Datasets:

ClarusC64
/

cross-domain-transfer-validity-stress-test-v0.1

id stringclasses 10 values	pair_id stringclasses 10 values	domain_a stringclasses 10 values	domain_b stringclasses 10 values	source_mechanism stringclasses 10 values	proposed_analogue stringclasses 10 values	invariant_mapping stringclasses 10 values	invalid_transfer_risks stringclasses 10 values	confounder_list stringclasses 10 values	minimal_validation_plan stringclasses 10 values	transfer_confidence_0_100 int64 68 78	stress_test_evidence stringclasses 10 values	constraints stringclasses 1 value	gold_checklist stringclasses 1 value
TVST-001	P-01	fibrosis prevention in wound healing	PTSD prevention after trauma	early inflammatory dampening reduces myofibroblast persistence and scar rigidity	early hyperarousal dampening reduces rigid memory consolidation and chronic PTSD	amplification→hardening; persistent activation→rigid remodeling	analogy fails if trauma exposure continues; psychological hardening may arise without arousal overshoot; intervention timing window may differ	social support, prior trauma load, substance use, sleep deprivation	RCT or matched cohort: early arousal-dampening protocol within 72h vs control; outcomes PTSD severity at 3 and 6 months; mediators sleep and arousal markers	78	Both have early amplification and later remodeling, but Domain B has strong context drivers and measurement noise	Under 260 words.	risks+confounders+plan+confidence
TVST-002	P-02	immune tolerance induction in allergy	phobia exposure therapy	graded antigen exposure retrains immune response toward tolerance	graded exposure retrains threat prediction toward safety	graded exposure schedule; reactivity recalibration	fear extinction is context-bound and can return; immune tolerance may be more durable; patient dropout alters trajectory	therapist quality, avoidance severity, comorbid depression, medication changes	Minimal test: compare graded vs massed exposure schedule; measure relapse at 1, 3, 6 months; track context generalization tasks	74	Transfer holds at schedule level but durability properties differ between domains	Under 260 words.	risks+confounders+plan+confidence
TVST-003	P-03	asthma rescue window dynamics	panic attack rescue window dynamics	early bronchodilation interrupts bronchospasm positive feedback	early grounding interrupts autonomic arousal positive feedback	positive feedback loop; early interruption window	panic triggers can be internal and self-sustaining; asthma triggers often external; grounding may not change physiology in severe panic	caffeine, sleep loss, interoceptive sensitivity, respiratory alkalosis	Minimal test: time-to-intervention vs symptom duration; randomized prompt to grounding within 2 minutes vs later; outcome recurrence over 4 weeks	76	Loop structure maps, but trigger source and physiology coupling differ	Under 260 words.	risks+confounders+plan+confidence
TVST-004	P-04	microbiome recolonization after antibiotics	sleep rhythm restoration after burnout	guided recolonization restores stable gut ecosystem	guided re-entrainment restores stable circadian rhythm	attractor restoration; stability through repeated inputs	sleep is strongly shaped by environment and intent; microbiome shaped by ecology; analogy fails if burnout includes psychiatric comorbidity	light exposure, work schedule, depression, alcohol	Minimal test: fixed wake time + morning light vs control; measure sleep regularity and fatigue; compare to microbiome stability metric conceptually	68	Shared attractor idea is valid but intervention levers differ in controllability	Under 260 words.	risks+confounders+plan+confidence
TVST-005	P-05	stroke motor rehab learned nonuse	aphasia language relearning compensatory avoidance	constraint-induced therapy prevents learned nonuse	forced language use prevents compensatory shortcuts	plasticity requires forced engagement; avoid compensation	language has social cost and fatigue; forced use may reduce participation; plateau meanings differ	caregiver support, education, depression, aphasia severity	Minimal test: intensive practice block vs standard; measure functional language and participation at 4 and 12 weeks; track fatigue and dropout	72	Plasticity mapping holds, but compliance pressure and social constraints are stronger in language	Under 260 words.	risks+confounders+plan+confidence
TVST-006	P-06	sepsis early stabilization bundle	delirium prevention bundle	early multi-stream stabilization prevents cascade collapse	early environment and sleep stabilization prevents cognitive collapse	restore coupling across streams early	delirium has multiple etiologies; sedation confounds; measurement of coherence is noisier	medication burden, ICU noise, baseline cognition, infection severity	Minimal test: bundle vs usual care; delirium incidence and duration; mediators sleep metrics and light/noise logs	77	The bundle concept transfers, but causality in delirium is less direct and confounded by drugs	Under 260 words.	risks+confounders+plan+confidence
TVST-007	P-07	bone callus scaffolding	habit scaffolding in behavior change	temporary scaffold supports remodeling then removal	temporary routine supports habit formation then autonomy	scaffold then remodel then remove	habits depend on meaning and social context; scaffold removal timing is subjective	life stress, social support, motivation, rewards	Minimal test: time-limited scaffold routine vs indefinite scaffold; outcomes habit persistence at 3 months; measure stress and context change	70	Structure transfer is plausible but control variables differ and measurement is subjective	Under 260 words.	risks+confounders+plan+confidence
TVST-008	P-08	central sensitization in chronic pain	threat sensitization in chronic anxiety	learning-based amplification converts transient signals into persistent prediction loops	learning-based amplification converts transient stress into persistent threat prediction	prediction bias loop; reinforcement maintains state	pain can have ongoing peripheral drivers; anxiety can persist without external driver; interventions differ	sleep, trauma history, medication, ongoing injury	Minimal test: measure prediction bias markers before and after intervention; compare pain vs anxiety cohorts; outcome symptom persistence and generalization	73	Loop concept transfers, but presence of peripheral drivers can break symmetry	Under 260 words.	risks+confounders+plan+confidence
TVST-009	P-09	tolerance induction via graded exposure	panic disorder interoceptive exposure	graded exposure builds tolerance	graded interoceptive exposure reduces panic reactivity	graded exposure invariant	panic physiology and catastrophizing can overpower exposure; dropout bias	therapist skill, comorbidity, medication	Minimal test: graded schedule vs random exposure; outcomes panic frequency and avoidance; track dropout	71	Transfer is plausible but adherence and cognition confound effect size	Under 260 words.	risks+confounders+plan+confidence
TVST-010	P-10	plateau interpretation in stroke rehab	plateau interpretation in aphasia rehab	plateau may signal consolidation or stalled adaptation	plateau may signal consolidation or stalled adaptation	plateau semantics invariant	plateaus measured differently; language gains appear in bursts; measurement instruments differ	testing frequency, practice quality, fatigue	Minimal test: classify plateau segments and compare outcomes after protocol change vs hold; measure later gains and relapse	69	Concept transfers but measurement mismatch is the main risk	Under 260 words.	risks+confounders+plan+confidence

What this dataset tests

Whether a model can stress-test a cross-domain transfer claim
by identifying invalidity risks, confounders, and a minimal validation plan.

Required outputs

invalid_transfer_risks
confounder_list
minimal_validation_plan
transfer_confidence_0_100

Typical failures

no boundary conditions
confounders not named
validation plan too broad to run
confidence score without justification

Suggested prompt wrapper

System

You stress-test a cross-domain mechanism transfer.

User

Domain A mechanism
{source_mechanism}

Proposed analogue
{proposed_analogue}

Invariant mapping
{invariant_mapping}

Return

invalid transfer risks
confounders
minimal validation plan
transfer confidence 0–100

Citation

ClarusC64 dataset family

Downloads last month: 15

Number of rows:

Total file size:

13.6 kB