ZICQ curates daily AI news and deep dives on models, open-source infrastructure, and developer field reports.
Submit an articleCurrent filter: Tag: #Character Error Rate Clear filters
Agentic· 2026-11-06#Agent Evaluation#Closed-Loop Evaluation#Provenance#Self-Calibration#Production Monitoring
By Mr.Xu1 views0 discussions
This paper proposes closed-loop evaluation for deployed AI agents, treating evaluation as …
LLMs & Foundation Models· 2026-11-06
Mr.Xu1 views0 discussions
RAGCompass addresses the fact that no retriever is consistently best for every query in re…
LLMs & Foundation Models· 2026-11-06
Mr.Xu1 views0 discussions
The DS@GT ARC team developed a four-stage fact-checking article generation pipeline for CL…
Research & Papers· 2026-11-06
Mr.Xu1 views0 discussions
The paper develops a theoretical framework for comparing Darwinian, Lamarckian, and agent-…
Agentic· 2026-11-06
Mr.Xu0 views0 discussions
The paper introduces the Trust Responsive Authorization Policy (TRAP), which converts runt…
Research & Papers· 2026-11-06
Mr.Xu1 views0 discussions
The paper introduces a simulated-collaboration methodology that replays proof-development …
Agentic· 2026-11-06
Mr.Xu0 views0 discussions
Enterprise AI billing is typically aggregated at the SKU and billing-period level, making …
Compute & Architecture· 2026-11-06
Mr.Xu0 views0 discussions
FlexFloor addresses idle worker time in reinforcement-learning post-training for mixture-o…
LLMs & Foundation Models· 2026-11-06
Mr.Xu1 views0 discussions
BlockLock addresses batch-dependent inference by fixing the numerical path observed by a p…
Agentic· 2026-11-06
Mr.Xu1 views0 discussions
KDASO is an automated skill-optimization framework for liability adjudication. It combines…
LLMs & Foundation Models· 2026-11-06
Mr.Xu1 views0 discussions
EditVeto addresses overcorrection in LLM-based OCR repair, where a model fixes genuine OCR…
LLMs & Foundation Models· 2026-11-06
Mr.Xu1 views0 discussions
Large language models tend to answer ambiguous references prematurely rather than ask ques…
