Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
coercion – Dataset by kargaranamir | AlphaNeural AI
You can deploy this model and start earning money today!
kargaranamir
/
coercion
like
0
100K<n<1M
parquet
optimized-parquet
text
datasets
dask
polars
mlcroissant
us
llm-robustness
belief-collapse
adversarial-reasoning
pluralistic-alignment
mmlu
meta-llama-Llama-3.3-70B-Instruct
Views
No views yet
Model card
Files and Versions
Community
API
Pressure-Coerced Self-Contradiction — mmlu
Paper: "When a Model Turns on Itself: Pressure-Coerced Reasoning as a Robustness Probe"
Parameters
Source mmlu
Model meta-llama/Llama-3.3-70B-Instruct
Pressure high
Reasoning length 1 sentences
Attribution self
Rows 2,052
Time 501.1 min
Date 2026-04-26 20:30 UTC
Key Results
Metric Value
Baseline correct 1,738 / 2,052
Any coercion succeeded 2,052 / 2,052
BCR 83.1% (1444… See the full description on the dataset page:
https://huggingface.co/datasets/kargaranamir/coercion
.
