When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models
- 类型：arxiv
- 标识：2606.27288
- 链接：https://arxiv.org/abs/2606.27288
- 主分类：agent
- 形态：method
- 被引：5
- 被引来源：Semantic Scholar
- S2被引：5
- OpenAlex被引：0
- 影响力被引：2
- TLDR：Multi-model LLM systems such as routing, voting, cascades, fusion, and mixture-of-agents are used to beat single-model accuracy, it is shown that their gain is capped by a quantity the field rarely reports, and combining models rarely beats the single best model without a strong query-level routing signal.
- OpenAlex ID：W7166155130
- OpenAlex DOI：10.48550/arxiv.2606.27288
- DOI：10.48550/arxiv.2606.27288
- DOI来源：OpenAlex
- 开放获取：green
- 开放获取链接：https://doi.org/10.48550/arxiv.2606.27288
- OpenAlex更新：2026-07-19
- 待LLM分类：否
- 标题中文：何时组合 LLM 更有帮助？——基于 67 个前沿模型的路由、投票与 Mixture-of-Agents 共失效上限研究
- TLDR中文：路由、投票、级联、融合与 Mixture-of-Agents 等多模型 LLM 系统常被用于超越单模型精度；研究表明其增益受限于一个该领域鲜少报告的量化指标，且在缺乏强查询级路由信号时，组合模型很少能胜过单一最佳模型。
- 来源文件：
- /inbox/tom/_candidates/2026-06-29-agent-memory-tool-use-candidates.json
- /inbox/tom/_candidates/2026-06-28-agent-rag-longcontext-candidates.json
- [S2 enrich]
- [OpenAlex backfill]