llm-as-a-Verifier-dsh

仓库内包含的插件

• @aispin/plugin-verifierv0.3.2✓

  LLM-as-a-Verifier for dsh with the Best-of-N conversation mode built in: rank N candidates with a fine-grained verifier (expected grade over the logprob distribution), on demand via the verify tool or automatically on every turn of a Best-of-N session. Independent implementation of arXiv:2607.05391. By Aispin

  homepage

  https://github.com/aispin-dev/llm-as-a-Verifier-dsh#readme ↗

  package.json

  package.json

  dsh.bundle.patch

  cordis.patch.yml

  官方核心包 (9)

  • @deepseek-ai/cordis4.0.1共享
  • @deepseek-ai/dsh-client-runtime^0.1.0-rc.7共享
  • @deepseek-ai/dsh-client-ui-settings^0.1.0-rc.7共享
  • @deepseek-ai/dsh-client-web-react^0.1.0-rc.7共享
  • @deepseek-ai/dsh-credentials^0.1.0-rc.7共享
  • @deepseek-ai/dsh-llm^0.1.0-rc.6共享
  • @deepseek-ai/dsh-settings^0.1.0-rc.7共享
  • @deepseek-ai/dsh-tools^0.1.0-rc.6共享
  • @deepseek-ai/schemastery^3.18.1共享

路径固定于提交 f7fbb062dc00

概览

LLM-as-a-Verifier (arXiv:2607.05391) as a dsh plugin — Best-of-N conversation mode: give DeepSeek V4 Flash test-time scaling. Bo5 self-verification hits 88% on Terminal-Bench 2.1, beating some frontier models at a fraction of the cost. Fine-grained logprob-expectation scoring, PPT tournament, zero-config.