Agentic· 2026-11-19
Mr.Xu0 views0 discussions
GUARD targets recursive self-improvement in agents that keep a frozen base model but conti…
GridForge combines an IEC 60287/60853-based thermal digital twin, calibrated synthetic tel…
An arXiv v1 study maps the failure modes of sub-4-bit quantization across ResNet and Effic…
RFBench aligns 10,000 two-port planar passive RF layouts with ADS Momentum simulations. Ea…
HardwareDx applies a graph foundation model to datacenter hardware diagnosis at Google, us…
An arXiv preprint reframes coding-agent economics as a form of software thermodynamics. Th…
Introspective Oversight reframes LLM self-reporting as a separately trained, third-person …
FutureBench evaluates how language models arbitrate between parametric knowledge and confl…
Researchers studying LLM-assisted research-question validation find a large gap between wh…
A study by Quinn Zhong and colleagues uses citation count as a proxy for factual prevalenc…
This paper uses iterated games with incomplete information to ask whether an LLM's visible…