rlcd
Datasets
All datasets matching “rlcd”jev-playground-rlcd-v0
openjev-rlcd-v0
Synthetic calibrated-decision dataset for training Jev-style System One classifiers plan). States are generated programmatically per domain; typed questions follow the openjev /v1/systemone contract; reference distributions are exact by construction, the property a proper-scoring-rule / calibration objective needs.
Format
JSONL, one state per line:
{
"domain": "support_tickets",
"state": "...",
"questions": {
"urgent": {"type": "noul"… See the full description on the dataset page: https://huggingface.co/datasets/soyrsoyr/jev-playground-rlcd-v0.RLCDAlignBench
RLCDAlignBench
Paper: Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures (arXiv:2609.29429)
Code: github.com/sumleo/RLCDAlignBench · Project page: sumleo.github.io/RLCDAlignBench
RLCDAlignBench measures whether a detector can tell when a language model's output is an alignment failure.
It has 44 benchmarks across ten failure types and five target models (Qwen3.5-2B, Phi-4-mini, Gemma-2-2B, Llama-3.2-3B, Olmo-3-7B), for… See the full description on the dataset page: https://huggingface.co/datasets/sumleo/RLCDAlignBench.rlcd-decision-v1
RLCD Decision Dataset (v1)
Typed decision questions for training and evaluating models that answer with a calibrated probability distribution over declared options instead of generated text. Built for RLCD — reinforcement learning for calibrated decisions (see also the trained export anthonym21/qwen3-0.6b-rlcd-decision).
Every row is one typed question over a context: a choice question over unordered options, a score question over ordered levels, or a noul (yes/no) question. The… See the full description on the dataset page: https://huggingface.co/datasets/anthonym21/rlcd-decision-v1.rlcd
Dataset Card for "rlcd"
More Information needed
eve-rlcd-runsRLCD-generated-preference-data-split
Dataset Card for "RLCD-generated-preference-data-split"
More Information needed