English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Consilience: A Confidence-Trajectory Framework for Verifier-Free Test-Time Scaling

Forum topic · 小凯 · 2026-08-12

Summary

This arXiv paper (2508.05137) by Lecheng Kong, Like Hui, and Haitao Mao addresses verifier-free test-time scaling (VF-TTS) for large language models. Confidence-based VF-TTS methods rank rollouts purely by model confidence with near-zero overhead and minimal access to internal states, but the authors identify a critical flaw: these methods catastrophically fail on complex tasks, where uniformly high confidence often signals failed exploration and confidently wrong answers. The key insight is that robust epistemic search requires a specific confidence trajectory: exploratory branching with low initial confidence that converges to high final confidence. To operationalize this, the paper introduces Consilience, a selection framework that explicitly evaluates the temporal asymmetry of confidence during reasoning via a composite metric that penalizes high initial confidence while strictly requiring final certainty. Extensive experiments on graduate-level math problems and free-form code generation show that Consilience outperforms existing baselines, validating this new perspective on completion confidence.

Paper Overview

Field: NLP Authors: Lecheng Kong, Like Hui, Haitao Mao Published: 2026-08-12 arXiv: 2508.05137

Abstract

Test-time scaling often uses an external verifier, such as compilers and test cases in coding or trained value functions in robotics applications, to obtain high-quality rollouts. Verifier-free test-time scaling (VF-TTS) is gaining extensive attention as a mechanism to enhance Large Language Model (LLM) reasoning, primarily because we do not have access to such high-quality verifiers in many real-world applications. Among existing VF-TTS methods, confidence-based VF-TTS methods, which compute and rank rollouts solely by confidence, are particularly promising. Such methods introduce near-zero overhead for sample evaluation and require minimal access to internal model states, making the methods highly flexible across models and tasks.

In this paper, the authors demonstrate a critical limitation of existing confidence-based VF-TTS methods: they catastrophically collapse on complex tasks. A notable observation is that uniformly high confidence typically indicates failed exploration, tending toward confidently wrong answers.

Key Idea

The core insight is that robust epistemic search requires a specific confidence trajectory pattern: methods should perform exploratory branching at the start (manifesting as low initial confidence) and converge to high final-confidence solutions.

Method: Consilience

To realize this insight, the authors introduce consilience, a novel selection framework that explicitly evaluates the temporal asymmetry of confidence during reasoning. It is implemented through a composite metric that:

  • Actively penalizes high initial confidence
  • Strictly requires final certainty

Results

Extensive experiments on graduate-level math problems and free-form code generation show that consilience effectively outperforms existing baselines, validating the novel perspective on completion confidence.

---

*Automatically collected on 2026-08-12*

Tags

#llm#test-time-scaling#verifier-free#confidence#reasoning#nlp#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178633379