Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Prompting Large Language Models for Training-Free Non-Intrusive Load Monitoring

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read Prompt-only LLMs can perform basic non-intrusive load monitoring without training, but their accuracy trails specialized deep-learning models; their strengths are cross-house transfer and readable explanations.

desk verdict Prompt-only LLMs for NILM are honestly benchmarked and fall short of DL baselines, but the cross-region generalization claim is not supported by the experiments as designed. read the letter →

arxiv 2505.06330 v3 pith:SQIMMC2M submitted 2025-05-09 cs.LG cs.AIeess.SP

classification cs.LGcs.AIeess.SP
keywords non-intrusiveloadmonitoringenergydisaggregationlargelanguagemodelspromptengineeringin-contextlearningzero-shottransferexplainabilityappliancestatedetection
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper asks whether a general-purpose large language model, fed only a text prompt and given no training on power data, can do non-intrusive load monitoring: inferring which appliances are on from a home's aggregate electricity draw. The answer it defends is a boundary claim rather than a breakthrough in accuracy. With a carefully assembled prompt—appliance statistics, one example, and the previous window's predicted states—the best LLM reaches F1 scores of about 0.44 on an unseen US house and 0.61 on a UK house, well below the 0.73–0.87 scores of CNN baselines. The model's distinctive value appears in what trained models lack: it transfers to a new house and a new regional appliance set by rewriting the prompt's knowledge block, and it emits a human-readable reason for every ON/OFF decision. The paper concludes that for now LLMs should complement, not replace, accurate but opaque disaggregation models.

What carries the argument

The carrying mechanism is the prompt itself, specifically the knowledge-injection and context blocks of the framework. Appliance features estimated by statistics from a few training houses are written into the prompt, and the state predictions from the previous window are appended as context; the LLM is instructed to output JSON state sequences. Ablations show that adding context delivers the largest single performance boost and that power range is the most decisive appliance feature: removing it drops the F1 score from 0.5376 to 0.3008, while removing pattern description only lowers it to 0.5725. The framework's sensitivity analyses establish that window length and context length must be tuned to model capacity—longer windows help larger LLMs but cause smaller ones to violate the required output format.

What would settle it

Run the same framework with knowledge extracted from REDD houses 2–3 but test on an unseen UK-DALE house (house 1 or 5) rather than house 2; if overall F1 collapses toward chance, the cross-region generalization claim fails. As a second check, replace the injected power ranges with deliberately wrong values and confirm the F1 falls back to the no-knowledge level of about 0.30, which would show the prompt, not some latent model knowledge, is carrying the capability.

Watch

Extended reading notes

Core claim

The paper's central discovery is a capability boundary for prompt-only LLMs in NILM. Using the proposed framework, a prompt composed of a role/task description, statistical appliance knowledge (standby power, power range, typical duration, usage pattern), a one-shot example, and the previous window's predicted states turns a general LLM into a zero-shot appliance-state detector. On REDD house 1 and UK-DALE house 2 the optimized prompt yields overall F1 scores of 0.4397 and 0.6120 for the stronger LLM, versus 0.7289 and 0.8721 for the Seq2Point CNN baseline; power estimation is weaker still, with signal aggregate error of about 0.425 versus 0.170 for the baseline. Appliances with sharp, high-power signatures or strong periodicity are detected most reliably, while multi-stage devices such as dishwashers are frequently masked by other loads. The same prompt, with only the knowledge block rewritten, transfers to the unseen US house and to the UK appliance set, and the model supplies natural-language justifications that can be checked against the aggregate signal. The paper reads these results as defining the boundary: prompt-only LLMs are not accuracy-competitive, but they offer training-free deployment, easy adaptation, and explainability.

Load-bearing premise

The claim that the approach generalizes rests on the assumption that appliance characteristics measured in a few known houses describe the same appliance types in unseen houses well enough for a text-only prompt to recognize them.

Editorial extensions

If this is right

  • A zero-training NILM deployment is possible: adapting to a new house or a different regional appliance set means editing the prompt's knowledge block, not retraining a model.
  • LLM-based disaggregation is usable where transparency matters more than peak accuracy, since every ON/OFF decision comes with a textual rationale that can be compared with the aggregate power trace.
  • The accuracy gap quantifies what prompt-only knowledge can capture: distinctive signatures (peak power, periodic cycling) transfer, while multi-stage signatures like a dishwasher's are lost among other loads.
  • Practitioners should pair window length and context length with model capacity; beyond a window of about 30 samples, smaller models' output format stability degrades sharply.
  • Power-value estimation is not yet practical by prompting alone, so near-term applications should treat appliance state detection as the deliverable.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A strict cross-region test remains open: the UK knowledge and the UK test set both come from house 2, so applying REDD-derived knowledge to an unseen UK house would be the decisive experiment for the regional generalization claim.
  • Because power range alone carries most of the gain, automatically mining sharper appliance signatures (e.g., per-cycle envelopes or duration distributions) from unlabeled aggregate data could cheaply raise zero-shot accuracy.
  • The emitted rationales could be used as machine-checkable evidence rather than decoration: a threshold rule can verify each explanation against the aggregate signal and flag disagreements, turning explainability into a quality-control loop.
  • The context block acts as a hand-written state smoother; an agent that maintains a longer internal belief state across windows might close part of the gap with trained models without fine-tuning.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper introduces LLM4NILM, a prompt-based framework that uses general-purpose large language models (LLMs) for non-intrusive load monitoring (NILM) without any model training. The framework injects appliance knowledge (power range, standby power, duration, usage pattern), one-shot examples, timestamps, and contextual information from previous windows into a prompt. The authors evaluate state-detection performance on REDD and UK-DALE, compare against Seq2Seq and Seq2Point baselines, and report that LLMs lag specialized DL models in F1 score but claim strong zero-shot generalization across houses and regions, as well as human-readable explainability. The paper also includes ablations of prompt components, knowledge-injection factors, window size, context length, and model capacity, concluding that LLMs are not ready to replace DL methods but may complement them in adaptable, transparent systems.

Significance. If the claims hold, this is a useful capability-boundary study: it is among the first systematic evaluations of prompt-only LLMs for NILM, and it honestly documents a clear accuracy gap while exploring generalization and explainability. The ablations are informative, particularly the finding that power range is the most influential knowledge component and that context length has a non-monotonic effect. The headline negative result (LLMs lag specialized models) is consistently supported by Table 3 across both datasets. The main advertised positive claim, however, is cross-region generalization, and that claim is currently under-supported by the experimental design.

major comments (4)
  1. [Section 6, Section 6.2, Section 3.1] The cross-region generalization claim (RQ2, abstract, Section 6.2) is not actually tested. Section 6 states that for UK-DALE, "we use data from house 2, splitting it into training (70%), validation (10%), and testing (20%)," while Section 3.1 says appliance knowledge is "extracted from historical data using statistical methods." Thus the UK-DALE test set is a later time slice of the same household used to derive the prompt knowledge, not an unseen house in a new region. The REDD experiment (knowledge from houses 2/3, test on unseen house 1) supports cross-house generalization, but no experiment evaluates an unseen UK household. The claim of "generalization ... across different houses and even regions" therefore rests on a single in-house temporal split and should be revised, or the experiment should be rerun on a held-out UK-DALE house.
  2. [Section 6, Table 3] The LLM results are stochastic API outputs, but each reported F1 score appears to come from a single run with no variance, repetition, temperature setting, or seed information. Given the large gaps between models (e.g., GPT-4.1-mini vs. DeepSeek-V3-0324 on REDD), single-run point estimates are insufficient to establish stable model rankings or reliable performance levels. The paper should report mean and standard deviation over multiple calls, or at least specify the sampling parameters and justify that the reported values are representative.
  3. [Section 6, Section 3.1] It is unclear whether the UK-DALE appliance knowledge (standby power, power range, duration, usage pattern) is computed from the training split only or from the full house-2 dataset. If the knowledge statistics are computed before the temporal split, information from the test period could leak into the prompts. The manuscript should state explicitly which portion of house 2 is used for knowledge extraction and, if necessary, recompute the statistics on the training split.
  4. [Section 6.3, Appendix B] The explainability evaluation is purely qualitative. The paper states that explanations are "verifiable" and compares them against ground truth and the aggregate signal, but no systematic protocol, metric, or inter-annotator assessment is provided. As explainability is one of the three central research questions, the RQ3 claim would be stronger with a defined evaluation procedure, even a simple correctness-rate measure over a sample of explanations.
minor comments (5)
  1. [Figure 6a] The caption states that the dishwasher is omitted from the figure, yet the overall F1 score includes it; this makes the overall trend hard to interpret because the composition of the average changes with appliance coverage. A note explaining how the overall score is computed in the figure would help.
  2. [Table 2] The rows of Table 2 are not labeled, so the reader must infer which knowledge combinations correspond to each row. Adding row labels or checkmarks for Power, Duration, and Pattern would improve readability.
  3. [Section 4.3] The Output Normalizer uses forward padding or truncation for misaligned outputs. Since the LLM is penalized for length errors, the choice of padding direction could affect appliance-state boundaries; a brief discussion of this effect would be useful.
  4. [Appendix A] The prompt template is helpful but omits the concrete appliance knowledge values used for REDD and UK-DALE (e.g., the actual power ranges and durations). Providing the exact knowledge entries in an appendix or supplementary file would make the study reproducible.
  5. [Section 5.1] The phrase "unseen REDD test set" in the caption of Figure 4 is slightly misleading because the knowledge is derived from houses 2 and 3, so the house is unseen but the appliance types and their statistical features are known. The wording could be clarified.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: prompt-only NILM evaluation is empirical and self-contained against external benchmarks.

full rationale

The paper's central claims are empirical comparisons, not derivations that reduce to their inputs. Appliance knowledge is 'extracted from historical data using statistical methods' (Section 3.1) from REDD houses 2/3 and the UK-DALE house 2 training split, then injected into prompts and evaluated on the REDD house 1 test set (an unseen house) and the UK-DALE house 2 test split. That is a normal transfer/evaluation protocol: the knowledge is fit to a training source and tested on held-out data; the reported F1 scores are not algebraically forced by the knowledge values. Prompt-component ablations and window/context sensitivity analyses are similarly empirical. The paper's self-citations (e.g., [11], [27], [29]) appear in background or formatting contexts and are not load-bearing; no uniqueness theorem or ansatz is imported from the authors' prior work. The only notable weakness is that the claimed 'cross-region' generalization for UK-DALE is tested on a temporal split of the same house (house 2) used to derive the knowledge, so the evidence does not fully support generalization to a new UK household. That is a limitation in the support for a generalization claim, not a circular derivation; accordingly, no circular step is identified.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claims rest on a handful of data-derived prompt statistics and threshold labels rather than a derivation. The key ledger items are the ground-truth thresholds and the knowledge features; both are legitimate engineering choices but they are fitted inputs, and the paper's 'training-free' framing masks that the prompts are constructed from labeled historical data (statistics extracted from training houses).

free parameters (4)
  • Ground-truth ON/OFF power thresholds = microwave 200W, fridge 50W, dishwasher 10W, washing machine 20W, kettle 2000W
    Fixed thresholds used to derive appliance state labels from appliance power data; every F1 score in the paper compares against these labels, so the thresholds define the target of both LLM and DL models.
  • Prompt knowledge features (power range, standby power, ON duration, usage pattern, cycle duration) = per-appliance statistics estimated from REDD houses 2/3 and UK-DALE house 2 training data
    These data-derived statistics are embedded in the prompts and are the mechanism claimed to enable zero-shot transfer; they are fitted to training-house data, not to the test houses.
  • Window size and context length = window 100, context 10 for large-scale evaluation
    Selected through ablation studies on a one-day REDD slice (Table 1, Figure 6); note context 30 gave the best F1 (0.6193) in Section 5.4 but the large-scale runs used context 10.
  • Output normalizer rules = forward padding or truncation for misaligned outputs; malformed outputs discarded
    Ad hoc post-processing that modifies the scored output for misaligned LLM responses; its effect on metrics is not quantified.
assumptions (4)
  • domain assumption Fixed power thresholds faithfully represent true appliance ON/OFF states (Section 4.1)
    The ground truth itself is threshold-derived; if an appliance runs below threshold (e.g., dishwasher low-power stages), the label is OFF by definition, and both LLM and DL scores inherit this definition.
  • domain assumption Appliance knowledge features estimated from the training houses transfer to unseen houses (Section 3.1, Section 6.2)
    This is the load-bearing premise for the RQ2 zero-shot generalization claim, supported only by the REDD cross-house result; the UK-DALE test does not use an unseen house.
  • domain assumption JSON-mode API outputs, after normalization, are a fair representation of LLM performance (Section 4.3)
    Malformed outputs are unprocessable and misaligned outputs are padded/truncated; the incidence of each error type is not reported, so the reported F1 may not reflect raw model behavior.
  • standard math The NILM disaggregation problem is ill-posed and requires priors (Section 1, citing Hart 1992)
    Background justification for why prompting with appliance priors is a sensible strategy; not contentious.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Prompting Large Language Models for Training-Free Non-Intrusive Load Monitoring." pith.science (2026). https://pith.science/paper/SQIMMC2M

@misc{pith2026250506330,
  author       = {Pith},
  title        = {Pith review of: Prompting Large Language Models for Training-Free Non-Intrusive Load Monitoring},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SQIMMC2M}},
  note         = {Machine review of arXiv:2505.06330}
}
read the original abstract

Non-intrusive load monitoring (NILM) aims to disaggregate total electricity consumption into individual appliance usage, thus enabling more effective energy management. While deep learning has advanced NILM, it remains limited by its dependence on labeled data, restricted generalization, and lack of explainability. This paper introduces the first prompt-based NILM framework that leverages large language models (LLMs) with in-context learning. We design and evaluate prompt strategies that integrate appliance features, contextual information, and representative time-series examples through extensive case studies. Extensive experiments on the REDD and UK-DALE datasets show that LLMs guided solely by prompts deliver only basic NILM capabilities, with performance that lags behind traditional deep-learning models in complex scenarios. However, the experiments also demonstrate strong generalization across different houses and even regions by simply adapting the injected appliance features. It also provides clear, human-readable explanations for the inferred appliance states. Our findings define the capability boundaries of using prompt-only LLMs for NILM tasks. Their strengths in generalization and explainability present a promising new direction for the field.

Figures

Figures reproduced from arXiv: 2505.06330 by the authors.

Figure 1
Figure 1. Limitations of traditional DL-based NILM methods. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. An overview of our proposed LLM4NILM framework. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. The NILM capability of mainstream general LLMs. [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Illustrative predictions on a one-day segment of the REDD house 1 test set. The results are generated by DeepSeek-V3- [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Illustrative predictions on a one-day segment of the UK-DALE house 2 test set. The results are generated by DeepSeek [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Sensitivity analysis results for window size (left) and context length (right). [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Output Formatting Error Rates Across Different [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. UST-SSM: Unified Spatio-Temporal State Space Models for Point Cloud Video Modeling

    cs.CV 2025-08 unverdicted novelty 6.0 of 10

    A point cloud video action recognition model is claimed (UST-SSM), but the submitted full text is an unrelated energy disaggregation paper, so the central claim is unverifiable.

Reference graph

Works this paper leans on

38 extracted references · 28 canonical work pages · cited by 1 Pith paper

  1. [1]

    Nonintrusive appliance load monitoring

    George W Hart. Nonintrusive appliance load monitoring. Proceedings of the IEEE, 80(12):1870–1891, 1992

  2. [2]

    Non-intrusive load monitoring approaches for disaggregated energy sensing: A survey

    Ahmed Zoha, Alexander Gluhak, Muhammad Ali Imran, and Sutharshan Ra- jasegarar. Non-intrusive load monitoring approaches for disaggregated energy sensing: A survey. Sensors, 12(12):16838–16866, 2012

  3. [3]

    Redd: A public data set for energy disag- gregation research

    J Zico Kolter and Matthew J Johnson. Redd: A public data set for energy disag- gregation research. In Workshop on data mining applications in sustainability (SIGKDD), San Diego, CA , volume 25, pages 59–62. Citeseer, 2011

  4. [4]

    Nonintrusive load monitoring (nilm) performance evaluation: A unified approach for accuracy reporting

    Stephen Makonin and Fred Popowich. Nonintrusive load monitoring (nilm) performance evaluation: A unified approach for accuracy reporting. Energy Efficiency, 8:809–814, 2015

  5. [5]

    Deep learning

    Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. Deep learning. nature, 521(7553):436–444, 2015

  6. [6]

    Neural nilm: Deep neural networks applied to energy disaggregation

    Jack Kelly and William Knottenbelt. Neural nilm: Deep neural networks applied to energy disaggregation. In Proceedings of the 2nd ACM International Conference on Embedded Systems for Energy-Efficient Built Environments , pages 55–64, 2015

  7. [7]

    A regression approach to single-channel speech separation via high-resolution deep neural networks

    Jun Du, Yanhui Tu, Li-Rong Dai, and Chin-Hui Lee. A regression approach to single-channel speech separation via high-resolution deep neural networks. IEEE/ACM Transactions on Audio, Speech, and Language Processing , 24(8):1424– 1437, 2016

  8. [8]

    Sequence-to-point learning with neural networks for non-intrusive load monitoring

    Chaoyun Zhang, Mingjun Zhong, Zongzuo Wang, Nigel Goddard, and Charles Sutton. Sequence-to-point learning with neural networks for non-intrusive load monitoring. In Proceedings of the 32nd AAAI Conference on Artificial Intelligence , 2018

Show all 38 references
  1. [9]

    Bert4nilm: A bidirectional transformer model for non-intrusive load monitoring

    Zhenrui Yue, Camilo Requena Witzig, Daniel Jorde, and Hans-Arno Jacobsen. Bert4nilm: A bidirectional transformer model for non-intrusive load monitoring. In Proceedings of the 5th International Workshop on Non-Intrusive Load Monitoring, pages 89–93, 2020

  2. [10]

    Auglpn-nilm: Augmented light- weight parallel network for nilm embedding attention module over sequence to point

    Wuqing Yu, Linfeng Yang, and Xiangyu Liu. Auglpn-nilm: Augmented light- weight parallel network for nilm embedding attention module over sequence to point. Sustainable Energy, Grids and Networks , 38:101378, 2024

  3. [11]

    Towards real-world deployment of nilm systems: Challenges and practices

    Junyu Xue, Yu Zhang, Xudong Wang, Yi Wang, and Guoming Tang. Towards real-world deployment of nilm systems: Challenges and practices. In 2024 IEEE International Conference on Sustainable Computing and Communications (Sus- tainCom), pages 16–23, 2024

  4. [12]

    A survey of large language models

    Wayne Xin Zhao, Kun Zhou, Junyi Li, Tianyi Tang, Xiaolei Wang, Yupeng Hou, Yingqian Min, Beichen Zhang, Junjie Zhang, Zican Dong, et al. A survey of large language models. arXiv preprint arXiv:2303.18223, 2023

  5. [13]

    A survey on in-context learning

    Qingxiu Dong, Lei Li, Damai Dai, Ce Zheng, Jingyuan Ma, Rui Li, Heming Xia, Jingjing Xu, Zhiyong Wu, Baobao Chang, et al. A survey on in-context learning. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 1107–1128, 2024. Junyu Xu...

  6. [14]

    A survey of zero-shot learning: Settings, methods, and applications

    Wei Wang, Vincent W Zheng, Han Yu, and Chunyan Miao. A survey of zero-shot learning: Settings, methods, and applications. ACM Transactions on Intelligent Systems and Technology (TIST), 10(2):1–37, 2019

  7. [15]

    Generalizing from a few examples: A survey on few-shot learning

    Yaqing Wang, Quanming Yao, James T Kwok, and Lionel M Ni. Generalizing from a few examples: A survey on few-shot learning. ACM computing surveys (csur), 53(3):1–34, 2020

  8. [16]

    The uk-dale dataset, domestic appliance- level electricity demand and whole-house demand from five uk homes

    Jack Kelly and William Knottenbelt. The uk-dale dataset, domestic appliance- level electricity demand and whole-house demand from five uk homes. Scientific data, 2(1):1–14, 2015

  9. [17]

    Approximate inference in additive factorial hmms with application to energy disaggregation

    J Zico Kolter and Tommi Jaakkola. Approximate inference in additive factorial hmms with application to energy disaggregation. In Artificial intelligence and statistics, pages 1472–1482. PMLR, 2012

  10. [18]

    Gradient-based learning applied to document recognition

    Yann LeCun, Léon Bottou, Yoshua Bengio, and Patrick Haffner. Gradient-based learning applied to document recognition. Proceedings of the IEEE , 86(11):2278– 2324, 1998

  11. [19]

    Long short-term memory

    Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory. Neural Computation, 9(8):1735–1780, 1997

  12. [20]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. In Advances in Neural Information Processing Systems (NeurIPS) , volume 30, 2017

  13. [21]

    A review of current methods and challenges of advanced deep learning-based non-intrusive load monitoring (nilm) in residential context

    Hasan Rafiq, Prajowal Manandhar, Edwin Rodriguez-Ubinas, Omer Ahmed Qureshi, and Themis Palpanas. A review of current methods and challenges of advanced deep learning-based non-intrusive load monitoring (nilm) in residential context. Energy and Buildings, page 113890, 2024

  14. [22]

    Using explain- ability tools to inform nilm algorithm performance: a decision tree approach

    Rachel Stephen Mollel, Lina Stankovic, and Vladimir Stankovic. Using explain- ability tools to inform nilm algorithm performance: a decision tree approach. In Proceedings of the 9th ACM International Conference on Systems for Energy- Efficient Buildings, Cities, and Transporta...

  15. [23]

    Toward transparent load disaggregation—a framework for quantitative evaluation of explainability using explainable ai

    Djordje Batic, Vladimir Stankovic, and Lina Stankovic. Toward transparent load disaggregation—a framework for quantitative evaluation of explainability using explainable ai. IEEE Transactions on Consumer Electronics , 70(1):4345–4356, 2023

  16. [24]

    Toward explainable nilm: Real-time event-based nilm framework for high-frequency data

    Grigorii Gerasimov, Ilia Kamyshev, Sahar Moghimian Hoosh, Elena Gryazina, and Henni Ouerdane. Toward explainable nilm: Real-time event-based nilm framework for high-frequency data. arXiv preprint arXiv:2501.16841, 2025

  17. [25]

    Transfer learning for multi-objective non-intrusive load monitoring in smart building

    Dandan Li, Jiangfeng Li, Xin Zeng, Vladimir Stankovic, Lina Stankovic, Changjiang Xiao, and Qingjiang Shi. Transfer learning for multi-objective non-intrusive load monitoring in smart building. Applied Energy, 329:120223, 2023

  18. [26]

    Adaptive fusion feature transfer learning method for nilm

    Keqin Li, Jian Feng, Juan Zhang, and Qi Xiao. Adaptive fusion feature transfer learning method for nilm. IEEE Transactions on Instrumentation and Measurement, 72:1–12, 2023

  19. [27]

    Evsense: A robust and scalable approach to non-intrusive ev charging detection

    Xudong Wang, Guoming Tang, Yi Wang, Srinivasan Keshav, and Yu Zhang. Evsense: A robust and scalable approach to non-intrusive ev charging detection. In Proceedings of the Thirteenth ACM International Conference on Future Energy Systems, pages 307–319, 2022

  20. [28]

    A semi-supervised load identification method with class incremental learning

    Leixin Qiu, Tao Yu, and Chaofan Lan. A semi-supervised load identification method with class incremental learning. Engineering Applications of Artificial Intelligence, 131:107768, 2024

  21. [29]

    Fednilm: Applying federated learning to nilm applications at the edge

    Yu Zhang, Guoming Tang, Qianyi Huang, Yi Wang, Kui Wu, Keping Yu, and Xun Shao. Fednilm: Applying federated learning to nilm applications at the edge. IEEE Transactions on Green Communications and Networking , 7(2):857–868, 2022

  22. [30]

    Federated learning-based non-intrusive load monitoring adaptive to real-world heterogeneities

    Qingquan Luo, Chaofan Lan, Tao Yu, Minhang Liang, Wencong Xiao, and Zhen- ning Pan. Federated learning-based non-intrusive load monitoring adaptive to real-world heterogeneities. Scientific Reports, 15(1):18223, 2025

  23. [31]

    Promptcast: A new prompt-based learning paradigm for time series forecasting

    Hao Xue and Flora D Salim. Promptcast: A new prompt-based learning paradigm for time series forecasting. IEEE Transactions on Knowledge and Data Engineering, 36(11):6851–6864, 2023

  24. [32]

    Large language models are zero-shot time series forecasters

    Nate Gruver, Marc Finzi, Shikai Qiu, and Andrew G Wilson. Large language models are zero-shot time series forecasters. Advances in Neural Information Processing Systems, 36:19622–19635, 2023

  25. [33]

    Large language models for time series: A survey

    Xiyuan Zhang, Ranak Roy Chowdhury, Rajesh K Gupta, and Jingbo Shang. Large language models for time series: A survey. arXiv preprint arXiv:2402.01801, 2024

  26. [34]

    Chain-of-thought prompting elicits reasoning in large language models

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems , 35:24824–24837, 2022

  27. [35]

    Unisep: Universal target audio separation with language models at scale

    Yuanyuan Wang, Hangting Chen, Dongchao Yang, Weiqin Li, Dan Luo, Guangzhi Li, Shan Yang, Zhiyong Wu, Helen Meng, and Xixin Wu. Unisep: Universal target audio separation with language models at scale. arXiv preprint arXiv:2503.23762, 2025

  28. [36]

    Lisa: Reasoning segmentation via large language model

    Xin Lai, Zhuotao Tian, Yukang Chen, Yanwei Li, Yuhui Yuan, Shu Liu, and Jiaya Jia. Lisa: Reasoning segmentation via large language model. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 9579–9589, 2024

  29. [37]

    Transfer learning for non-intrusive load monitoring

    Michele D’Incecco, Stefano Squartini, and Mingjun Zhong. Transfer learning for non-intrusive load monitoring. IEEE Transactions on Smart Grid, 11(2):1419–1429, 2019

  30. [38]

    APPLIANCE_NAMES[0]_status

    Rachel Stephen Mollel, Lina Stankovic, and Vladimir Stankovic. Explainability- informed feature selection and performance prediction for nonintrusive load monitoring. Sensors, 23(10):4845, 2023. A PROMPTS Role. You are an expert system specializing in Non-intrusive Load Monito...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.