Unverified50% confidenceBenchmarkExact time
In an ablation experiment, after replacing the model's latent thinking segments with segments from unrelated questions, accuracy dropped by 45%–51%, indicating that latent thinking carries substantive reasoning.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedFor the two models tested, enabling maximum reasoning capability reduced the capitulation rate on deep puzzles from 19.2% and 12.5% directly to 0%.78% similarUnverifiedTop academic deepfake detectors can see their accuracy plummet from 99% to below 60% when faced with previously unseen generative models, a phenomenon known as the cross-model generalization problem.77% similarUnverifiedWhen the context exceeds 40% of the model's context window, the model's reasoning ability noticeably degrades, exhibiting a decline in intelligence.77% similarUnverifiedAnthropic's research shows that even mild alignment training can cause a 2–8 percentage point decline in model performance on certain reasoning-intensive tasks.76% similarUnverifiedThe study found that models' accuracy on semantically plausible distorted statements is actually higher than their performance on meaningless random replacements.76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1072475API
curl https://kongchang.com/api/v1/knowledge/claims/1072475MCP
get_claim(id=1072475)