# VeriTrace: Evolving Mental Models for Deep Research Agents

arXiv: https://arxiv.org/abs/2605.26081 · HTML: https://mlsanity.com/paper/2605.26081/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 19 of its 47 references that are in the map.
- First posted on arXiv 25 May 2026.

## Placed in

- LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/)

## Builds on (most cited first, 19 of 19)

- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society](https://mlsanity.com/paper/2303.17760/) (2023)
- [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026)
- [ChatDev: Communicative Agents for Software Development](https://mlsanity.com/paper/2307.07924/) (2023)
- [WebThinker: Empowering Large Reasoning Models with Deep Research Capability](https://mlsanity.com/paper/2504.21776/) (2025)
- [DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents](https://mlsanity.com/paper/2506.11763/) (2025)
- [Towards End-to-End Automation of AI Research](https://mlsanity.com/paper/2606.15497/) (2026)
- [Deep Research Agents: A Systematic Examination And Roadmap](https://mlsanity.com/paper/2506.18096/) (2025)
- [Learning Latent Representations to Influence Multi-Agent Interaction](https://mlsanity.com/paper/2011.06619/) (2020)
- [WebWeaver: Structuring Web-Scale Evidence with Dynamic Outlines for Open-Ended Deep Research](https://mlsanity.com/paper/2509.13312/) (2025)
- [Prompt-based Adaptation in Large-scale Vision Models: A Survey](https://mlsanity.com/paper/2510.13219/) (2025)
- [Enterprise Deep Research: Steerable Multi-Agent Deep Research for Enterprise Analytics](https://mlsanity.com/paper/2510.17797/) (2025)
- [OmniThink: Expanding Knowledge Boundaries in Machine Writing through Thinking](https://mlsanity.com/paper/2501.09751/) (2025)
- [Not All Directions Matter: Towards Structured and Task-Aware Low-Rank Model Adaptation](https://mlsanity.com/paper/2603.14228/) (2026)
- [RhinoInsight: Improving Deep Research through Control Mechanisms for Model Behavior and Context](https://mlsanity.com/paper/2511.18743/) (2025)
- [EmoDebt: Bayesian-Optimized Emotional Intelligence for Strategic Agent-to-Agent Debt Recovery](https://mlsanity.com/paper/2503.21080/) (2025)
- [FS-Researcher: Test-Time Scaling for Long-Horizon Research Tasks with File-System-Based Agents](https://mlsanity.com/paper/2602.01566/) (2026)
- [EmoMAS: Emotion-Aware Multi-Agent System for High-Stakes Edge-Deployable Negotiation with Bayesian Orchestration](https://mlsanity.com/paper/2604.07003/) (2026)
- [Mind2Report: Expert-Level Commercial Report Synthesis via Cognitive Deep Research Agent](https://mlsanity.com/paper/2601.04879/) (2026)

## Abstract (arXiv)

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 errors propagate 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 an average of 4.82 pp on DeepResearch Bench (DRB) Insight (1.83 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.