Pith. sign in

REVIEW 6 cited by

The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrink

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2204.05149 v1 pith:W26O6OW4 submitted 2022-04-11 cs.LG cs.AIcs.GL

classification cs.LGcs.AIcs.GL
keywords emissionsbestcarbonenergypracticestrainingfootprintimportance
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Machine Learning (ML) workloads have rapidly grown in importance, but raised concerns about their carbon footprint. Four best practices can reduce ML training energy by up to 100x and CO2 emissions up to 1000x. By following best practices, overall ML energy use (across research, development, and production) held steady at <15% of Google's total energy use for the past three years. If the whole ML field were to adopt best practices, total carbon emissions from training would reduce. Hence, we recommend that ML papers include emissions explicitly to foster competition on more than just model quality. Estimates of emissions in papers that omitted them have been off 100x-100,000x, so publishing emissions has the added benefit of ensuring accurate accounting. Given the importance of climate change, we must get the numbers right to make certain that we work on its biggest challenges.

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Opti-Q: A Constraint-Based Optimization Framework for Multi-LLM Question Planning

    cs.AI 2026-06 conditional novelty 6.0 of 10

    Per-question database-style plan search over multi-LLM DAGs improves QA quality under budgets by ~58% (MMLU-Pro) and ~41% (SimpleQA) versus reimplemented baselines.

  2. Reasoning-Aware Multimodal Fusion for Hateful Video Detection

    cs.CV 2025-12 conditional novelty 6.0 of 10

    RAMF's three-stage adversarial VLM reasoning plus local-global/cross-head attention fusion improves hateful video classification on HateMM and MultiHateClip.

  3. Promptception: How Sensitive Are Large Multimodal Models to Prompts?

    cs.CV 2025-09 conditional novelty 6.0 of 10

    A study of 61 prompt variants across 10 vision-language models and 3 benchmarks finds accuracy swings of up to 15 points, with proprietary models more sensitive than open-source ones.

  4. SiPipe: Bridging the CPU-GPU Utilization Gap for Efficient Pipeline-Parallel LLM Inference

    cs.DC 2025-06 conditional novelty 6.0 of 10

    SiPipe claims 1.6x to 2.1x throughput gains over vLLM for pipeline-parallel LLM inference by using idle CPUs for sampling, double-buffered CUDA graphs, and structure-aware transmission.

  5. A Theory of Inference Compute Scaling: Reasoning through Directed Stochastic Skill Search

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A skill-graph random-walk model gives closed-form accuracy-versus-compute formulas for four reasoning strategies and connects them to training scaling.

  6. A Comprehensive Graph Framework for Question Answering with Mode-Seeking Preference Alignment

    cs.CL 2025-06 conditional novelty 4.0 of 10

    GraphMPA combines an embedding-similarity hierarchical graph with mode-seeking preference optimization to improve RAG question answering on six datasets.

Pith tools