用多阶段大模型流水线生成多语言幽默文本，效果与顶尖模型相当。
RAGthoven at SemEval-2026 Task 1: A Multi-Stage Pipeline Walks Into a Benchmark and Barely Clears the Bar
- 分四步走：规划、写作、自评、判别，结合幽默理论设计
- 中文和西班牙语表现优于基线，英语持平，均与顶级模型并列
- 复杂智能体架构未带来提升，说明高阶设计在强模型前边际效益低
我们提出 RAGthoven 系统，用于 SemEval-2026 任务1（MWAHAHA）子任务A——在英语、西班牙语和中文中生成受约束的幽默文本。该系统采用多阶段大语言模型流水线（规划者、最佳N写作、反思者自评、LLM作为裁判），基于计算幽默理论（良性违反理论、脚本语义幽默理论），并通过十次实验优化。最终配置中，规划阶段引入从精选笑话语料库的检索增强生成（RAG），以多样化的幽默机制引导生成。同时评估两种代理变体——顺序工具调用（Exp09）和自主多分支编排（Exp10），均通过确定性约束审计器控制流程。在保留的12个英文样本上，四个前沿模型测试中，两种代理变体均未产生优于非代理流水线的输出，尽管工具调用开销显著更高。RAGthoven 在三种语言中与 Gemini 2.5 Flash 基线并列第一，组织方报告的置信区间重叠。在西班牙语中领先42原始埃洛分（1182对1140），而在英语（1045对1081）和中文（1045对1053）中，基线仍略高，但处于同一统计平局。结果表明，一旦使用强大前沿模型，复杂的多阶段提示工程和代理结构带来的收益随语言不同而递减。
原文摘要 · Abstract (English)
We present RAGthoven, our system for SemEval-2026 Task 1 (MWAHAHA), Subtask A (multilingual constrained humor generation in English, Spanish, and Chinese). RAGthoven decomposes creative text generation into a multi-stage large language model (LLM) pipeline (Planner, Best-of-N Writer, Reflector for self-critique, LLM-as-a-judge Judge) grounded in computational humor theory (Benign Violation Theory, Script-based Semantic Theory of Humor) and refined across ten experiments. In our final configuration, we augment the Planner with retrieval-augmented generation (RAG) from a curated joke corpus, seeding generation with diverse joke mechanisms. We also evaluate two agentic variants -- ReAct-style sequential tool-calling (Exp09) and autonomous multi-branch orchestration (Exp10) -- that expose the same four stages with a deterministic ConstraintAudit checker. Across four frontier models on a held-out 12-instance English sample, neither agentic variant produced outputs we judged superior to the non-agentic pipeline despite substantially higher tool-call budgets. RAGthoven shares Rank 1 with the Gemini 2.5 Flash baseline in all three languages, with overlapping organizer-reported confidence intervals. In Spanish, it leads the baseline by 42 raw Elo points (1182 vs. 1140), while in English (1045 vs. 1081) and Chinese (1045 vs. 1053) the baseline holds the higher raw rating within the same statistical tie. Together, these results suggest language-dependent diminishing returns from elaborate multi-stage prompt engineering and agentic scaffolding once a strong frontier model is in the loop.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。