Read original ↗
paperarXivTrust 82 · PrimaryPublished 1mo agoLive · 1mo ago

When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models

Multi-model LLM systems such as routing, voting, cascades, fusion, and mixture-of-agents are used to beat single-model accuracy. We show that their gain is capped by a quantity the field rarely reports. For any policy whose output is one member model answer, accuracy cannot exceed one minus beta, where beta is the rate at which every model is wrong on the same query. In contrast, the usual diagnostic, average pairwise error correlation rho, cannot identify beta: error laws with identical marginals and pairwise correlations can have different all-wrong rates. A Clopper-Pearson bound on beta giv

Lineage graph

Paper → model → repo connections mined from source citations (Tier-1 exact match).

Why these links exist

Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.

  • FuzzySimilar title/name (fuzzy) · 87%NirDiamant/GenAI_Agents

    Fuzzy title match (0.94): “When Does Combining Language Models Help? A Co-Failure Ceili” ≈ “NirDiamant/GenAI_Agents”

  • FuzzySimilar title/name (fuzzy) · 84%Unity-Technologies/ml-agents

    Fuzzy title match (0.92): “When Does Combining Language Models Help? A Co-Failure Ceili” ≈ “Unity-Technologies/ml-agents”

  • FuzzySimilar title/name (fuzzy) · 59%datawhalechina/hello-agents

    Fuzzy title match (0.73): “When Does Combining Language Models Help? A Co-Failure Ceili” ≈ “datawhalechina/hello-agents”

  • FuzzySimilar title/name (fuzzy) · 59%TauricResearch/TradingAgents

    Fuzzy title match (0.73): “When Does Combining Language Models Help? A Co-Failure Ceili” ≈ “TauricResearch/TradingAgents”

  • FuzzySimilar title/name (fuzzy) · 59%jnMetaCode/agency-agents-zh

    Fuzzy title match (0.73): “When Does Combining Language Models Help? A Co-Failure Ceili” ≈ “jnMetaCode/agency-agents-zh”

  • FuzzySimilar title/name (fuzzy) · 59%Eigenwise/atomic-agents

    Fuzzy title match (0.73): “When Does Combining Language Models Help? A Co-Failure Ceili” ≈ “Eigenwise/atomic-agents”

Implements (incoming)

Related across the graph

Topics