Unverified50% confidenceFactExact time
For GPT-5.4, more frequent correct simulation masked an evidence-use loss of about ten points, making the net change indistinguishable from zero.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/20/2026
First Seen
Valid until: 1/18/2027
Sources
Related Entities
Related Claims
Unverified一项针对GPT-4的研究显示，对MMLU题目进行语义等价改写后，模型准确率在部分子集上下降超过10个百分点74% similarUnverifiedIn the controlled GPT-2 experiments, although projection updates were clearly higher in lexical hits, the strict answer score of all experimental groups remained zero.74% similarUnverifiedIn an editing benchmark, the same GPT model's score dropped from 99% to 70% depending on the framework used.72% similarUnverifiedFine-tuned DistilBERT achieves F1=0.9993 on in-distribution testing, but loses 12.9 mean average precision points under topic shift, a greater degradation than the linear model's 5.2 points.71% similarUnverifiedExperiments conducted by the research team on models such as GPT-2 showed that after only 5–9 generations of recursive training, the quality of model outputs degrades significantly.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073398API
curl https://kongchang.com/api/v1/knowledge/claims/1073398MCP
get_claim(id=1073398)