finding
active
finding:adding-a-single-disambiguating-example-12-9-21-aligns-divergent-m1-m4-interpretations-under-tested-seedsAdding a single disambiguating example (12−9=21) aligns divergent M1-M4 interpretations under tested seeds
E1 finding consistent with threshold-crossing: near-threshold state resolved by one additional anchor
Source paper
extracted_from(2025) · Edward Yi Chang · Kaya, Zeyneb N. · Ethan Chang
Neighborhood — ranked by edge-count
Claims (2)
claim
- Conclusion from E1 and central UCCT claim.
- Small prompt changes can yield threshold-like shifts because S crosses the critical value ScsupportsAuthors' explanation for abrupt behavioral changes
Hypotheses (1)
hypothesis
- Core testable hypothesis of UCCT about the nature of performance transitions under anchoring
Communities (3)
community
- Few-shot anchoring & latent structuremembers_ofHow minimal examples disambiguate and recruit latent arithmetic/reasoning interpretations in LLMs
- How minimal, task-specific prompt examples rebind model priors across threshold boundaries without weight updates, studied through arithmetic reasoning tasks.
- Disambiguation via single examplesmembers_ofOne counterintuitive arithmetic example aligns divergent model interpretations across random seeds
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.
- Ambiguous anchors (33-27=60, 11-9=20) yield four distinct arithmetic interpretations across M1-M4finding0.776Models produce different answers (240, 138, -240) from the same ambiguous prompt
- Core result of Experiment 3: cross-model semantic convergence under self-referential processing
- Generalization to arithmetic.
- Concurrent work result showing emergent misalignment occurs in small models
- Demonstrates emergent re-alignment is achievable with minimal data from same domain
- Ablation result showing neutrals are not strong indicators of diversity
- On DINOv3, classification and segmentation tolerate R²≈0.8, while depth needs R²≈0.9 to retain 95% accuracy.finding0.754Task-dependent distortion floors.
- Contrasts with temporal permutation results; constitutes the most suggestive evidence of potential consciousness phenomena in LLM representations.