Apodex Discovery: Reality Benchmarks and Environments for Evaluating and Building Discoverative Artificial Intelligence

• 类型：arxiv
• 标识：2608.11341
• 链接：https://arxiv.org/abs/2608.11341
• 主分类：evaluation
• 形态：benchmark
• 被引：0
• 被引来源：Semantic Scholar
• S2被引：0
• 影响力被引：0
• TLDR：This work introduces Apodex Discovery, a framework for building and evaluating discoverative AI through the heavy-duty solver, a system comprising a foundation model, harness, tools, and control policies that pursues extended, stateful, verifiable investigations.
• 待LLM分类：否
• 标题中文：Apodex Discovery：用于评估与构建探索型 AI 的现实基准与环境
• TLDR中文：提出 Apodex Discovery，一个通过 heavy-duty solver 构建和评估发现型 AI 的框架；该 solver 包含一个 foundation model、harness、工具和控制策略，用于执行长期的、有状态的、可验证的探索。
• 来源文件：
• /inbox/tom/_candidates/2026-08-18-rag-retrieval-reranking-candidates.json
• /inbox/tom/_candidates/2026-08-18-agent-rag-longcontext-candidates.json
• [S2 enrich]