2026-09-21 → 2026-09-27 · scanned 3855 unique arXiv papers · 3 candidate(s)
Large-language-model multi-agent systems (LLM-MAS) introduce a characteristic reliability problem: an error produced by one agent can be accepted as context by downstream agents and propagate across the workflow.
Authors: Uliana Elina
Affiliations: Unknown
Published: 2026-09-27
AI research agents must predict the effects of computational changes after budgeted experiments.
Authors: Jingjie Ning, Xueqi Li, Yibo Kong, Dongting Li
Affiliations: Unknown
Published: 2026-09-23
Financial question answering often requires precise numerical extraction, unit handling, and arithmetic over tables and text, but applying expensive verification uniformly wastes test-time compute.
Authors: Janet Jenq, Hongda Shen
Affiliations: Unknown
Published: 2026-09-27