{"paper":{"title":"GCA Framework: A GCC Countries-Grounded Dataset and Agentic Pipeline for Climate Decision Support","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Domain fine-tuning on a GCC-grounded climate dataset plus tool integration raises LLM reliability for regional decision support.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fahad Shahbaz Khan, Khawar Shehzad, Muhammad Haris Khan, Muhammad Umer Sheikh, Salman Khan","submitted_at":"2026-04-14T05:31:40Z","abstract_excerpt":"Climate decision-making in the GCC states increasingly demands systems that can translate heterogeneous scientific and policy evidence into actionable guidance, yet general-purpose large language models (LLMs) remain weak both in region-specific climate knowledge and grounded interaction with geospatial and forecasting tools. We present the GCA framework, which unifies (i) GCA-DS, a curated multimodal dataset grounded in the GCC states, and (ii) Gulf Climate Agent (GCA), a tool-augmented agent for climate analysis. GCA-DS comprises 200k question--answer pairs spanning governmental policies and"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Domain fine-tuning and tool integration substantially improve reliability over general-purpose baselines on climate tasks in the GCC states.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The curated 200k QA pairs and remote-sensing inputs are sufficiently representative and high-quality to ground the agent for real decision support.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A GCC-grounded multimodal dataset and tool-augmented agent improve LLM performance on regional climate analysis tasks over general baselines.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Domain fine-tuning on a GCC-grounded climate dataset plus tool integration raises LLM reliability for regional decision support.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9553a5d7441be93bf7f760ad568738ffe4b3e7d2656d352b67764245ffb94318"},"source":{"id":"2604.12306","kind":"arxiv","version":3},"verdict":{"id":"8e9fa35c-132a-4098-a532-02f7bf9a8779","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T16:06:56.894635Z","strongest_claim":"Domain fine-tuning and tool integration substantially improve reliability over general-purpose baselines on climate tasks in the GCC states.","one_line_summary":"A GCC-grounded multimodal dataset and tool-augmented agent improve LLM performance on regional climate analysis tasks over general baselines.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The curated 200k QA pairs and remote-sensing inputs are sufficiently representative and high-quality to ground the agent for real decision support.","pith_extraction_headline":"Domain fine-tuning on a GCC-grounded climate dataset plus tool integration raises LLM reliability for regional decision support."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.12306/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}