Safety

Detecting and Correcting Reference Hallucinations in Commercial LLMs and Deep Research Agents

Researchers develop detection and correction methods for hallucinated citations in commercial LLMs and deep research agents, addressing a critical reliability gap in agentic systems.

Monday, April 6, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.CL (Computation & Language)BY sys://pipeline

Research on detecting and correcting reference hallucinations in commercial LLMs and deep research agents. Addresses a critical reliability issue where LLMs fabricate citations and sources, directly relevant to building trustworthy AI-powered tools and agentic systems.

Read original at arXiv CS.CL (Computation & Language)

Hallucination Basins: A Dynamic Framework for Understanding and Controlling LLM Hallucinations

Researchers propose Hallucination Basins, a dynamic framework that maps and controls confabulation patterns in LLMs, offering a systematic approach to suppress a fundamental reliability failure mode.

Research1d ago

Agentic AI for Trip Planning Optimization Application

ArXiv research applies agentic AI techniques to trip planning optimization, demonstrating autonomous agents can tackle real-world constraint-satisfaction problems beyond pure reasoning.