Unverified50% confidenceBenchmarkExact time
"From Detection to Decision" evaluated Jev, two versions of Llama Prompt Guard 2, and an LLM judge on 47,611 held-out samples, concluding that achieving a 1% false negative rate requires escalating more than 91.8% of inputs to human handling.
1
Sources
50%
Confidence
Long-term
Relevance
10/18/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn a test of 520 verified comments, Jev 1.13 achieved 94.0% accuracy, detected 175/184 pieces of abusive content, falsely flagged 22/336 clean items, had a median latency of 250ms, and cost $0.05 per thousand calls.73% similarUnverifiedOn a test set of 600 synthetic invoices, the QLoRA fine-tuned model achieved a full-invoice exact rate of 83.0%, an auto-approval rate of 86.8%, and 0 out of 461 validation-coverage field errors in approved invoices (upper bound 0.8%).71% similarUnverifiedAfter adding a preprocessing function to decode base64/hex payloads in the command guard test, Jev's accuracy improved from 91% to 98.4%70% similarUnverifiedAfter disabling schema constraints and authentication, the erroneous backend dispatch rate spiked from 0% to 95.28% in replay/test evaluations.70% similarUnverifiedThe self-reflection loop (LLM review) architecture has an end-to-end success rate of 63.8% and a spec drift failure rate of 22.8%.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1060890API
curl https://kongchang.com/api/v1/knowledge/claims/1060890MCP
get_claim(id=1060890)