y0news
AnalyticsDigestsSourcesTopicsRSSAICrypto

#self-preference News & Analysis

1 article tagged with #self-preference. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.

1 articles
AIBullisharXiv – CS AI · Feb 276/107
🧠

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

Researchers introduce Duel-Evolve, a new optimization algorithm that improves LLM performance at test time without requiring external rewards or labels. The method uses self-generated pairwise comparisons and achieved 20 percentage points higher accuracy on MathBench and 12 percentage points improvement on LiveCodeBench.