{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/climategpt-towards-ai-synthesizing","title":"ClimateGPT: Towards AI Synthesizing Interdisciplinary Research on Climate Change","arxiv_id":"2401.09646","date":"2024-01-17","proceeding":null,"authors":["David Thulke","Yingbo Gao","Petrus Pelser","Rein Brune","Rricha Jalota","Floris Fok","Michael Ramos","Ian van Wyk","Abdallah Nasir","Hayden Goldstein","Taylor Tragemann","Katie Nguyen","Ariana Fowler","Andrew Stanco","Jon Gabriel","Jordan Taylor","Dean Moro","Evgenii Tsymbalov","Juliette de Waal","Evgeny Matusov","Mudar Yaghi","Mohammad Shihadah","Hermann Ney","Christian Dugast","Jonathan Dotan","Daniel Erasmus"],"abstract":"This paper introduces ClimateGPT, a model family of domain-specific large language models that synthesize interdisciplinary research on climate change. We trained two 7B models from scratch on a science-oriented dataset of 300B tokens. For the first model, the 4.2B domain-specific tokens were included during pre-training and the second was adapted to the climate domain after pre-training. Additionally, ClimateGPT-7B, 13B and 70B are continuously pre-trained from Llama~2 on a domain-specific dataset of 4.2B tokens. Each model is instruction fine-tuned on a high-quality and human-generated domain-specific dataset that has been created in close cooperation with climate scientists. To reduce the number of hallucinations, we optimize the model for retrieval augmentation and propose a hierarchical retrieval strategy. To increase the accessibility of our model to non-English speakers, we propose to make use of cascaded machine translation and show that this approach can perform comparably to natively multilingual models while being easier to scale to a large number of languages. Further, to address the intrinsic interdisciplinary aspect of climate change we consider different research perspectives. Therefore, the model can produce in-depth answers focusing on different perspectives in addition to an overall answer. We propose a suite of automatic climate-specific benchmarks to evaluate LLMs. On these benchmarks, ClimateGPT-7B performs on par with the ten times larger Llama-2-70B Chat model while not degrading results on general domain benchmarks. Our human evaluation confirms the trends we saw in our benchmarks. All models were trained and evaluated using renewable energy and are released publicly.","url_abs":"https://arxiv.org/abs/2401.09646v1","url_pdf":"https://arxiv.org/pdf/2401.09646v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"climategpt-towards-ai-synthesizing","repo_url":"https://github.com/eci-io/climategpt-evaluation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"climategpt-towards-ai-synthesizing","repo_url":"https://github.com/nlp-rise/climateeval","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2401.09646","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2401.09646"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/eci-io/climategpt-evaluation","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nlp-rise/climateeval","reach":{"status":"ok"}}],"summary":{"ran":19,"unverified":1},"by_repo_kind":{"official":{"samples":20,"ran":19,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":20,"samples":[{"code_sha256_prefix":"a9c6563fa889a383","entry":"agg_f1","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/pira_mcq/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/pira_mcq/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a9c6563fa889a383"}},{"code_sha256_prefix":"5b9f76792332a73c","entry":"agg_f1","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/climabench/clima_text/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/climabench/clima_text/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5b9f76792332a73c"}},{"code_sha256_prefix":"f9b12c852066f5eb","entry":"agg_f1","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/climabench/climate_eng/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/climabench/climate_eng/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f9b12c852066f5eb"}},{"code_sha256_prefix":"6af37f58f53ce56c","entry":"convert_to_int","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/pira_mcq/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/pira_mcq/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6af37f58f53ce56c"}},{"code_sha256_prefix":"9ca40505158717f6","entry":"f1","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/exeter/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/exeter/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9ca40505158717f6"}},{"code_sha256_prefix":"1eb73c7c88a4339a","entry":"f1","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/pira_mcq/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/pira_mcq/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1eb73c7c88a4339a"}},{"code_sha256_prefix":"6ca45f4106331bb0","entry":"f1","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/climabench/clima_text/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/climabench/clima_text/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6ca45f4106331bb0"}},{"code_sha256_prefix":"b4d2038fc921868c","entry":"filter_binary","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/climabench/clima_text/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/climabench/clima_text/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b4d2038fc921868c"}},{"code_sha256_prefix":"1adec589d9a04b82","entry":"filter_binary","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/climabench/climate_fever/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/climabench/climate_fever/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1adec589d9a04b82"}},{"code_sha256_prefix":"90b219e118424010","entry":"filter_main","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/climabench/cdp_qa/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/climabench/cdp_qa/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"90b219e118424010"}},{"code_sha256_prefix":"1cd2e2080bc7652b","entry":"filter_multi","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/climabench/cdp_qa/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/climabench/cdp_qa/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1cd2e2080bc7652b"}},{"code_sha256_prefix":"ca267189a9d78cfe","entry":"filter_multi","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/climabench/climate_eng/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/climabench/climate_eng/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ca267189a9d78cfe"}},{"code_sha256_prefix":"713068816eb94d0f","entry":"get_claim","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/exeter/gen_exeter_data.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/exeter/gen_exeter_data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"713068816eb94d0f"}},{"code_sha256_prefix":"b0580a8baab609a0","entry":"get_claim_label","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/exeter/gen_exeter_data.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/exeter/gen_exeter_data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b0580a8baab609a0"}},{"code_sha256_prefix":"61628eeb493a5514","entry":"multi_f1","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/exeter/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/exeter/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"61628eeb493a5514"}},{"code_sha256_prefix":"b478e459d66cd559","entry":"process_docs","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/exams/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/exams/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b478e459d66cd559"}},{"code_sha256_prefix":"d7b3301efb4113f2","entry":"process_docs_filter_science","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/exams/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/exams/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d7b3301efb4113f2"}},{"code_sha256_prefix":"8270397adb29614e","entry":"relabel_labels","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/exams/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/exams/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8270397adb29614e"}},{"code_sha256_prefix":"b421e04e83d5f1d4","entry":"select_first_1100","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/climabench/cdp_qa/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/climabench/cdp_qa/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b421e04e83d5f1d4"}},{"code_sha256_prefix":"15e0548c0a4040d5","entry":"agg_f1","repo":"eci-io/climategpt-evaluation","repo_kind":"official","path":"tasks/exeter/utils.py","file_url":"https://github.com/eci-io/climategpt-evaluation/blob/HEAD/tasks/exeter/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"15e0548c0a4040d5"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}