{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:L3PNEVFRT2OW2JUOGWMCFSY2LA","short_pith_number":"pith:L3PNEVFR","schema_version":"1.0","canonical_sha256":"5eded254b19e9d6d268e359822cb1a581fefa8b1b081c67341970e00dcc3ed9b","source":{"kind":"arxiv","id":"1910.09700","version":2},"attestation_state":"computed","paper":{"title":"Quantifying the Carbon Emissions of Machine Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"A tool called the Machine Learning Emissions Calculator approximates the carbon emissions of training neural networks based on server location, energy grid, training duration, and hardware.","cross_cats":["cs.LG"],"primary_cat":"cs.CY","authors_text":"Alexandra Luccioni, Alexandre Lacoste, Thomas Dandres, Victor Schmidt","submitted_at":"2019-10-21T23:57:32Z","abstract_excerpt":"From an environmental standpoint, there are a few crucial aspects of training a neural network that have a major impact on the quantity of carbon that it emits. These factors include: the location of the server used for training and the energy grid that it uses, the length of the training procedure, and even the make and model of hardware on which the training takes place. In order to approximate these emissions, we present our Machine Learning Emissions Calculator, a tool for our community to better understand the environmental impact of training ML models. We accompany this tool with an expl"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":true},"canonical_record":{"source":{"id":"1910.09700","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CY","submitted_at":"2019-10-21T23:57:32Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"53c718faef53841f47f977dc8641503cb9d4fc1a9e2602604343414e85ae24bc","abstract_canon_sha256":"d4ce882e1d4e646bf26fc5ac7996efe579ab08f03782965320b110c07728eadd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:53.339536Z","signature_b64":"QILvlcBCkPI1w9QBbixIcxekDUIFFLXHWRWm6KOv7x/RPgRz/+VvrdHMSVJjuDuBxAM2p8RVYeunwkSFlbplCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5eded254b19e9d6d268e359822cb1a581fefa8b1b081c67341970e00dcc3ed9b","last_reissued_at":"2026-05-17T23:38:53.338922Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:53.338922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Quantifying the Carbon Emissions of Machine Learning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"A tool called the Machine Learning Emissions Calculator approximates the carbon emissions of training neural networks based on server location, energy grid, training duration, and hardware.","cross_cats":["cs.LG"],"primary_cat":"cs.CY","authors_text":"Alexandra Luccioni, Alexandre Lacoste, Thomas Dandres, Victor Schmidt","submitted_at":"2019-10-21T23:57:32Z","abstract_excerpt":"From an environmental standpoint, there are a few crucial aspects of training a neural network that have a major impact on the quantity of carbon that it emits. These factors include: the location of the server used for training and the energy grid that it uses, the length of the training procedure, and even the make and model of hardware on which the training takes place. In order to approximate these emissions, we present our Machine Learning Emissions Calculator, a tool for our community to better understand the environmental impact of training ML models. We accompany this tool with an expl"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we present our Machine Learning Emissions Calculator, a tool for our community to better understand the environmental impact of training ML models","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The listed factors (server location, energy grid, training length, hardware) are sufficient to accurately approximate emissions and that the tool's estimates will be reliable enough to guide mitigation decisions.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Presents a calculator tool for estimating carbon emissions from ML model training along with mitigation actions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A tool called the Machine Learning Emissions Calculator approximates the carbon emissions of training neural networks based on server location, energy grid, training duration, and hardware.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"c2958f2972083cbf7e70fa9f5432665589dd39acaa2d66bf5216f7daf89b6e1f"},"source":{"id":"1910.09700","kind":"arxiv","version":2},"verdict":{"id":"c9340fbc-6a49-48f0-b80c-0d6996b91fde","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T05:47:52.538039Z","strongest_claim":"we present our Machine Learning Emissions Calculator, a tool for our community to better understand the environmental impact of training ML models","one_line_summary":"Presents a calculator tool for estimating carbon emissions from ML model training along with mitigation actions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The listed factors (server location, energy grid, training length, hardware) are sufficient to accurately approximate emissions and that the tool's estimates will be reliable enough to guide mitigation decisions.","pith_extraction_headline":"A tool called the Machine Learning Emissions Calculator approximates the carbon emissions of training neural networks based on server location, energy grid, training duration, and hardware."},"references":{"count":25,"sample":[{"doi":"","year":1906,"title":"Energy and policy considerations for deep learning in nlp","work_id":"33ac678f-b75d-4caf-b8a6-9a4d65b1748c","ref_index":1,"cited_arxiv_id":"1906.02243","is_internal_anchor":true},{"doi":"","year":1907,"title":"Emma Strubell, Ananya Ganesh, and Andrew McCallum","work_id":"2f2ecc7f-a9df-4003-a924-900d790909d9","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2006,"title":"Institute for Global Environmental Strategies Hayama, Japan, 2006","work_id":"cdc5d24e-6cb6-4209-968e-bd6036fa39c8","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2002,"title":"Ghg emissions from electricity consumption: A case study of hong kong from 2002 to 2015 and trends to 2030","work_id":"19c7fb3a-bb23-477b-82b7-b6fc85cb31a7","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2011,"title":"Electricity- specific emission factors for grid electricity.Ecometrica, Emissionfactors","work_id":"a2d83cb4-911a-40f6-a569-85381c48b0f5","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":25,"snapshot_sha256":"d699f907ffce6da5e415eb0c900a3077e07088ebcd840e41a04f1db644cae7f2","internal_anchors":8},"formal_canon":{"evidence_count":2,"snapshot_sha256":"4047b89aadfcf197b412157147df3a98871e4082651894db2a7213dc3f5d8493"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"1910.09700","created_at":"2026-05-17T23:38:53.339010+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.09700v2","created_at":"2026-05-17T23:38:53.339010+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.09700","created_at":"2026-05-17T23:38:53.339010+00:00"},{"alias_kind":"pith_short_12","alias_value":"L3PNEVFRT2OW","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_16","alias_value":"L3PNEVFRT2OW2JUO","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_8","alias_value":"L3PNEVFR","created_at":"2026-05-18T12:33:21.387695+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":36,"internal_anchor_count":36,"sample":[{"citing_arxiv_id":"2204.06745","citing_title":"GPT-NeoX-20B: An Open-Source Autoregressive Language Model","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2503.08223","citing_title":"Will LLMs Scaling Hit the Wall? Breaking Barriers via Distributed Resources on Massive Edge Devices","ref_index":127,"is_internal_anchor":true},{"citing_arxiv_id":"2503.10666","citing_title":"Green Prompting: Characterizing Prompt-driven Energy Costs of LLM Inference","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06597","citing_title":"UniSD: Towards a Unified Self-Distillation Framework for Large Language Models","ref_index":52,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22393","citing_title":"Nf-PEAK: Process-Based Energy Attribution for Nextflow Workflows on Kubernetes Clusters","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18893","citing_title":"Position: Graph Condensation Needs a Reset -- Move Beyond Full-dataset Training and Model-Dependence","ref_index":92,"is_internal_anchor":true},{"citing_arxiv_id":"2511.06943","citing_title":"PlantTraitNet: An Uncertainty-Aware Multimodal Framework for Global-Scale Plant Trait Inference from Citizen Science Data","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2511.04776","citing_title":"Quantifying the Climate Risk of Generative AI: Region-Aware Carbon Accounting with G-TRACE and the AI Sustainability Pyramid","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14624","citing_title":"An Amortized Efficiency Threshold for Comparing Neural and Heuristic Solvers in Combinatorial Optimization","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18893","citing_title":"Position: Graph Condensation Needs a Reset -- Move Beyond Full-dataset Training and Model-Dependence","ref_index":92,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19605","citing_title":"deadtrees.earth-aerial: A Multi-Resolution Aerial Image Dataset for Tree Cover and Mortality Detection","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2507.02850","citing_title":"LLM Hypnosis: Exploiting User Feedback for Unauthorized Knowledge Injection to All Users","ref_index":32,"is_internal_anchor":true},{"citing_arxiv_id":"2511.04776","citing_title":"Quantifying the Climate Risk of Generative AI: Region-Aware Carbon Accounting with G-TRACE and the AI Sustainability Pyramid","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2511.16719","citing_title":"SAM 3: Segment Anything with Concepts","ref_index":63,"is_internal_anchor":true},{"citing_arxiv_id":"2303.09014","citing_title":"ART: Automatic multi-step reasoning and tool-use for large language models","ref_index":129,"is_internal_anchor":true},{"citing_arxiv_id":"2601.22487","citing_title":"Coordinating GPU Data Centers and Power Grid Regulation Service for Exogenous Carbon Benefits","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14055","citing_title":"PEML: Parameter-efficient Multi-Task Learning with Optimized Continuous Prompts","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14624","citing_title":"An Amortized Efficiency Threshold for Comparing Neural and Heuristic Solvers in Combinatorial Optimization","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14249","citing_title":"EnergyLens: Predictive Energy-Aware Exploration for Multi-GPU LLM Inference Optimization","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14550","citing_title":"Multi-Dimensional Model Integrity and Responsibility Assessment Index and Scoring Framework","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2110.08207","citing_title":"Multitask Prompted Training Enables Zero-Shot Task Generalization","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11764","citing_title":"Decomposing the Generalization Gap in PROTAC Activity Prediction: Variance Attribution and the Inter-Laboratory Ceiling","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11733","citing_title":"Position: LLM Inference Should Be Evaluated as Energy-to-Token Production","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2309.14509","citing_title":"DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models","ref_index":158,"is_internal_anchor":true},{"citing_arxiv_id":"2402.19173","citing_title":"StarCoder 2 and The Stack v2: The Next Generation","ref_index":220,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L3PNEVFRT2OW2JUOGWMCFSY2LA","json":"https://pith.science/pith/L3PNEVFRT2OW2JUOGWMCFSY2LA.json","graph_json":"https://pith.science/api/pith-number/L3PNEVFRT2OW2JUOGWMCFSY2LA/graph.json","events_json":"https://pith.science/api/pith-number/L3PNEVFRT2OW2JUOGWMCFSY2LA/events.json","paper":"https://pith.science/paper/L3PNEVFR"},"agent_actions":{"view_html":"https://pith.science/pith/L3PNEVFRT2OW2JUOGWMCFSY2LA","download_json":"https://pith.science/pith/L3PNEVFRT2OW2JUOGWMCFSY2LA.json","view_paper":"https://pith.science/paper/L3PNEVFR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.09700&json=true","fetch_graph":"https://pith.science/api/pith-number/L3PNEVFRT2OW2JUOGWMCFSY2LA/graph.json","fetch_events":"https://pith.science/api/pith-number/L3PNEVFRT2OW2JUOGWMCFSY2LA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L3PNEVFRT2OW2JUOGWMCFSY2LA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L3PNEVFRT2OW2JUOGWMCFSY2LA/action/storage_attestation","attest_author":"https://pith.science/pith/L3PNEVFRT2OW2JUOGWMCFSY2LA/action/author_attestation","sign_citation":"https://pith.science/pith/L3PNEVFRT2OW2JUOGWMCFSY2LA/action/citation_signature","submit_replication":"https://pith.science/pith/L3PNEVFRT2OW2JUOGWMCFSY2LA/action/replication_record"}},"created_at":"2026-05-17T23:38:53.339010+00:00","updated_at":"2026-05-17T23:38:53.339010+00:00"}