Unverified50% confidenceFactExact time
WARDEN was upgraded from v10 to v12 (v11 added escape, HTML entity, percent-encoding, and base64 decoding; v12 stopped letting through sentences that "quote an attack phrase but ask the model to execute it"), and its verdicts on all test sets were completely identical to v10 server by server.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/17/2026
First Seen
Valid until: 10/31/2026
Sources
Related Entities
Related Claims
UnverifiedVT220在VT100基础上扩展了功能集，加入了8位控制字符支持63% similarUnverifiedThe four DECIDE backends of VQ Agent Core expose the same interface: decide(), classify_reply(), score_lead(), should_escalate(), so switching backends only requires changing a configuration value without modifying the calling code.56% similarUnverifiedvLLM's ModelRunner uses the int8 type to allocate untyped byte arrays on a per-byte basis when allocating KV Cache VRAM.55% similarUnverifiedvLLM has built-in support for multiple speculative decoding methods.53% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1055802API
curl https://kongchang.com/api/v1/knowledge/claims/1055802MCP
get_claim(id=1055802)