Cognition · Software component

Quality-Weighted Vote Aggregator

Software componentCognitionCognition & Memoryarc:QualityWeightedVoteAggregator

A self-consistency aggregator that weights each sampled path's vote by its reasoning-quality score, and can average stated probabilities weighted by quality.

Responsibility. Selects the answer by reasoning-quality-weighted voting across samples.

Also known as: Reasoning-Aware Self-Consistency (RASC) aggregator, Weighted voting

Variant of Self-Consistency Aggregator abstract

When to choose. Choose for medium-to-hard problems with larger sample counts (k≈10-40) or cost-constrained high-stakes deployments where extracting more signal from fewer samples justifies ~5% quality-scoring overhead.

Direct neighbourhood (hover for relationship types)

Relationships

receives data from dynamic

Reasoning Path Quality Classifier Ch5.3

alternative to variability

Design guidance

SHOULD be combined with difficulty-adaptive sample allocation to minimise total samples.

Quantitative guidance

As stated by the sources; verify before use.

RASC with k=10 matches standard Self-Consistency at k=40 (~70% sample reduction) at ~5% extra tokens per sample (Ch5.3).
RASC plus adaptive k averages k=6-8 samples for accuracy comparable to k=40, cutting cost 80-85% (Ch5.3).
Quality weighting raises effective sample efficiency 2-3x at k=15-40 (Ch5.3).

Classification

Patterns: Reasoning-Aware Self-Consistency (RASC)Quality-weighted probability averaging
Quality attributes: Performance efficiency (ISO/IEC 25010)Cost efficiencyFunctional suitability: correctness and validity (ISO/IEC 25010 | NIST AI RMF: valid)
Risks mitigated: Low-quality paths with lucky answers dominating the voteOutlier suggestions from superficial reasoning

Sources

Ch5.3: T. Nguyen, "Self-Consistency Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.3. ISBN: 9798244538229.