mlsanityData to 14 Oct 2026

Do LLM Agents Know How to Ground, Recover, and Assess? A Benchmark for Epistemic Competence in Information-Seeking Agents

arXiv 2509.22391 ↗First posted 26 Sep 2025cs.AIcited by 7

On the map: LLM agents and tool use → Web, search, and research agents

Abstract (arXiv)

Recent work has explored training Large Language Model (LLM) search agents with reinforcement learning (RL) for open-domain question answering (QA). However, most evaluations focus solely on final answer accuracy, overlooking how these agents reason with and act on external evidence. We introduce SeekBench, the first benchmark for evaluating the textitepistemic competence of LLM search agents through step-level analysis of their response traces. SeekBench comprises 190 expert-annotated traces with over 1,800 response steps generated by LLM search agents, each enriched with evidence annotations for granular analysis of whether agents (1) generate reasoning steps grounded in observed evidence, (2) adaptively reformulate searches to recover from low-quality results, and (3) have proper calibration to correctly assess whether the current evidence is sufficient for providing an answer.

Key facts. Cited by 7 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 27 of its 39 references that are in the map. First posted on arXiv 26 Sep 2025.

Builds on

27 of 39 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025Beyond Ten Turns: Unlocking Long-Horizon Agentic Search with Large-Scale Asynchronous RL11 Aug
  2. 2025A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challenges3 Aug
  3. 2025WebSailor: Navigating Super-human Reasoning for Web Agent3 Jul
  4. 2025InfoDeepSeek: Benchmarking Agentic Information Seeking for Retrieval-Augmented Generation21 May

Other references · 23

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  4. 2024Qwen2.5 Technical Reportcited by 4,942
  5. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  6. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,955
  7. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,444
  8. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,282
  9. 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,128
  10. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,075
  11. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 916
  12. 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 698
11 more
  1. 2025R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learningcited by 336
  2. 2025DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environmentscited by 286
  3. 2025The Landscape of Agentic Reinforcement Learning for LLMs: A Surveycited by 205
  4. 2024Making Reasoning Matter: Measuring and Improving Faithfulness of Chain-of-Thought Reasoningcited by 110
  5. 2025Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Searchcited by 53
  6. 2024Direct Evaluation of Chain-of-Thought in Multi-hop Reasoning with Knowledge Graphscited by 32
  7. 2023ThoughtSource: A central hub for large language model reasoning datacited by 23
  8. 2025Premise-Augmented Reasoning Chains Improve Error Identification in Math reasoning with LLMscited by 22
  9. 2025Mapping the Minds of LLMs: A Graph-Based Analysis of Reasoning LLMcited by 21
  10. 2025Agent models: Internalizing Chain-of-Action Generation into Reasoning modelscited by 17
  11. 2025KG-TRACES: Enhancing Large Language Models with Knowledge Graph-constrained Trajectory Reasoning and Attribution Supervisioncited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 7.

7 papers.

  1. 2026A Survey of Agent Memory in the Second Half: Towards Self-Evolving and Long-Horizon Agentscited by 44
  2. 2026Time to REFLECT: Can We Trust LLM Judges for Evidence-based Research Agents?cited by 10
  3. 2026Do Web Agents Investigate Before They Decide?cited by 0
  4. 2026Co-ReAct: Rubrics as Step-Level Collaborators for ReAct Agentscited by 0
  5. 2026When English Rewrites Local Knowledge: Global Narrative Dominance in Large Language Modelscited by 0
  6. 2026CIVI: A Framework for Diagnosing Search Agent Failures in Civic Informationcited by 0
  7. 2026DoGBench: Can Agents Meet Expert Standards for User-Facing Documentation?cited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026SearchAtlas: Analyzing Agentic Search Strategies via Evidential Query Graphs
  2. 2025Knowledge-Aware Iterative Retrieval for Multi-Agent Systems
  3. 2025An Empirical Study on Reinforcement Learning for Reasoning-Search Interleaved LLM Agents
  4. 2025Reinforcement Fine-Tuning for Reasoning towards Multi-Step Multi-Source Search in Large Language Models
  5. 2026PaperSearchQA: Learning to Search and Reason over Scientific Papers with RLVR
  6. 2026Search, Do not Guess: Teaching Small Language Models to Be Effective Search Agents
  7. 2026When Search Agents Should Ask: DiscoBench for Clarification-Aware Deep Search
  8. 2025Beyond Correctness: Rewarding Faithful Reasoning in Retrieval-Augmented Generation