Unverified50% confidenceFactExact time
This study derives closed-form optimal policies for the pass@k and max@k objectives under KL regularization, showing that @k optimization is essentially redistributing learning pressure according to problem difficulty, shifting the optimization focus from easy problems to hard ones.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
Unverified@k pressure can be fully realized through single-completion training, matching the performance of multi-completion @k optimizers at only a fraction of their cost.75% similarUnverifiedIn behavior-constrained reinforcement learning, KL regularization can be achieved with access only to the gradient of the reference policy, and the gradient of the reference policy can be obtained through a velocity score transformation.73% similarUnverifiedIn the MoE architecture, a routing mechanism dynamically selects the most relevant small subset of experts for each input, typically employing a Top-K strategy (usually K=2), and a load-balancing loss function is introduced during training to prevent uneven load distribution.70% similarUnverifiedReverse KL favors refining the token ranking the student is already active on (mode-seeking), while forward KL is responsible for exploring candidates that the teacher's support covers but the student does not sufficiently attend to (mass-covering).70% similarUnverifiedKimi K2.6能根据问题复杂度自适应调整输出长度,这一能力通过Post Training阶段的长度校准技术实现69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1130869API
curl https://kongchang.com/api/v1/knowledge/claims/1130869MCP
get_claim(id=1130869)