Unverified50% confidenceFactExact time
The training of deep learning models is not a fully interpretable process, and researchers often discover capabilities that were not explicitly designed during post-training evaluations, such as in-context learning, chain-of-thought reasoning, tool-calling tendencies, and so on.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedModern deep learning models generally exhibit a tendency toward overconfidence, which stems both from the training objective (minimizing cross-entropy) itself not directly optimizing calibration and from excessive model capacity leading to memorization of the training distribution.78% similarUnverifiedThe lifecycle of deep learning is typically divided into two phases: Training and Inference.77% similarUnverifiedTransductive learning does not attempt to build a general model but directly infers from known training samples to specific test samples; inductive learning extracts generalizable rules from training samples.77% similarUnverifiedDeepMind's Chinchilla research pointed out that previous models were generally oversized and undertrained, meaning they had too many parameters but insufficient training data.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1031695API
curl https://kongchang.com/api/v1/knowledge/claims/1031695MCP
get_claim(id=1031695)