FAIR+S extends FAIR with sustainability metrics and is validated via expert survey confirming importance but revealing awareness gaps in green practices.
hub Canonical reference
Quantifying the Carbon Emissions of Machine Learning
Canonical reference. 71% of citing Pith papers cite this work as background.
41
Pith papers citing it
Background
71% of classified citations
abstract
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 explanation of the factors cited above, as well as concrete actions that individual practitioners and organizations can take to mitigate their carbon emissions.
hub tools
citation-role summary
background 7
citation-polarity summary
roles
background 7representative citing papers
Enginuity is the first open benchmark dataset for VLMs on engineering diagrams, with evaluations showing models identify parts but produce low-fidelity descriptions and struggle with factual reasoning.
Releases DTE-aerial-train (385K patches) and DTE-aerial-bench (25 global orthoimages) as the first harmonized multi-resolution datasets for joint tree cover and mortality segmentation across biomes.
Introduces the Amortized Efficiency Threshold (AET) to identify the deployment volume at which neural combinatorial optimization solvers achieve lower total energy use than heuristic baselines after accounting for training costs.
Nearly every arXiv submission leaks hidden sensitive information through its source files, existing cleaners fail, and ALC-NG provides a more reliable fix.
SAM 3 introduces promptable concept segmentation that doubles accuracy of prior systems on images and videos while improving standard SAM segmentation performance.
A promptable model trained on 1B masks achieves competitive zero-shot segmentation performance across tasks and is released publicly with its dataset.
OPT releases open decoder-only transformers up to 175B parameters that match GPT-3 performance at one-seventh the carbon cost, along with code and training logs.
Multitask fine-tuning of an encoder-decoder model on prompted datasets produces zero-shot generalization that often beats models up to 16 times larger on standard benchmarks.
Manual audit shows web-scraped Lombard corpora are largely noisy and biased toward Western varieties over Eastern ones.
BaltiVoice supplies the first public read-speech corpus and fine-tuned Whisper ASR for Balti, reaching 26.74% WER on a speaker-disjoint validation set.
Nf-PEAK is a containerized method that attributes energy to Nextflow tasks with 6.6% MAPE in isolated runs and 10.9% under co-located load, outperforming Kepler on nf-core workflows.
EnergyLens predicts multi-GPU LLM inference energy consumption with 9-13% MAPE and identifies configurations with up to 52x energy efficiency differences.
PEML co-optimizes continuous prompts and low-rank adaptations to deliver up to 6.67% average accuracy gains over existing multi-task PEFT methods on GLUE, SuperGLUE, and other benchmarks.
Inter-laboratory measurement variance dominates the generalization gap in PROTAC activity prediction, capping LOTO AUROC near 0.67 across models and architectures.
UniSD unifies self-distillation components for autoregressive LLMs and its full integrated version improves base models by 5.4 points and baselines by 2.8 points across six benchmarks.
PlantTraitNet is an uncertainty-aware multimodal deep learning framework that infers four plant traits from citizen science images and produces global trait maps that outperform prior products when validated against independent survey data.
A single attacker can use strategic upvoting and downvoting on language model outputs to inject facts, security flaws, or fake news that persist in the model for all users after preference tuning.
SAM 2 delivers more accurate video segmentation with 3x fewer user interactions and 6x faster image segmentation than the original SAM by training a streaming-memory transformer on the largest video segmentation dataset collected to date.
StarCoder2-15B matches or beats CodeLlama-34B on code tasks despite being smaller, and StarCoder2-3B outperforms prior 15B models, with open weights and exact training data identifiers released.
DeepSpeed-Ulysses keeps communication volume constant for sequence-parallel attention when sequence length and device count scale together, delivering 2.5x faster training on 4x longer sequences than prior SOTA.
ART automatically generates multi-step reasoning programs with tool integration for LLMs, yielding substantial gains over few-shot and auto-CoT prompting on BigBench and MMLU while matching hand-crafted CoT on most tasks.
BLOOM is a 176B-parameter open-access multilingual language model trained on the ROOTS corpus that achieves competitive performance on benchmarks, with improved results after multitask prompted finetuning.
GPT-NeoX-20B is a publicly released 20B parameter autoregressive language model trained on the Pile that shows strong gains in five-shot reasoning over similarly sized prior models.
citing papers explorer
-
FAIR+S: A validation study of a framework for sustainable research data and software
FAIR+S extends FAIR with sustainability metrics and is validated via expert survey confirming importance but revealing awareness gaps in green practices.
-
Enginuity: A Dataset and Benchmark for Vision-Language Understanding of Engineering Diagrams
Enginuity is the first open benchmark dataset for VLMs on engineering diagrams, with evaluations showing models identify parts but produce low-fidelity descriptions and struggle with factual reasoning.
-
An Amortized Efficiency Threshold for Comparing Neural and Heuristic Solvers in Combinatorial Optimization
Introduces the Amortized Efficiency Threshold (AET) to identify the deployment volume at which neural combinatorial optimization solvers achieve lower total energy use than heuristic baselines after accounting for training costs.
-
Hidden Secrets in the arXiv: Discovering, Analyzing, and Preventing Unintentional Information Disclosure in Source Files of Scientific Preprints
Nearly every arXiv submission leaks hidden sensitive information through its source files, existing cleaners fail, and ALC-NG provides a more reliable fix.
-
SAM 3: Segment Anything with Concepts
SAM 3 introduces promptable concept segmentation that doubles accuracy of prior systems on images and videos while improving standard SAM segmentation performance.
-
Segment Anything
A promptable model trained on 1B masks achieves competitive zero-shot segmentation performance across tasks and is released publicly with its dataset.
-
OPT: Open Pre-trained Transformer Language Models
OPT releases open decoder-only transformers up to 175B parameters that match GPT-3 performance at one-seventh the carbon cost, along with code and training logs.
-
"Chi nas dal soch el sent de legn" -- Auditing Text Corpora for Lombard
Manual audit shows web-scraped Lombard corpora are largely noisy and biased toward Western varieties over Eastern ones.
-
BaltiVoice: A Speech Corpus and Fine-tuned Whisper ASR System for the Balti Language
BaltiVoice supplies the first public read-speech corpus and fine-tuned Whisper ASR for Balti, reaching 26.74% WER on a speaker-disjoint validation set.
-
EnergyLens: Predictive Energy-Aware Exploration for Multi-GPU LLM Inference Optimization
EnergyLens predicts multi-GPU LLM inference energy consumption with 9-13% MAPE and identifies configurations with up to 52x energy efficiency differences.
-
PEML: Parameter-efficient Multi-Task Learning with Optimized Continuous Prompts
PEML co-optimizes continuous prompts and low-rank adaptations to deliver up to 6.67% average accuracy gains over existing multi-task PEFT methods on GLUE, SuperGLUE, and other benchmarks.
-
UniSD: Towards a Unified Self-Distillation Framework for Large Language Models
UniSD unifies self-distillation components for autoregressive LLMs and its full integrated version improves base models by 5.4 points and baselines by 2.8 points across six benchmarks.
-
PlantTraitNet: An Uncertainty-Aware Multimodal Framework for Global-Scale Plant Trait Inference from Citizen Science Data
PlantTraitNet is an uncertainty-aware multimodal deep learning framework that infers four plant traits from citizen science images and produces global trait maps that outperform prior products when validated against independent survey data.
-
LLM Hypnosis: Exploiting User Feedback for Unauthorized Knowledge Injection to All Users
A single attacker can use strategic upvoting and downvoting on language model outputs to inject facts, security flaws, or fake news that persist in the model for all users after preference tuning.
-
ART: Automatic multi-step reasoning and tool-use for large language models
ART automatically generates multi-step reasoning programs with tool integration for LLMs, yielding substantial gains over few-shot and auto-CoT prompting on BigBench and MMLU while matching hand-crafted CoT on most tasks.
-
BLOOM: A 176B-Parameter Open-Access Multilingual Language Model
BLOOM is a 176B-parameter open-access multilingual language model trained on the ROOTS corpus that achieves competitive performance on benchmarks, with improved results after multitask prompted finetuning.
-
SEVA: Self-Evolving Verification Agent with Process Reward for Fact Attribution
SEVA trains a verification agent with a decomposed process reward to produce structured fact attributions, enabling a self-evolution loop that matches GPT-4o-mini F1 on ClearFacts while generating richer output.
-
Lost in the Evidence? Reproducing Document Position and Context Size Effects in RAG
Reproducibility study shows position and context size effects in RAG depend on topic sampling and retrieval quality, proposes calibration for stable trends, and releases code after finding discrepancies with prior industry work.
-
CroCo: Cross-Lingual Contrastive Preference Tuning on Self-Generations
CroCo applies English-reward-ranked self-generations for contrastive preference tuning that improves two LLMs on structured and open-ended tasks across 14 languages without language-specific annotations.
-
Mapping the Schedule x Bit-Width Boundary in Sub-100M Quantisation-Aware Training
Factorial experiments with over 1300 runs falsify the hypothesis that INT6 QAT needs a different LR schedule from higher precision and identify a 50M-parameter boundary for INT4 schedule sensitivity.
-
Multi-Dimensional Model Integrity and Responsibility Assessment Index and Scoring Framework
MIRAI is a unified index that combines five responsibility dimensions into one score for tabular models, demonstrating that predictive performance does not ensure high overall integrity.
-
Position: LLM Inference Should Be Evaluated as Energy-to-Token Production
LLM inference should be reframed and evaluated as energy-to-token production with a Token Production Function that accounts for power, cooling, and efficiency ceilings.
-
Agentic Insight Generation in VSM Simulations
A two-step agentic system for extracting insights from VSM simulations achieves up to 86% accuracy with top LLMs by using progressive data discovery and slim context.
-
Frugal Knowledge Graph Construction with Local LLMs: A Zero-Shot Pipeline, Self-Consistency and Wisdom of Artificial Crowds
A frugal zero-shot local-LLM pipeline extracts relations at F1 0.70 and reaches 0.55 EM on multi-hop QA through self-consistency, cross-model oracles, and confidence routing, while identifying an agreement paradox where strong consensus signals hallucination.
-
ChatGPT, is this real? The influence of generative AI on writing style in top-tier cybersecurity papers
Top-tier cybersecurity papers exhibit a post-2022 increase in AI marker words and higher lexical complexity, suggesting generative AI is influencing academic writing style.
-
MedicalRec: Medical recommender system for image classification without retraining
A transformer recommender system trained on a new benchmark of over 5,000 model performances from medical imaging papers achieves up to 75.5% HitRate@100.
-
From Cradle to Cloud: A Life Cycle Review of AI's Environmental Footprint
A review of AI sustainability studies finds inconsistent life cycle definitions and predominant reliance on coarse CO2e proxies, with limited coverage of water, materials, and multi-impact assessments.
-
Quantifying the Climate Risk of Generative AI: Region-Aware Carbon Accounting with G-TRACE and the AI Sustainability Pyramid
G-TRACE provides region-aware estimates of GenAI carbon emissions including 4309 MWh and 2068 tCO2 for a 2024-2025 image generation trend, paired with a seven-level AI Sustainability Pyramid for policy guidance.
-
Green Prompting: Characterizing Prompt-driven Energy Costs of LLM Inference
Empirical tests on three LLMs show prompt semantics and task keywords drive inference energy costs more than length, with varying patterns by task.
-
Will LLMs Scaling Hit the Wall? Breaking Barriers via Distributed Resources on Massive Edge Devices
Position paper claiming that distributed training across massive edge devices can overcome data depletion and centralized compute monopolies in LLM scaling.