mlsanityData to 14 Oct 2026

From Knowing to Acting: Benchmarking Self-Awareness Capability of LLM Agents

arXiv 2606.20661 ↗First posted 9 Jun 2026cs.AI cs.CL cs.LGcited by 2

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

The integration of external tools has transitioned LLM agents from passive responders to autonomous systems. However, current benchmarks prioritize execution success, neglecting self-awareness capability, the ability to discern whether a problem requires necessary external resources or can be solved via internal parametric knowledge. To address this, we introduce KAPRO (Knowing-Acting Quadrant PRObe), a framework that evaluates cognitive-behavioral alignment by decoupling an agent's metacognitive judgment (Knowing) from its spontaneous execution (Acting). We further construct KAware, a dataset rigorously partitioning tasks into external, internal, and hybrid subspaces to systematically probe these epistemic boundaries. Extensive experiments across diverse agent architectures show that self-awareness capability is strongly correlated with task success but degrades sharply in internal-capability settings. Moreover, open-source and instruction-following models exhibit stronger tool overuse due to shallow pattern matching, while proprietary and reasoning-oriented models demonstrate more reliable cognitive gating. Benchmark and codes are available at https://github.com/AI-Santiago/KAware.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 28 of its 44 references that are in the map. First posted on arXiv 9 Jun 2026.

Builds on

28 of 44 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Synergistic Multi-Agent Framework with Trajectory Learning for Knowledge-Intensive Tasks13 Jul
  2. 2024τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains17 Jun

Other references · 26

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  3. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  4. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,141
  5. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,947
  6. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  7. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
  8. 2023Gorilla: Large Language Model Connected with Massive APIscited by 1,332
  9. 2023GAIA: a benchmark for General AI Assistantscited by 1,128
  10. 2026DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligencecited by 1,040
  11. 2017Selective Classification for Deep Neural Networkscited by 852
  12. 2025DeepSeek-V3.2: Pushing the Frontier of Open Large Language Modelscited by 815
14 more
  1. 2022Teaching Models to Express Their Uncertainty in Wordscited by 751
  2. 2024ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilitiescited by 295
  3. 2023MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Usecited by 217
  4. 2023TaskBench: Benchmarking Large Language Models for Task Automationcited by 144
  5. 2021Machine Learning with a Reject Option: A surveycited by 109
  6. 2023T-Eval: Evaluating the Tool Utilization Capability of Large Language Models Step by Stepcited by 93
  7. 2024Planning, Creation, Usage: Benchmarking LLMs for Comprehensive Tool Utilization in Real-World Complex Scenarioscited by 86
  8. 2025TOUCAN: Synthesizing 1.5M Tool-Agentic Data from Real-World MCP Environmentscited by 69
  9. 2025ToolHop: A Query-Driven Benchmark for Evaluating Large Language Models in Multi-Hop Tool Usecited by 50
  10. 2025MCP-RADAR: A Multi-Dimensional Benchmark for Evaluating Tool Use Capabilities in Large Language Modelscited by 48
  11. 2025MCPEval: Automatic MCP-based Deep Evaluation for AI Agent Modelscited by 36
  12. 2025RECODE-H: A Benchmark for Research Code Development with Interactive Human Feedbackcited by 25
  13. 2025Ready Jurist One: Benchmarking Language Agents for Legal Intelligence in Dynamic Environmentscited by 13
  14. 2026The Tool-Overuse Illusion: Why Does LLM Prefer External Tools over Internal Knowledge?cited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Judged Useless, Queried Anyway: Tool-Using Agents Rarely Turn Their Own Evidence Judgments into Stopping Decisionscited by 0
  2. 2026Agentic Self-Awareness: Aligning Knowing, Acting, and Holdingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Agentic Knowledgeable Self-awareness
  2. 2025What Do LLM Agents Do When Left Alone? Evidence of Spontaneous Meta-Cognitive Patterns
  3. 2026EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer
  4. 2026AgentAbstain: Do LLM Agents Know When Not to Act?
  5. 2026MetaCogAgent: A Metacognitive Multi-Agent LLM Framework with Self-Aware Task Delegation
  6. 2024Competence-Aware AI Agents with Metacognition for Unknown Situations and Environments (MUSE)
  7. 2026TRACE: A Self-Evolving Skill Bank for Consistent, Limit-Aware LLM Agents
  8. 2026Know When to Trust the Skill: Delayed Appraisal and Epistemic Vigilance for Single-Agent LLMs