Unverified50% confidenceBenchmarkExact time
While PTC-Bias reduces B-WER, U-WER (unbiased word error rate) remains largely unchanged.
1
Sources
50%
Confidence
Long-term
Relevance
9/25/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOverly aggressive biasing can often reduce B-WER but simultaneously raises U-WER, resulting in limited or even negative improvement in overall WER.69% similarUnverifiedOn real scientific claim tasks, PAPT consistently reduces repetition rates but accuracy declines.55% similarUnverifiedROUGE metrics exhibit length bias: models that generate shorter summaries have an advantage in precision, models that generate longer summaries have an advantage in recall, and ROUGE cannot measure factual consistency.54% similarUnverifiedRETD does not modify the follow-on trace or importance sampling ratios; it stores the emphatic TD signal in a leaky scalar state and releases the correction signal through a delay mechanism.54% similarUnverifiedHigh-recall verification systems face four mutually conflicting constraints: reading less, maintaining high recall, not losing minority evidence, and not converting unchecked into nothing found.54% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/949148API
curl https://kongchang.com/api/v1/knowledge/claims/949148MCP
get_claim(id=949148)