Researchers unveiled RealtimeBench, a new suite that tests large language […]
Category: Research
LHTB 2.0: 77‑Task Long‑Horizon Benchmark
Long‑Horizon‑Terminal‑Bench 2.0 (LHTB 2.0) expands the original 46‑task benchmark to […]
ClawLoop: Self‑Improving Agents
ClawLoop is a new framework that lets large language model […]
LLM-Referenced C-Test Gauges Supra-Intelligence
Researchers have unveiled a pioneering benchmark that leverages Levin’s Kt […]
RL Value Functions Encode World Models
Researchers have shown that a Proximal Policy Optimization (PPO) agent’s […]
VEPA: Adaptive Formal Guardrails for LLM Agents
VEPA, a new framework presented in arXiv:2610.12813v1, offers a verified, […]
LLM-Driven Hardware Knowledge Graphs
A new LLM‑driven pipeline turns dense hardware datasheets into hierarchical […]
EvoBelief: Self‑Evolving World Models
EvoBelief, a new framework from arXiv, lets large‑language‑model agents update […]
MagicBeanStalk: New RLVR Data Synthesis
MagicBeanStalk introduces a novel top‑down, bottom‑up data‑synthesis framework that reshapes […]
CLEAT: Value‑Guided Clarification Boosts LLM Ag…
Researchers unveiled CLEAT, a co‑training framework that teaches large‑language‑model agents […]