Earth-Agent-Pro: Towards Real-World Full-Chain Earth Observation with Agents

• URL: http://arxiv.org/abs/2609.12533v1
• Date: Fri, 11 Sep 2026 07:41:36 GMT
• Title: Earth-Agent-Pro: Towards Real-World Full-Chain Earth Observation with Agents
• Abstract summary: Real-world Earth observation (EO) agents must translate high-level scientific questions into executables.<n>Earth-Bench-Pro instantiates 248 expert-curated task cores as 744 questions under three matched regimes.<n>With a shared GPT-5 backbone, Earth-Agent-Pro achieves 66.13% LLM-as-Judge accuracy.
• Score: 41.45703910271276
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Real-world Earth observation (EO) agents must translate high-level scientific questions into executable workflows to acquire observations, prepare data, perform domain computations, and derive conclusions from runtime evidence. Existing EO agents typically start from supplied observations, while benchmarks typically provide prepared inputs or candidate answers, leaving full-chain open-world EO execution largely untested. We present Earth-Agent-Pro, an execution-adaptive Plan-and-Execute framework using expert-authored skills to constrain planning and runtime tool use. Workflow-centered structured memory records planned steps, accepted evidence, and their dependencies, enabling repair of only the affected workflow suffix when runtime evidence invalidates a step. Separate large language model adapters use sequence-level supervised fine-tuning for planner workflow composition and node-level group relative policy optimization with locally verifiable rewards for executor tool-argument grounding. Earth-Bench-Pro instantiates 248 expert-curated task cores as 744 questions under three matched regimes. Its 248 Open-World Execution questions span RGB imagery, spectral observations, and remote sensing products, pairing high-level requests with runtime data requirements, executable trajectories, and open-ended answers grounded in execution evidence. With a shared GPT-5 backbone, Earth-Agent-Pro achieves 66.13% LLM-as-Judge accuracy, exceeding ReAct by 20.95 points in this metric and 24.44 points in Tools-In-Order. Joint adapter tuning raises Qwen3.5-9B LLM-as-Judge accuracy from 38.31% to 50.00%, an 11.69-point gain over the untuned configuration. Planning-only evaluation and execution with the reference workflow show that the adapters improve workflow composition and argument grounding, respectively. Code and datasets will be released soon.

Related papers

• GeoForge: Non-Parametric Self-Evolving Agents for Earth-Observation Reasoning [25.382191805848013]
  We present GeoForge, a training-free, self-evolving framework that transforms completed trajectories into a structured nonparametric execution state.<n>GeoForge constrains the operation space according to the sensing context, then retrieves a task-conditioned prior from three complementary memories.<n>After each task, a safety-gated distillation process converts grounded trajectories into reusable execution knowledge for future retrieval.
  arXiv  Detail & Related papers  (2026-08-11T05:11:10Z)
• A corrective agentic hybrid RAG and an operations-grounded evaluation for a scientific facility [0.0]
  We present APS-RAG, Advanced Photon Source Retrieval Augmented Generation.<n>The retrieval engine fuses dense, sparse, and knowledge-graph channels with query-type-adaptive reciprocal-rank fusion.<n>We construct APS-Bench, a 50-question, question-answering (QA) dataset with auditable gold answers.
  arXiv  Detail & Related papers  (2026-07-27T17:01:30Z)
• ACCORD: Action-Conditioned Contextual Grounding for Language Agents [69.73525608707764]
  ACCORD (Action-Conditioned Contextual Grounding) is an agent framework for adaptive grounding.<n>We show that current agents often fail to do so. They act from assumed rather than observed specifics, overlook information they could have gathered, and fail to incorporate evidence that has already been returned.
  arXiv  Detail & Related papers  (2026-06-15T09:05:55Z)
• EEVEE: Towards Test-time Prompt Learning in the Real World for Self-Improving Agents [64.96332056338923]
  EEVEE is the first multi-dataset test-time prompt learning framework for LLM agents.<n>It enables test-time prompt learning under real-world task streams.<n>EEVEE improves average multi-benchmark scores by 10.38 and 24.32 points over Qwen3-4B-Instruct and DeepSeek-V3.2.
  arXiv  Detail & Related papers  (2026-06-09T17:57:16Z)
• GTA-2: Benchmarking General Tool Agents from Atomic Tool-Use to Open-Ended Workflows [90.35728421223673]
  GTA-2 is a hierarchical benchmark for General Tool Agents (GTA)<n>Built on real-world authenticity, it leverages real user queries, deployed tools, and multimodal contexts.<n>Experiments reveal a pronounced capability cliff: while frontier models already struggle on atomic tasks, top models achieving only 14.39% success.
  arXiv  Detail & Related papers  (2026-04-17T05:36:00Z)
• GUIDE: Interpretable GUI Agent Evaluation via Hierarchical Diagnosis [17.279226216630065]
  We introduce GUIDE (GUI Understanding and Interpretable Diagnostic Evaluation), a framework that decomposes trajectory assessment into three sequential stages.<n>Subtask Diagnosis evaluates each unit in context, assigning a completion verdict and generating a structured error analysis with corrective recommendations.<n>We validate GUIDE on three benchmarks: an industrial e-commerce dataset of 932 trajectories, AGENTREWARDBENCH spanning five web agent tasks with 1302 trajectories, and AndroidBench for mobile device control.
  arXiv  Detail & Related papers  (2026-04-06T03:58:43Z)
• EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair [18.22665454212814]
  EigenData is an integrated, self-evolving platform that automates the full data lifecycle through a multi-agent architecture.<n>We apply EigenData to audit and repair the Berkeley Function-Calling Leaderboard (BFCL-V3)<n>We demonstrate that the repaired benchmark, coupled with outcome-aware metrics, produces model rankings substantially better correlated with human judgments of functional correctness.
  arXiv  Detail & Related papers  (2026-03-05T04:58:38Z)
• IC-EO: Interpretable Code-based assistant for Earth Observation [5.490577355567582]
  Earth Observation (EO) analysis remains difficult to perform for the laymen.<n>Many systems return black-box predictions that are difficult to audit or reproduce.<n>This study proposes a conversational, code-language agent that transforms natural-language queries into executable Python queries.
  arXiv  Detail & Related papers  (2026-01-27T10:49:29Z)
• Visual Document Understanding and Question Answering: A Multi-Agent Collaboration Framework with Test-Time Scaling [83.78874399606379]
  We propose MACT, a Multi-Agent Collaboration framework with Test-Time scaling.<n>It comprises four distinct small-scale agents, with clearly defined roles and effective collaboration.<n>It shows superior performance with a smaller parameter scale without sacrificing the ability of general and mathematical tasks.
  arXiv  Detail & Related papers  (2025-08-05T12:52:09Z)
• ThinkGeo: Evaluating Tool-Augmented Agents for Remote Sensing Tasks [64.86209459039313]
  ThinkGeo is an agentic benchmark designed to evaluate tool-augmented agents on remote sensing tasks via structured tool use and multi-step planning.<n>We implement a ReAct-style interaction loop and evaluate both open and closed-source LLMs on 486 structured agentic tasks with 1,773 expert-verified reasoning steps.<n>Our analysis reveals notable disparities in tool accuracy and planning consistency across models.
  arXiv  Detail & Related papers  (2025-05-29T17:59:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.