CASE FILE / Decision automation
Introducing the Papers with Code MCP server
Introducing the Papers with Code MCP server! To celebrate its release, I've asked Claude Code to research the architecture of Jev by @typesafeai using the `search_papers` tool. Here's what it came up with: "If Jev is architecturally anything public, it's a large schema-conditioned bidirectional encoder with parallel label-query heads (GLiNER/ML-Decoder shape) trained with a proper-scoring-rule RL objective (RLCR shape), scaled far past the ~150M-param range those papers operate in — "neither small nor an LLM" fits that. The 40–200× speedup is consistent with removing autoregression, not with any exotic mechanism."
Original post
Introducing the Papers with Code MCP server! To celebrate its release, I've asked Claude Code to research the architecture of Jev by @typesafeai using the `search_papers` tool. Here's what it came up with: "If Jev is architecturally anything public, it's a large schema-conditioned bidirectional encoder with parallel label-query heads (GLiNER/ML-Decoder shape) trained with a proper-scoring-rule RL objective (RLCR shape), scaled far past the ~150M-param range those papers operate in — "neither small nor an LLM" fits that. The 40–200× speedup is consistent with removing autoregression, not with any exotic mechanism."
Chinese translation
介绍带有Code的论文LCP服务器！ 为了庆祝它的发布，我请Claude Code通过@typesafeai使用“search_papers”工具研究Jev的架构。 这是它的结果： “如果Jev在架构上是任何公共的东西，那么它就是一个大型模式条件双向编码器，具有并行标签查询头（GLiNER/ML-Decoder形状），用适当的评分规则RL目标（RLCC形状）进行训练，扩展范围远远超过了这些论文操作的~ 150 M参数范围--“既不小，也不LLM”符合这一点。40-200倍的加速与消除自回归是一致的，而不是任何奇异机制。"
What this case shows
- Use JEV as a typed, fast decision layer
- Wire probability into business rules
- Open the original post for the full demo