Quantum-Inspired Trace-Augmented Evidence Selection for Reasoning over Structured Hypothesis Spaces

2026-06-05

AI Summary

This work addresses the limitations of large language models in evidence-intensive reasoning tasks such as legal judgment, where performance is often undermined by biased evidence selection and the suppression of high-quality minority viewpoints under majority voting schemes. To overcome these challenges, the authors propose EP-HUBO, a novel framework that integrates high-order unconstrained binary optimization (HUBO) with quantum-inspired computation. EP-HUBO constructs hypothesis-specific evidence pools and performs fine-grained evidence evaluation and aggregation by weighting evidence according to relevance, specificity, and discriminability. This approach enables high-quality minority hypotheses to prevail, thereby transcending the constraints of conventional majority voting. Experimental results demonstrate that EP-HUBO significantly improves accuracy on two legal reasoning benchmarks, with particularly strong performance in low-contamination, non-overfitted settings involving state-of-the-art models.

Abstract

Large language models (LLMs) now solve a wide range of expert-level exams at or above human level, yet remain brittle on specialised, evidence-intensive domains such as law. On these tasks, errors arise not only from gaps in world knowledge but also from subtle distinctions between pieces of evidence and inconsistent use of supporting evidence. The most common aggregator over sampled chain-of-thought (CoT) traces, majority vote, returns the most popular answer regardless of whether its evidence is actually strongest. We propose to treat the selection of CoT reasoning fragments into a set of evidence as an explicit combinatorial optimisation problem, allowing well-supported but minority hypotheses to override noisy majorities, and to evaluate the approach on legal-reasoning benchmarks that are particularly sensitive to evidence quality. We introduce EP-HUBO (Evidence Pool Higher-Order Binary Optimisation), which generates multiple CoT traces with a small local model, parses fragments into per-hypothesis evidence pools, solves a higher-order unconstrained binary optimisation per pool with quality-derived weights (relevance, specificity, distinctiveness), and delegates a single adjudication call per question to a frontier model. We evaluate EP-HUBO on two evidence-intensive legal benchmarks using both simulated annealing on classical hardware and the Dirac-3 photonic entropy-quantum machine from Quantum Computing Inc. HUBO-style optimisation gives a principled way to aggregate reasoning fragments while preserving minority-but-correct hypotheses, and is most valuable in low-contamination domains where frontier models have not already absorbed the benchmark material.

Problem

• evidence selection
• legal reasoning
• chain-of-thought
• hypothesis aggregation
• evidence-intensive reasoning

Innovation

• evidence selection
• higher-order binary optimisation
• chain-of-thought reasoning
• quantum-inspired optimisation
• legal reasoning