# TRACE: Trajectory-Aware Comprehensive Evaluation for Deep Research Agents

arXiv: https://arxiv.org/abs/2602.21230 · HTML: https://mlsanity.com/paper/2602.21230/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 26 of its 48 references that are in the map.
- First posted on arXiv 5 Feb 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, 26 of 26)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [GAIA: a benchmark for General AI Assistants](https://mlsanity.com/paper/2311.12983/) (2023)
- [ReTool: Reinforcement Learning for Strategic Tool Use in LLMs](https://mlsanity.com/paper/2504.11536/) (2025)
- [ToolRL: Reward is All Tool Learning Needs](https://mlsanity.com/paper/2504.13958/) (2025)
- [WebThinker: Empowering Large Reasoning Models with Deep Research Capability](https://mlsanity.com/paper/2504.21776/) (2025)
- [WebSailor: Navigating Super-human Reasoning for Web Agent](https://mlsanity.com/paper/2507.02592/) (2025)
- [OpenThinkIMG: Learning to Think with Images via Visual Tool Reinforcement Learning](https://mlsanity.com/paper/2505.08617/) (2025)
- [WebWatcher: Breaking New Frontier of Vision-Language Deep Research Agent](https://mlsanity.com/paper/2508.05748/) (2025)
- [SafeArena: Evaluating the Safety of Autonomous Web Agents](https://mlsanity.com/paper/2503.04957/) (2025)
- [Explorer: Scaling Exploration-driven Web Trajectory Synthesis for Multimodal Web Agents](https://mlsanity.com/paper/2502.11357/) (2025)
- [ARTEMIS: Autoregressive End-to-End Trajectory Planning with Mixture of Experts for Autonomous Driving](https://mlsanity.com/paper/2504.19580/) (2025)
- [WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learning](https://mlsanity.com/paper/2509.13305/) (2025)
- [Scaling Agents via Continual Pre-training](https://mlsanity.com/paper/2509.13310/) (2025)
- [Recent Advances of Multimodal Continual Learning: A Comprehensive Survey](https://mlsanity.com/paper/2410.05352/) (2024)
- [GHPO: Adaptive Guidance for Stable and Efficient LLM Reinforcement Learning](https://mlsanity.com/paper/2507.10628/) (2025)
- [A Systematic Review of Key Retrieval-Augmented Generation (RAG) Systems: Progress, Gaps, and Future Directions](https://mlsanity.com/paper/2507.18910/) (2025)
- [Characterizing Deep Research: A Benchmark and Formal Definition](https://mlsanity.com/paper/2508.04183/) (2025)
- [Synergizing RAG and Reasoning: A Systematic Review](https://mlsanity.com/paper/2504.15909/) (2025)
- [Physics-Driven Data Generation for Contact-Rich Manipulation via Trajectory Optimization](https://mlsanity.com/paper/2502.20382/) (2025)
- [Benchmarking Large Language Models on Multiple Tasks in Bioinformatics NLP with Prompting](https://mlsanity.com/paper/2503.04013/) (2025)
- [MMLU-Reason: Benchmarking Multi-Task Multi-modal Language Understanding and Reasoning](https://mlsanity.com/paper/2505.16459/) (2025)
- [How Well Do LLMs Handle Cantonese? Benchmarking Cantonese Capabilities of Large Language Models](https://mlsanity.com/paper/2408.16756/) (2024)
- [ConSurv: Multimodal Continual Learning for Survival Analysis](https://mlsanity.com/paper/2511.09853/) (2025)

## Built on by (most cited first, 8 of 8)

- [Towards Security-Auditable LLM Agents: A Unified Graph Representation](https://mlsanity.com/paper/2605.06812/) (2026)
- [Models Recall What They Violate: Constraint Adherence in Multi-Turn LLM Ideation](https://mlsanity.com/paper/2604.28031/) (2026)
- [TraceSIR: A Multi-Agent Framework for Structured Analysis and Reporting of Agentic Execution Traces](https://mlsanity.com/paper/2603.00623/) (2026)
- [Diagnosing Search Behavior and Failure Modes in Long-Horizon Search Agents](https://mlsanity.com/paper/2608.01913/) (2026)
- [Holistic Evaluation and Failure Diagnosis of AI Agents](https://mlsanity.com/paper/2605.14865/) (2026)
- [GUI-CIDER: Mid-training GUI Agents via Causal Internalization and Density-aware Exemplar Reselection](https://mlsanity.com/paper/2605.28534/) (2026)
- [Memory Has Geometry: Non-Uniform Geometric Memory for Long-Horizon Personalized AI](https://mlsanity.com/paper/2609.17969/) (2026)
- [OS-SPEAR: A Toolkit for the Safety, Performance,Efficiency, and Robustness Analysis of OS Agents](https://mlsanity.com/paper/2604.24348/) (2026)

## Abstract (arXiv)

The evaluation of Deep Research Agents is a critical challenge, as conventional outcome-based metrics fail to capture the nuances of their complex reasoning. Current evaluation faces two primary challenges: 1) a reliance on singular metrics like Pass@1, creating a "high-score illusion" that ignores the quality, efficiency, and soundness of the reasoning process; and 2) the failure of static benchmarks to quantify crucial attributes like robustness and latent capability. To address these gaps, we introduce TRACE (Trajectory-Aware Comprehensive Evaluation), a framework that holistically assesses the entire problem-solving trajectory. To counter the "high-score illusion", we propose a Hierarchical Trajectory Utility Function that quantifies process efficiency and cognitive quality, including evidence grounding, alongside accuracy. To measure deeper attributes, TRACE introduces a Scaffolded Capability Assessment protocol, quantifying an agent's latent ability by determining the minimum guidance needed for success. Our contributions include the TRACE framework, its novel metrics, and the accompanying DeepResearch-Bench with controllable complexity. Experiments show TRACE delivers a granular ranking that uncovers critical trade-offs between agent accuracy, efficiency, and robustness entirely missed by singular metrics.