Unverified50% confidenceFactExact time
The filtering strategy sets a threshold and directly removes low-quality demonstrations; the training set becomes smaller but cleaner, with the problems being threshold selection and the sharp drop in data volume after filtering.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOver-filtering leads to lower recall; when filter conditions are too narrow and the candidate set is empty, the system needs to degrade gracefully rather than return irrelevant snippets77% similarUnverifiedDo not exclude blocked trajectories from training, because filtering out blocked trajectories is still selecting trajectories that obtain high reward by evading the monitor, implicitly inversely selecting blocked trajectories, and due to the nature of softmax, increasing the probability of evasion necessarily comes at the cost of lowering the probability of non-evasion.77% similarUnverified评分校准是关键技术难点，模型必须能大量给出低分，否则过滤系统失去意义74% similarUnverifiedApproaches to mitigating the coverage problem include introducing diversity penalty terms at the decoding stage, upsampling training on Semantic IDs of long-tail items, and adding coverage constraints at the ranking stage, but these methods often involve trade-offs against immediate metrics such as click-through rate.74% similarUnverifiedThere is a significant difference in result quality between pre-filtering (filter then search) and post-filtering (search then filter); post-filtering may result in insufficient recall because a large number of candidates are eliminated after the fact.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1027479API
curl https://kongchang.com/api/v1/knowledge/claims/1027479MCP
get_claim(id=1027479)