今日速览
| 序号 | 标题 | 来源 | 日期 | 主题 | 推荐等级 |
|---|---|---|---|---|---|
| 1 | Digital Pantheon: Simulating and Auditing Coalition Formation with LLM Agents | arXiv | 2026-07-16 | RAG | 高 |
| 2 | BrainPilot: Automating Brain Discovery with Agentic Research | arXiv | 2026-07-16 | AI-Agent | 高 |
重点论文与技术动态
1. Digital Pantheon: Simulating and Auditing Coalition Formation with LLM Agents
- 来源:arXiv
- 日期:2026-07-16
- 作者/机构:Dylan Van Mulders, Matthias Bogaert, Dirk Van den Poel
- 主题标签:
RAG,arXiv - 推荐等级:高
- 分类:cs.CL, cs.AI, cs.MA
一句话结论
该框架构建了一个透明、可扩展的测试平台,用于事前探索政党兼容性和形式主义者调解的妥协。
核心内容
- 框架整合SFT、DPO和RAG技术,使LLM能维持政党特定行为,避免RLHF导致的中立性偏见。
- 引入Multi-Layered Information Lineage Topology (MILT)追踪谈判条款来源至政党宣言,分类为五种证明状态,并计算Coalition Influence Score (CIS)量化政党影响。
- 三个独立模拟稳定预测2019年弗兰德斯选举中N-VA领先于CD&V和Open Vld的胜者排名,且宣言锚定追溯可靠预测现实结果,幻觉内容不可靠。
方法与数据
- 方法:结合SFT、DPO和RAG;使用MILT和CIS进行可解释性追踪。
- 数据:基于2019年弗兰德斯选举的政党官方宣言。
价值判断
- 值得关注:成功解决LLM在政治模拟中的中立性偏见问题,实现政党行为的稳定模拟。
- 可复用点:MILT和CIS框架可推广至其他政治谈判场景,提供可解释性分析工具。
- 局限/待核查:摘要未明确具体局限,但需验证生成内容真实性以避免幻觉内容影响预测可靠性。
摘要
The formation of political coalitions is a complex negotiation driven by both concrete policy objectives and deep-seated ideological convictions. While Large Language Models (LLMs) open new avenues for computational political science, the neutrality and helpfulness biases instilled by Reinforcement Learning from Human Feedback (RLHF) prevent them from sustaining steadfast partisan behaviour. We present a multi-agent framework that reconciles factual grounding with ideological alignment by combining Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Retrieval-Augmented Generation (RAG): DPO instils aggressive party-specific personas, while a per-party RAG pipeline keeps each agent bounded to its official manifesto. We operationalize the framework on the 2019 Flemish election, deploying the partisan agents in a hub-and-spoke negotiation arbitrated by a formateur. To make the emergent negotiation interpretable, we introduce a Multi-Layered Information Lineage Topology (MILT) that traces every clause in the final agreement back to its manifesto origin and classifies it into five provenance states, a Coalition Influence Score (CIS) that aggregates these traceable contributions to identify which party shaped the agreement, and a real-world grounding pass that benchmarks each simulated provision against the historically adopted coalition agreement. Across three independent simulations the framework yields a stable winner and ranking (N-VA ahead of CD\&V and Open Vld), and manifesto-anchored lineage reliably predicts real-world materialization whereas hallucinated content does not. The result is a transparent, scalable testbed for the ex-ante exploration of party compatibility and formateur-mediated compromise.2. BrainPilot: Automating Brain Discovery with Agentic Research
- 来源:arXiv
- 日期:2026-07-16
- 作者/机构:Haoxuan Li, Tianci Gao, Jianhe Li, Yang Fan, Runze Shi, Weiran Wang
- 主题标签:
AI-Agent,arXiv - 推荐等级:高
- 分类:cs.AI
一句话结论
BrainPilot通过多代理系统实现脑科学发现的自动化,提供可追溯日志和代理验证结果,性能媲美先进框架且成本更低。
核心内容
- BrainPilot是完全开源的多代理系统,主研究员代理协调专家代理,基于7,233项索引知识库和72个方法单元的领域知识,覆盖七个研究领域。
- Graph of Trace记录所有工作流步骤,链接子目标、工具使用、证据和主张,支持审计和专家干预。
- 审计员代理集成防造假检查,解决当前代理易虚构和漂移问题,确保结果可信。
方法与数据
- 知识库含7,233个索引项和72个方法单元,评估使用三个任务(来自Agents’ Last Exam)、BrainPilotBench-v0基准及端到端案例研究。
- 摘要未明确数据收集方法。
价值判断
- 值得关注:解决脑科学中AI代理缺乏领域知识、易造假等关键缺陷,提升研究可靠性。
- 可复用点:开源架构、可审计工作流设计、防造假机制及成本效益可推广至其他科学自动化领域。
- 局限/待核查:摘要未明确。