Reflective Confidence: Correcting Reasoning Flaws via Online Self-Correction

TL;DR

Reflective Confidence introduces a self-correction mechanism for large language models that uses confidence signals to trigger reflections and improve reasoning accuracy without additional computational overhead.

Contribution

The paper proposes a novel reflective confidence framework that transforms low-confidence signals into reflection prompts, enabling proactive self-correction during reasoning tasks.

Findings

Significant accuracy improvements on mathematical reasoning benchmarks.

Outperforms early-stopping baselines at similar computational costs.

Validates proactive self-correction as an effective strategy.

Abstract

Large language models (LLMs) have achieved strong performance on complex reasoning tasks using techniques such as chain-of-thought and self-consistency. However, ensemble-based approaches, especially self-consistency which relies on multiple reasoning trajectories, often incur substantial computational overhead. To improve efficiency, prior work has leveraged internal confidence signals, where early stopping strategies such as DeepConf reduce cost by terminating low-confidence trajectories. However, this strategy discards incomplete reasoning paths and wastes partial computation. We propose reflective confidence, a novel reasoning framework that transforms low-confidence signals from termination indicators into reflection triggers. When confidence falls below a threshold, instead of stopping generation, the model produces a reflection prompt to analyze the current reasoning state, identify potential errors, and continue generation along a corrected trajectory. Experiments on mathematical reasoning benchmarks, including AIME 2025, demonstrate significant accuracy improvements over advanced early-stopping baselines at comparable computational cost, validating the effectiveness of proactive self-correction over passive discarding.