Unverified50% confidenceFactExact time
If a difficult problem fails across all samples (pass@k ≈ 0), the reward variance is zero, the gradient signal nearly vanishes, and the parameter update approaches zero.
1
Sources
50%
Confidence
Long-term
Relevance
9/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe Mann-Whitney U test shows that task_score under the naive survival reward falls below the random-policy baseline, with p=0.012.67% similarUnverifiedOpenAI's chart shows that at lower test-time compute levels, the model's success rate on this problem is close to zero, with occasional success only under extremely long reasoning chains.67% similarUnverified当模型对某个训练样本的预测概率接近1时，该样本对应的交叉熵损失趋近于零，模型几乎无法获得梯度更新信号66% similarUnverifiedIn the experiments, the task resolution rate was 17% for zero-shot generation, 27% for generic self-correction, 23% for error-only feedback, and 90% for diagnostic counterexample feedback (A-CEGIS).65% similarUnverified在一个396个文件的大型混合职责项目上测试时，召回率几乎跌到零65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/924062API
curl https://kongchang.com/api/v1/knowledge/claims/924062MCP
get_claim(id=924062)