arXiv (All)
open access
The Decomposition Tax: LLM Pipelines Lose Up to 40 Accuracy Points at Their Own Interfaces
A four-stage LLM pipeline gives up as much as 40.5 accuracy points at its own interfaces (gemma-3-12B on MATH-500, Holm-corrected p = 1.66e-19; the largest tax in the primary family). We hold model, problem, stages, stage prompts and completion budget fixed, vary only whether each stage can still se
Conceptio Open Archive
Analyze & explore related research in Conceptio
Connect this record directly into your research workflow or AI agent. Explore citations across open repositories, query full text, and export structured proof bundles.
This document is indexed with metadata only: full text is not available in the archive for this record. Open the official source ↗
Related documents
- The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines#610090
- The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines#609028
- FAPO: Fully Autonomous Prompt Optimization of Multi-Step LLM Pipelines#290656
- KV-Pipe: On the Relation Between KV Sharing and Pipeline Parallel Efficiency in LLMs#617059
- KV-Pipe: On the Relation Between KV Sharing and Pipeline Parallel Efficiency in LLMs#616559
Record · ID 1117156
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Bring this data into your agent: Connect via MCP or API · 20 free credits to start · no card required
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive