchmark. A malicious or source-aware agent could reconstruct worlds. For meaningful held-out evaluation, run the environment on a trusted machine, provide a private
evaluator_secretinResearchSession, keep it off the Hub, and use newly sampled private task seeds. The private evaluator must not expose world objects, source files or oracle details to the model. - Do not publish private salts, oracle states, or answers. Do not train on test tasks.
- No LLM weights are trained by this repository. The example policies are hand-coded baselines, not neural training results.
- The simulator is research software, not evidence of real-world scientific discovery.
Repository structure
data/{train,val,test}.jsonl: task descriptors (not answers)researchforge/hub.py:load_tasks,ResearchSession,rolloutresearchforge/env.py: underlying step/reset/reward engineresearchforge/server.py: isolated-process JSONL protocol (not a hardened security sandbox)researchforge/training_data.py: SFT / preference data utilitiesresearchforge/evaluation.py: paired evaluation utilitiesexamples/: agent and benchmark examplestests/: unit and integration tests
License
MIT. See LICENSE.