Unverified50% confidenceOpinionExact time
The key to determining whether a new relation can generalize is not whether the fact can be recalled, but whether its shared structure is easy to read out linearly in the pretrained representation.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIf you want a model to learn a new relation through fine-tuning, you should first check whether the corresponding old structure is already linearly encoded in its pretrained representation; for relations with poor linear readability, simply adding fine-tuning samples may yield limited benefit.72% similarUnverifiedThe revised knowledge takes effect not only in sentence patterns seen during training but also in expressions never encountered before.70% similarUnverifiedThe author believes that the 'Chain-of-Thought' output of LLMs is more likely a highly fitted reproduction of reasoning patterns in training data, rather than genuine logical deduction in the true sense.69% similarUnverifiedRedesigning the representation aggregation mechanism so that relational information is not destroyed during aggregation is a worthwhile future direction to explore67% similarUnverifiedThe paper proposes that representations obtained through contrastive learning inherently contain a family of geometries rather than a single geometry, and that the same set of fixed vector representations can be specialized to different semantic structures without retraining the model.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1041963API
curl https://kongchang.com/api/v1/knowledge/claims/1041963MCP
get_claim(id=1041963)