datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
poder_judiciallegal-judicial-doctrinal-consistency-drift-v0.1What this dataset is
You receive
prior opinion pattern
current opinion pattern
citation shift
test or factor weighting shift
explanation quality
external reaction signals
You decide
Does the judge remain doctrinally consistent
Answer
coherent
or
incoherent
Why this matters
Doctrinal drift predicts
dissents and fractures
en banc pressure
reversal risk
loss of precedential durability
legal-judicial-sentencing-coherence-drift-v0.1What this dataset is
You receive
offence facts
factors invoked
guideline anchor
sentence imposed
departure rationale
review signals
You decide
Does the sentence match the stated factors and anchor
Answer
coherent
or
incoherent
Why this matters
When sentencing coherence breaks
appeals rise
remands rise
disparity grows
trust collapses
legal-judicial-reasoning-chain-coherence-v0.1Clarus Judicial Reasoning Chain Coherence v0.1
This dataset tests whether a model can detect structural breakdown inside a judicial opinion.
The focus is not outcome prediction.
The focus is reasoning integrity.
Does the chain from facts to remedy hold together.
Or does it fracture before the conclusion.
Core structure tested
facts
rule
application
holding
remedy
Courts sometimes reach a result that the reasoning cannot support.
That is coherence decay.
This dataset detects that.
Task
Input… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/legal-judicial-reasoning-chain-coherence-v0.1.judicialhelplegal-judicial-panel-coherence-convergence-v0.1What this dataset is
You receive
majority frame
concurrence frame
dissent frame
fracture signals
review signals
You decide
Does the panel converge on the same core legal frame
Answer
coherent
or
incoherent
Why this matters
Fractured panels predict
en banc risk
higher court review risk
weak precedent
unstable doctrine