{"slug": "which-models-work-well-together-measuring-heterogeneity-for-llm-team-selection", "title": "Which Models Work Well Together? Measuring Heterogeneity for LLM Team Selection", "summary": "A new arXiv paper (2609.38274v1) proposes a heterogeneity-driven team selection framework for multi-LLM systems that profiles each model's capability plus two complementary signals: error-pattern decorrelation to reduce co-failures and predictive-behavior divergence to capture strategy diversity. The authors formulate team selection as a standardized quality-complementarity combinatorial objective solved with greedy search, and report that the framework consistently outperforms quality-only baselines across multiple benchmarks under controlled candidate pools and team sizes.", "body_md": "arXiv:2609.38274v1 Announce Type: new \nAbstract: The performance ceiling of an LLM team is constrained not only by individual model capabilities, but also by inter-member error resonance and predictive differences. Although heterogeneous teaming is often observed to be effective in practice, existing approaches lack complementarity metrics that are computable, interpretable, and optimizable, leaving team composition to rely on heuristics. We propose a heterogeneity-driven team selection framework that performs offline profiling to characterize individual capability along with two complementary signals: one captures decorrelation in error patterns to reduce co-failures, while the other measures divergence in predictive behavior to capture strategy diversity. We formulate team selection as a standardized quality--complementarity combinatorial objective and apply an efficient greedy search to select a small team from a candidate pool. Experiments across multiple benchmarks demonstrate that our framework consistently outperforms quality-only baselines under controlled candidate pools and team sizes, establishing reusable selection principles for multi-LLM systems.", "url": "https://wpnews.pro/news/which-models-work-well-together-measuring-heterogeneity-for-llm-team-selection", "canonical_source": "https://arxiv.org/abs/2609.38274", "published_at": "2026-10-01 04:00:00+00:00", "updated_at": "2026-10-01 04:19:16.095151+00:00", "lang": "en", "topics": ["large-language-models", "machine-learning", "ai-research", "artificial-intelligence"], "entities": ["arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/which-models-work-well-together-measuring-heterogeneity-for-llm-team-selection", "markdown": "https://wpnews.pro/news/which-models-work-well-together-measuring-heterogeneity-for-llm-team-selection.md", "text": "https://wpnews.pro/news/which-models-work-well-together-measuring-heterogeneity-for-llm-team-selection.txt", "jsonld": "https://wpnews.pro/news/which-models-work-well-together-measuring-heterogeneity-for-llm-team-selection.jsonld"}}