Unverified50% confidenceFactExact time
The research team used Bayesian optimization to search over the TR and TL coefficients in order to find better intervention directions.
1
Sources
50%
Confidence
Long-term
Relevance
11/2/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedExperimental results showed that the intervention directions identified by TR/TL-based optimization outperform the original task vectors in performance.77% similarUnverifiedPUCT 在每次选择节点时，除了利用 UCT 的探索-利用权衡项，还引入了一个基于策略网络的先验预测项，用来指导搜索往更有希望的分支扩展。67% similarUnverifiedOptuna implements optimization algorithms such as TPE, CMA-ES, and Grid Search, and uses a pruning mechanism to early-terminate unpromising trials.67% similarUnverifiedCommon hyperparameter tuning strategies include grid search, random search, and Bayesian optimization; the latter models historical evaluation results to guide the next sampling and typically achieves better results with fewer trials.64% similarUnverifiedAt inference time, PUN employs test-time scaling: it uses the predicted global field to guide a bounded local search, thereby improving planning quality while maintaining computational efficiency.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1124971API
curl https://kongchang.com/api/v1/knowledge/claims/1124971MCP
get_claim(id=1124971)