arXiv:2605.13743v2 Announce Type: replace Abstract: Open datasets and benchmarks for entity-level carbon-emission prediction remain fragmented across access, scale, granularity, and evaluation. We introduce GHGbench, an open dataset and benchmark for company- and building-level greenhouse-gas prediction. The company track contains 32,000+ company-year records from 12,000+ firms with Scope 1+2 and Scope 3 disclosures and financial/sectoral signals; the building track harmonises 491,591 building-year records from 13 open sources into a single schema across 26 metropolitan areas (10 U.S., 15 Aust
The increasing focus on Scope 3 emissions and corporate sustainability reporting drives the need for standardized, entity-level carbon prediction and benchmarking tools.
This benchmark provides critical open data and evaluation tools for developing AI models that predict greenhouse gas emissions, essential for accountability, carbon accounting, and strategic sustainability initiatives.
The availability of GHGbench enables more robust and consistent development of AI solutions for granular carbon emission forecasting, moving beyond fragmented datasets.
- · AI/ML researchers
- · ESG reporting platforms
- · Companies with high carbon emissions
- · Climate tech startups
- · Companies with opaque carbon reporting
- · Manual carbon accounting services
Improved accuracy and comparability of AI-driven carbon emission predictions for companies and buildings becomes possible.
Enhanced public and regulatory scrutiny of corporate carbon footprints, potentially influencing investment decisions and market valuations.
Standardized emission prediction models could form the basis for dynamic carbon pricing mechanisms or new forms of carbon credit markets.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG