Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

Duel-Evolve: Reward-Free Test-Time Scaling via LLM Self-Preferences

Sweta Karlekar, Carolina Zheng, Magnus Saebo, Nicolas Beltran-Velez +4 more

Published

Feb 25, 2026

Citations

0

Trust level

High

Usefulness score

65/100 (Medium)

Extraction confidence

80% (High)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Feb 25, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics Math Pairwise Preference

Should you rely on this paper?

This paper has useful evaluation signal, but protocol completeness is partial; pair it with related papers before deciding implementation strategy.

Use this as a practical starting point for protocol research, then validate against the original paper.

Best use

Secondary protocol comparison source

Use if you need

A benchmark-and-metrics comparison anchor.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

No major weakness surfaced.

Human feedback signal
Detected
From extracted signals
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
65/100
Moderate-confidence candidate

Useful as a secondary reference; validate protocol details against neighboring papers.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

Many applications seek to optimize LLM outputs at test time by iteratively proposing, scoring, and refining candidates over a discrete output space. Existing methods use a calibrated scalar evaluator for the target objective to guide search, but for many tasks such scores are unavailable, too sparse, or unreliable. Pairwise comparisons, by contrast, are often easier to elicit, still provide useful signal on improvement directions, and can be obtained from the LLM itself without external supervision. Building on this observation, we introduce Duel-Evolve, an evolutionary optimization algorithm that replaces external scalar rewards with pairwise preferences elicited from the same LLM used to generate candidates. Duel-Evolve aggregates these noisy candidate comparisons via a Bayesian Bradley-Terry model, yielding uncertainty-aware estimates of candidate quality. These quality estimates guide allocation of the comparison budget toward plausible optima using Double Thompson Sampling, as well as selection of high-quality parents to generate improved candidates. We evaluate Duel-Evolve on MathBench, where it achieves 20 percentage points higher accuracy over existing methods and baselines, and on LiveCodeBench, where it improves over comparable iterative methods by over 12 percentage points. Notably, the method requires no reward model, no ground-truth labels during search, and no hand-crafted scoring function. Results show that pairwise self-preferences provide strong optimization signal for test-time improvement over large, discrete output spaces.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

strong

Pairwise Preference

Directly usable for protocol triage.

"Many applications seek to optimize LLM outputs at test time by iteratively proposing, scoring, and refining candidates over a discrete output space."

Evaluation Modes

strong

Automatic Metrics

Includes extracted eval setup.

"Many applications seek to optimize LLM outputs at test time by iteratively proposing, scoring, and refining candidates over a discrete output space."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Many applications seek to optimize LLM outputs at test time by iteratively proposing, scoring, and refining candidates over a discrete output space."

Benchmarks / Datasets

strong

LiveCodeBench, Mathbench

Useful for quick benchmark comparison.

"We evaluate Duel-Evolve on MathBench, where it achieves 20 percentage points higher accuracy over existing methods and baselines, and on LiveCodeBench, where it improves over comparable iterative methods by over 12 percentage points."

Reported Metrics

strong

Accuracy

Useful for evaluation criteria comparison.

"We evaluate Duel-Evolve on MathBench, where it achieves 20 percentage points higher accuracy over existing methods and baselines, and on LiveCodeBench, where it improves over comparable iterative methods by over 12 percentage points."

Benchmarks and datasets

LiveCodeBenchMathbench

Reported metrics

accuracy
Human feedback details
Uses human feedback
Yes
Feedback types
Pairwise Preference
Rater population
Not reported
Unit of annotation
Pairwise
Expertise required
Math
Evaluation details
Evaluation modes
Automatic Metrics
Agentic eval
None
Quality controls
Not reported
Evidence quality
High
Use this page as
Secondary protocol comparison source

Research brief

Metadata summary

Many applications seek to optimize LLM outputs at test time by iteratively proposing, scoring, and refining candidates over a discrete output space.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • Many applications seek to optimize LLM outputs at test time by iteratively proposing, scoring, and refining candidates over a discrete output space.
  • Existing methods use a calibrated scalar evaluator for the target objective to guide search, but for many tasks such scores are unavailable, too sparse, or unreliable.
  • Pairwise comparisons, by contrast, are often easier to elicit, still provide useful signal on improvement directions, and can be obtained from the LLM itself without external supervision.

Researcher actions

  • Compare this paper against others mentioning LiveCodeBench.
  • Validate inferred eval signals (Automatic metrics) against the full paper.
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

Pairwise preference evaluation LiveCodeBench Accuracy

Contribution summary

  • Building on this observation, we introduce Duel-Evolve, an evolutionary optimization algorithm that replaces external scalar rewards with pairwise preferences elicited from the same LLM used to generate candidates.
  • We evaluate Duel-Evolve on MathBench, where it achieves 20 percentage points higher accuracy over existing methods and baselines, and on LiveCodeBench, where it improves over comparable iterative methods by over 12 percentage points.
  • Results show that pairwise self-preferences provide strong optimization signal for test-time improvement over large, discrete output spaces.

Why it matters for eval

  • Building on this observation, we introduce Duel-Evolve, an evolutionary optimization algorithm that replaces external scalar rewards with pairwise preferences elicited from the same LLM used to generate candidates.
  • Results show that pairwise self-preferences provide strong optimization signal for test-time improvement over large, discrete output spaces.

Researcher checklist

  • Human feedback protocol is explicit

    Detected: Pairwise Preference

  • Evaluation mode is explicit

    Detected: Automatic Metrics

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    Detected: LiveCodeBench, Mathbench

  • Metric reporting is present

    Detected: accuracy