Data to 7 Oct 2026

SciMaster: Towards General-Purpose Scientific AI Agents, Part I. X-Master as Foundation: Can We Lead on Humanity's Last Exam?

arXiv 2507.05241 ↗First posted 7 Jul 2025cs.AI cs.CLcited by 45

Not on the map. Why?

Cited by 29 in LLM agents and tool use, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

The rapid advancements of AI agents have ignited the long-held ambition of leveraging them to accelerate scientific discovery. Achieving this goal requires a deep understanding of the frontiers of human knowledge. As such, Humanity's Last Exam (HLE) provides an exceptionally challenging touchstone for evaluating scientific AI agents. In this work, we aim to construct the foundational architecture for general-purpose agents and validate the capabilities through leading performance on HLE. To achieve this, we introduce X-Master, a tool-augmented reasoning agent designed to emulate human researchers by interacting flexibly with external tools during its reasoning process. This agent, guided by the conceptualization of code as an interaction language, can flexibly leverage built-in Python libraries and our customized tools to augment the reasoning. We further scale its capabilities through X-Masters, a scattered-and-stacked agentic workflow that systematically enhances breadth and depth of reasoning. Our open-source solution, X-Masters, sets a new state-of-the-art record on HLE with a score of 32.1%, surpassing OpenAI's and Google's Deep Research (26.6% and 26.9%) and becoming the first to exceed the 30% threshold. This work allows us to gain a deeper understanding of complex task-solving and accumulates valuable experience that can inform future advancements, guiding subsequent model training.

Key facts. Cited by 45 papers in the map (data through 7 Oct 2026). Builds on 19 of its 29 references that are in the map. First posted on arXiv 7 Jul 2025.

Builds on

19 of 29 references.

References · 19

2022Training language models to follow instructions with human feedbackcited by 15,712
2024The Llama 3 Herd of Modelscited by 14,192
2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
2024Qwen2.5 Technical Reportcited by 4,713
2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Frameworkcited by 1,761
2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,371
2023ChatDev: Communicative Agents for Software Developmentcited by 893
2025Humanity's Last Examcited by 467
2025Search-o1: Agentic Search-Enhanced Large Reasoning Modelscited by 423
2025ReTool: Reinforcement Learning for Strategic Tool Use in LLMscited by 377
2025Accelerating scientific discovery with Co-Scientistcited by 353
2025Absolute Zero: Reinforced Self-play Reasoning with Zero Datacited by 323

7 more

2025R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learningcited by 322
2025WebThinker: Empowering Large Reasoning Models with Deep Research Capabilitycited by 320
2025Learning to Reason without External Rewardscited by 199
2025WebDancer: Towards Autonomous Information Seeking Agencycited by 180
2025ToRL: Scaling Tool-Integrated RLcited by 166
2025Maximizing Confidence Alone Improves Reasoningcited by 97
2025MAS-GPT: Training LLMs to Build LLM-based Multi-Agent Systemscited by 45

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 23; 2026: 22.

45 papers.

Sort:

By area:

2025AgentFold: Long-Horizon Web Agents with Proactive Context Managementcited by 95
2025Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solvingcited by 90
2025From AI for Science to Agentic Science: A Survey on Autonomous Scientific Discoverycited by 74
2025WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learningcited by 59
2025Tongyi DeepResearch Technical Reportcited by 53
2025CASCADE: Cumulative Agentic Skill Creation through Autonomous Development and Evolutioncited by 33
2025LLM-based Agentic Reasoning Frameworks: A Survey from Methods to Scenarioscited by 25
2025BrowseMaster: Towards Scalable Web Browsing via Tool-Augmented Programmatic Agent Paircited by 16
2025TUMIX: Multi-Agent Test-Time Scaling with Tool-Use Mixturecited by 13
2026SciAgentGym: Benchmarking Multi-Step Scientific Tool-use in LLM Agentscited by 12
2025Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoningcited by 11
2025Scaling Generalist Data-Analytic Agentscited by 11

33 more

2025FlowSearch: Advancing deep research with dynamic structured knowledge flowcited by 10
2026Can We Predict Before Executing Machine Learning Agents?cited by 10
2026Toward Generalist Autonomous Research via Hypothesis-Tree Refinementcited by 10
2026AgentSwing: Adaptive Parallel Context Management Routing for Long-Horizon Web Agentscited by 9
2025RhinoInsight: Improving Deep Research through Control Mechanisms for Model Behavior and Contextcited by 8
2025Bohrium + SciMaster: Building the Infrastructure and Ecosystem for Agentic Science at Scalecited by 8
2025Multimodal Reasoning for Science: Technical Report and 1st Place Solution to the ICML 2025 SeePhys Challengecited by 6
2026InnoEval: On Research Idea Evaluation as a Knowledge-Grounded, Multi-Perspective Reasoning Problemcited by 6
2025AlphaApollo: A System for Deep Agentic Reasoningcited by 5
2025CoT-Evo: Evolutionary Distillation of Chain-of-Thought for Scientific Reasoningcited by 5
2025MarsRL: Advancing Multi-Agent Reasoning System via Reinforcement Learning with Agentic Pipeline Parallelismcited by 4
2025Unifying Dynamic Tool Creation and Cross-Task Experience Sharing through Cognitive Memory Architecturecited by 4
2026ASI-Evolve: AI Accelerates AIcited by 4
2026EvoMaster: A Foundational Evolving Agent Framework for Agentic Science at Scalecited by 4
2026Rewarding the Scientific Process: Process-Level Reward Modeling for Agentic Data Analysiscited by 4
2026AutoSci: A Memory-Centric Agentic System for the Full Scientific Research Lifecyclecited by 4
2025InfoMosaic-Bench: Evaluating Multi-Source Information Seeking in Tool-Augmented Agentscited by 3
2025QMBench: A Research Level Benchmark for Quantum Materials Researchcited by 3
2026Research on World Models Is Not Merely Injecting World Knowledge into Specific Taskscited by 3
2026PRISM: Pushing the Frontier of Deep Think via Process Reward Model-Guided Inferencecited by 3
2026Discovery and Reinforcement of Tool-Integrated Reasoning Chains via Rollout Treescited by 2
2026HarmonyCell: Automating Single-Cell Perturbation Modeling under Semantic and Distribution Shiftscited by 2
2025DualResearch: Entropy-Gated Dual-Graph Retrieval for Answer Reconstructioncited by 1
2026Towards LLM-enabled autonomous combustion research: A literature-aware agent for self-corrective modeling workflowscited by 1
2026SEA-TS: Self-Evolving Agent for Autonomous Code Generation of Time Series Forecasting Algorithmscited by 1
2026SciNav: A General Agent Framework for Scientific Coding Taskscited by 1
2026The Agentification of Scientific Research: A Physicist's Perspectivecited by 1
2026SciResearcher: Scaling Deep Research Agents for Frontier Scientific Reasoningcited by 1
2026ReasFlow: Assisting Reasoning-Centric Scientific Discovery in Applied Mathematics via a Knowledge-Based Multi-Agent Systemcited by 1
2026What is Missing from AI Post-Training AI: An Empirical Analysiscited by 1
2025Constraint-Driven Small Language Models Based on Agent and OpenAlex Knowledge Graph: Mining Conceptual Pathways and Discovering Innovation Points in Academic Paperscited by 0
2026ReCrit: Transition-Aware Reinforcement Learning for Scientific Critic Reasoningcited by 0
2026Towards Direct Latent-Space Synthesis for Parallel Branches in LLM-Agent Workflowscited by 0

Similar, not cited

Close in wording, no citation either way