인터페이스는 현지화되어 있으며, Skill 소개와 설명 문서는 선택한 언어로 AI가 자동 번역합니다. 코드·명령어·경로·전문 용어는 원문을 유지하며 원작자 문서가 우선합니다.
Skills 둘러보기
생산성 및 협업 · SKILL.md

constraint_composition

constraint_composition

Compose multiple FUDGE future constraints, especially topic-word constraints, by summing weighted log probabilities.

Independent GitHub snapshot · Automated classification · Original documentation governs

Skill 문서

GitHub

name: constraint_composition description: Compose multiple FUDGE future constraints, especially topic-word constraints, by summing weighted log probabilities.

Constraint Composition for Topics

Use this skill when a FUDGE decoder must condition on several target words or independent attributes. For topic control, compute (lambda/N) * sum_j log P(w_j | prefix+candidate).

Inputs

  • Candidate-to-target probability table.
  • Ordered list of target words or attributes.
  • lambda_value and whether to normalize by N.

Outputs

  • Composed future log score per candidate.
  • Audit contributions per candidate and per target.

Workflow

  1. Ensure every candidate has every target probability.
  2. Clip probabilities to avoid log(0).
  3. Compute the target weight.
  4. Sum weighted log-probability contributions.
  5. Pass the composed score to the FUDGE rescoring skill.

Validation

Run python tests/test_composition.py.

저장한 Skills