Unverified50% confidenceFactExact time
Models that more substantially reconstruct their own representations based on example content actually benefit more from few-shot prompting, which is contrary to the popular disruption hypothesis.
1
Sources
50%
Confidence
Long-term
Relevance
9/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThis research conclusion refutes the popular claim that models adaptively adjust their computation based on problem difficulty.77% similarUnverifiedResearchers have found that fine-tuning only a small number of model parameters is sufficient to remove safety alignment (de-alignment attacks), and that certain jailbreak prompts can transfer across different models.76% similarUnverified小模型对数据分布更敏感，同样的数据补充量在 4B 模型上的边际效果往往比大模型更明显，但也更容易过拟合或遗忘75% similarUnverifiedThe larger the context window, the more likely model performance is to degrade—the more information fed to the model, the worse it may perform.74% similarUnverified通过'不要按常规套路'这样的约束性提示词，本质上是在调整模型的采样分布，降低高频方案的输出权重，迫使模型探索低概率但可能更具创新性的输出区域73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/924781API
curl https://kongchang.com/api/v1/knowledge/claims/924781MCP
get_claim(id=924781)