Unverified50% confidenceBenchmarkExact time
When the input pipeline was changed to select each day one stock with news-driven activity that day, the same local Gemma model's accuracy exceeded its paired baseline by 7.6 percentage points (p≈0.02).
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedPiPMRE achieves an average improvement of 5.6 percentage points in recall over the previous state-of-the-art methods on two public datasets.62% similarUnverifiedIn enterprise knowledge base scenarios, using semantic chunking strategies typically yields recall accuracy more than 15% higher than fixed-length chunking, and introducing a Reranker re-ranking model can further improve accuracy by 10–30%61% similarUnverified在歧义点的四类行为中，'先搜再问'成功率最高，平均通过率达93.4%，远高于'直接猜'的56.5%，'搜多次再猜'为51.5%61% similarUnverified针对金融领域微调的嵌入模型（如FinBERT）相比通用嵌入模型在财报问答任务上准确率高出15%以上61% similarVerifiedReAct's errors amplify in a chain: if the accuracy per step is p, the overall success rate after n steps is approximately p to the power of n; with a per-step accuracy of 95%, the overall success rate drops to approximately 59.9% after 10 steps and only 35.8% after 20 steps.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/903350API
curl https://kongchang.com/api/v1/knowledge/claims/903350MCP
get_claim(id=903350)