Abstract:Deep research agents face vast, interdependent, and pervasively uncertain information. Existing systems explore what evolving intermediate representations should look like, but leave their evolution to the LLM's implicit reasoning. Without explicit regulation, the intermediate layer is easily contaminated by mixed-quality information and propagates errors along its dependencies, so model scale often ends up substituting for absent regulation. We argue that an agent's mental model should instead evolve through explicit feedback that continuously aligns task understanding with reality, and identify three regulatory loops: interpretive update, deviation feedback, and schema revision. We realise this in VeriTrace, a cognitive-graph framework that explicitly implements the three loops. Using matched Qwen3.5-27B backbones, VeriTrace improves over the strongest matched baseline by 4.22 pp on DeepResearch Bench (DRB) Insight (1.49 pp Overall) and by 5.9 pp Overall win rate on DeepConsult. With Config-DeepSeek, it achieves the strongest reproducible open-source result on DRB.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.26081 [cs.AI] |
| (or arXiv:2605.26081v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2605.26081 arXiv-issued DOI via DataCite (pending registration) |
From: Haolang Zhao [view email]
[v1]
Mon, 25 May 2026 17:46:57 UTC (4,434 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。