Unverified50% confidenceBenchmarkExact time
At N=5000, compared with full-label training, CAPS is 1.5 percentage points lower in accuracy on the 0.6B model and 0.6 percentage points lower on the 1.7B model (versus one full-label seed), and propensity weighting brought no gain on this log.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/13/2026
First Seen
Valid until: 2/11/2027
Sources
Related Entities
Related Claims
UnverifiedWhen sampling eight candidates without constraints from the model, 10.2% of inputs cannot produce any valid output under limited training data, and this proportion drops to 0.4% when using the full training data73% similarUnverifiedThe quality ceiling of weak supervision labels depends on the accuracy of the zero-shot model itself; if the model makes systematic errors, the training set will inherit that bias.72% similarUnverifiedCompared with the best-performing simple generator, the model leads on 5 of 6 items on the main test set, with advantages of 0.7 to 2.5 percentage points, and the test can prove the gap for 1 of those items.72% similarUnverifiedOn repeated questions, index reinforcement raised the top-1 hit rate from 0.571 to 0.743 without any model training throughout the process.72% similarUnverifiedThe expert models trained by FuRA-S are only 1-5 MB in size, with the proportion of updated parameters only 0.015-0.05%.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1165002API
curl https://kongchang.com/api/v1/knowledge/claims/1165002MCP
get_claim(id=1165002)