Organization · Asia
Antarctica
Antarctica is a climate technology company delivering enterprise-grade measurement, observability, and intelligent automation for AI and IT systems.
Scientific Paper
Don't Look Up: Evaluating the Tradeoff Between Performance and Sustainability of Text Classification Using Open LLMs
Hamilton-Palicki et al. · 2026-09-02
What it covers
The study compares performance, energy use, processing time, and carbon emissions across dictionaries, trained classifiers, and self-hosted open LLMs for three text-analysis tasks.
Who it helps
Intended audience has not been documented.
Limitations
Source-specific limitations have not been reviewed. Inclusion and topic labels do not establish effectiveness or reliability.
Show citation
Hamilton-Palicki et al.. 2026-09-02. Don't Look Up: Evaluating the Tradeoff Between Performance and Sustainability of Text Classification Using Open LLMs. https://journals.sagepub.com/doi/10.1177/08944393261473230 (AI & Environment Resource Hub; record atlas-744312b6b2a2; collection snapshot 2026-10-04).
Classification and context
Included in the original Scientific Paper collection. Labels below are metadata based suggestions. They are not verified findings, claims of effectiveness, or endorsements.
topic
lifecycle
Available source metadata
- Author or creator
- Hamilton-Palicki et al.
- Publisher
- Social Science Computer Review
- Publication date
- 2026-09-02
- Date precision
- day
- Geographic scope
- Unknown
- Language
- unknown
- Peer review
- unknown
- Source access
- Automated access restricted
- Metadata confidence
- Metadata unconfirmed
- Last verified
- Not yet verified
- Snapshot import
- 2026-09-23
- Legacy domain
- AI & Data Centers/Hardware
- Legacy subdomain
- Sustainability of Text Classification Using Open LLMs
- journal
- Social Science Computer Review
- source evidence
- https://journals.sagepub.com/doi/10.1177/08944393261473230
- metadata reviewed at
- 2026-09-20
Provenance and review
- Source access
- Automated access restricted
- Metadata confidence
- Metadata unconfirmed
- Editorial relevance
- Relevance has not been separately reviewed.
- Evidence quality
- Evidence quality has not been separately assessed.
Source sheet: Scientific Paper · Row 523 · Original ID: paper-519.
A paper listing is not a quality assessment. Peer review and findings require source-level confirmation. Source blocks automated access · checked 2026-09-28. Metadata not yet corroborated. The import date is not the original date added.
Suggest a correction for this record →Newsletter appearances
Publication, placement, and evidence are tracked separately. A legacy coverage flag alone does not prove an appearance.
- Issue 59 · published · resource list · Issue evidence ↗ · Preserved from paper-519
Exact source link in saved original issue resource section: Scientific Papers 📄
Related through shared environmental topics
Ordered by the number of shared provisional topic labels. This indicates a browsing connection, not agreement between sources.
Shared topics: Cross-cutting sustainability; Energy and electricity; Climate and greenhouse gases
Shared topics: Cross-cutting sustainability; Research tools and geospatial methods; Energy and electricity
Scientific Paper · 2026-02-23
Brain-Inspired Energy-Efficient Technologies for Next-Generation Artificial Intelligence
The paper argues that neurobiological principles can inform ultra-low-power AI at both hardware and software levels, and calls for long-horizon industry-academic partnerships in energy-efficient NeuroAI.
Shared topics: Research tools and geospatial methods; Energy and electricity; Climate and greenhouse gases
Tool
CLEER Dashboard
Sustainable AI Group's public dashboard compares modeled AI inference energy and CO₂-eq by model, provider and session, including data center overhead assumptions. Water estimates and API are marked coming soon.
Shared topics: Research tools and geospatial methods; Energy and electricity; Climate and greenhouse gases
Scientific Paper · 2026-10-16
Energy-Aware Earth System Science: A Community Roadmap
This preprint examines computing demands and model skill in Earth system science and proposes joint reporting of predictive performance and energy use, including for machine learning weather models.