method
active
method:multiplicative-rubric-reward-legibility-substantiveness-probe-scoreMultiplicative rubric reward (legibility × substantiveness × probe score)
Overall reward is product of judge-based and probe-based terms.
Neighborhood — ranked by edge-count
Papers (1)
paper
Frameworks (1)
framework
- Core pipeline turning probe-read features into scalable reward functions for open-ended tasks.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Prior approach to reduce underspecification.
- Score = (sum of completed quartet values) × (number of quartets), making portfolio composition consequential.
- Primary scoring method: scorer sees three reference responses at known quality levels alongside each target to eliminate inflation
- design rationale for game mechanics
- Interpretation of resampling.
- Rubric where LLM rates how well a feature's interpretation matches the activating text.
- Reward probes attend to entity and relation tokens; Failed Fix shows higher attention on punctuationfinding0.724Attention map observation.
- 14-point scoring rubric for human evaluation of feature interpretability covering confidence, activation consistency, logit consistency, and specificity