claim

active

claim:fine-tuning-reduces-mismatch-dr-retrieval-increases-effective-cohesion-d-and-few-shot-adjusts-the-budget-k

Fine-tuning reduces mismatch dr, retrieval increases effective cohesion ρd, and few-shot adjusts the budget k

Unified interpretation of different adaptation methods via UCCT terms

Source paper

extracted_from

The Guanyin Protocol: A Framework for Immediately Establishing an Understanding of Both Causality and Compassion in LLM Systems Using Semantic Anchoring

(2025) · Edward Yi Chang · Kaya, Zeyneb N. · Ethan Chang

Neighborhood — ranked by edge-count

Findings (2)

finding

Cross-base fine-tuning yields asymmetric transfer: B10 transfers most robustly, B9 least
supports
In-base gains accompanied by uneven OOD drops; higher-density priors more robust.
Cross-base transfer: B10 transfers most robustly
supports
B10 fine-tuning yields smallest OOD drops when transferring to other bases

Communities (2)

community

Few-shot anchoring & latent structure
members_of
How minimal examples disambiguate and recruit latent arithmetic/reasoning interpretations in LLMs
Model base robustness and transfer learning asymmetries
members_of
Investigates how neural network bases (B10 vs B9) differ in fine-tuning stability and cross-base transfer, with density as a key predictor of robustness.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

Fine-tuning reduces dr; retrieval increases effective ρd; few-shot k trades budget against bothhypothesis0.912
UCCT's unified view of adaptation methods
Hypothesis: Fine-tuning reduces mismatch dr between prior and targethypothesis0.870
UCCT's theoretical prediction about how fine-tuning maps onto the anchoring score
Fine-tuning an emergently misaligned model on 120 secure code samples (35 steps, batch size 4) fully restores alignmentfinding0.819
Demonstrates emergent re-alignment is achievable with minimal data from same domain
SOO fine-tuning could be extended to align AI representations of its own goals with human user preferences, reducing misalignment by fostering coherence between self-related and other-related preferenceshypothesis0.816
Future work hypothesis about extending SOO to direct value alignment
Fine-tuning induces the behavioral pattern of self-correction but does not improve the underlying ability to correct effectivelyclaim0.804
Key interpretive conclusion from the dissociation between attempt rate and improvement rate in fine-tuning experiments
SOO fine-tuning could complement RLHF and Constitutional AI by fostering internal coherence that promotes honest behaviorsclaim0.798
Integration claim positioning SOO as additive to existing alignment approaches
Fine-tuning is universal in increasing the strength of helpful and harmless intentions across model families.claim0.797
Conclusion from Experiment 3 and HH intent analysis.
Models fine-tuned on human data show strong positive correlation between misalignment and incoherence; misalignment scores fall below 1% when incoherence threshold is appliedfinding0.790
Human data fine-tuning effect is distinct from synthetic emergent misalignment and likely caused by off-policy training