claim
active
claim:under-statistical-independence-the-reward-decomposes-into-additive-concept-level-terms-in-the-exponentUnder statistical independence, the reward decomposes into additive concept-level terms in the exponent
Eq. 3 assumption.
Source paper
extracted_from(2026) · Leon Bergen · Usha Bhalla · Sidharth Baskaran · Max Loeffler +13
Neighborhood — ranked by edge-count
Papers (1)
paper
Findings (2)
finding
- Most style interventions have strong off-target effects on non-targeted formatting attributesassociated_withcontradictsFig. 9.
- Fig. 8.
Claims (1)
claim
- Unifying Bayesian lens.
Questions (1)
question
- Case-study motivating question.
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.
- Claim about broader applicability of the scaling argument
- Expressivity, sample efficiency, and learning-dynamics accounts are complementary, not competingclaim0.766Limitations.
- Core interpretive claim of the paper connecting ETIs to connectionist learning
- Assumption D.3 formalized in the theoretical framework; empirically validated with coin-flip sequence experiments
- Justification for reward probes.
- Central framing.
- §4 Discussion.