Pith. sign in

REVIEW 4 major objections 5 minor 61 references

A Soft Sensor Method with Uncertainty-Awareness and Self-Explanation Based on Large Language Models Enhanced by Domain Knowledge Retrieval

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper claims that a two-stage LLM pipeline using in-context learning can replace supervised training for industrial soft sensing, achieving state-of-the-art prediction errors on two process datasets while generating explanations and…

desk verdict A genuinely novel LLM-based soft sensor framework with a promising no-training pipeline, but the RAC retrieval protocol and the temperature-0 uncertainty setup both need fixing before the SOTA claim is credible. read the letter →

arxiv 2501.03295 v2 pith:V2VUBKCL submitted 2025-01-06 cs.LG cs.AIeess.SP

Shuo Tong , Han Liu , Runyuan Guo , Wenqing Wang , Xueqiong Tian , Lingyun Wei , Lin Zhang , Huayong Wu
show 2 more authors
Ding Liu Youmin Zhang
This is my paper · ORCID
classification cs.LGcs.AIeess.SP
keywords softsensorin-contextlearninglargelanguagemodelsretrieval-augmentedgenerationuncertaintyquantificationself-explanationindustrialprocessmonitoringfew-shotprediction
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 claims that soft sensors, models that estimate hard-to-measure quality variables from easy-to-measure process variables, can be built without any supervised training by reformulating the task as text-based in-context learning. The proposed LLM-FUESS pipeline has two stages: LLM-ZAVS selects auxiliary variables zero-shot using domain knowledge retrieved from a vector store, and LLM-UFSS turns a few labeled samples into text demonstrations in a prompt to predict the target variable. On a penicillin fermentation simulation and a polypropylene production dataset, the paper reports that this prompt-only method achieves lower mean absolute error and root mean square error than random forest, MLP, k-NN, and PCR baselines, handles up to 50% missing values, and supplies confidence intervals, confidence scores, and human-readable explanations. If correct, the result means an industrial soft sensor could be deployed by filling in a prompt template, with no model training, no normalization, and no imputation.

What carries the argument

The load-bearing mechanism is in-context learning applied to text-formatted structured data: each training sample becomes a line such as 'Aeration rate: 6.5, Sugar feed rate: 0.1, ..., Penicillin concentration: 15.2', and the LLM predicts the primary variable for a test line from a handful of such demonstrations in the prompt. Supporting machinery includes the Industrial Knowledge Vector Store and the Industrial Process Data Vector Store, both enabled by retrieval-augmented generation, the AVS-PT and SS-PT fill-in-the-blank prompt templates with role, chain-of-thought, and emotional-stimulus prompts, and two uncertainty channels: repeated sampling to build confidence intervals and a prompted confidence score. The paper's Average Selection Consistency Score (ASCS) measures how consistently LLM-ZAVS picks the same variables across repeated runs.

What would settle it

Repeat the LLM-UFSS-RAC protocol with an explicitly held-out test set: build the IPDVS only from training samples, forbid any retrieved neighbor from being within a short time window of the test sample, and compare the resulting MAE with random forest trained on the same retrieved samples. If the gap closes or reverses, the reported advantage came from target leakage rather than in-context reasoning.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that large language models can act as soft sensors through in-context learning alone, matching or exceeding conventional trained regressors without parameter updates. The authors introduce and evaluate LLM-FUESS, a two-stage framework in which LLM-ZAVS performs zero-shot auxiliary variable selection with retrieval-augmented generation from an industrial knowledge vector store, and LLM-UFSS performs few-shot prediction by formatting structured process data as text and inserting the selected variables and retrieved similar samples into a soft-sensor prompt template. In the fixed-context variant, LLM-UFSS-FSC achieves the lowest MAE and RMSE on both datasets among all compared methods; in the retrieval-augmented variant, LLM-UFSS-RAC lowers MAE further by retrieving similar samples from an industrial process data vector store, while also producing confidence scores and confidence intervals that track prediction error. The paper also reports that prompting the model to explain its reasoning and rate its confidence improves accuracy, and that replacing missing values with 'N/A' preserves useful predictions at missing rates up to 50%.

Load-bearing premise

The retrieval-augmented results assume the test samples are not part of, or near-duplicates of, the 200 samples stored in the Industrial Process Data Vector Store, because the prompt returns the nearest stored samples' target values.

Editorial extensions

If this is right

  • If the central claim holds, soft sensor models can be developed for a new process by editing a few placeholders in a prompt template, removing the need for dataset-specific training pipelines and machine-learning expertise.
  • The absence of parameter updates means the usual failure modes of trained soft sensors, such as overfitting, underfitting, gradient instability, and sensitivity to initialization, are bypassed by construction.
  • Missing values can be passed through as 'N/A' rather than imputed, so data preprocessing is simplified while retaining predictive accuracy at missing rates up to 50%.
  • Operators would receive not only a prediction but a confidence interval, a confidence score, and a natural-language explanation, giving them information about when to trust the output.
  • Choice of underlying large language model matters: the paper's ablation shows a large performance gap between weaker and stronger base models, so the benefit is tied to the capability of the base model.

Reading between the lines

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

  • The strongest caution a reader should carry is about the retrieval-augmented experiments: the paper builds the IPDVS from the same 200 samples used in the fixed-context variant and does not state that test samples are excluded, so nearest-neighbor retrieval could return the test sample itself or a temporally adjacent point whose target value is nearly identical, which would inflate the reported ga
  • A fair comparison against trained baselines would require a strict split: build the data vector store only from training samples, ensure retrieved neighbors are not within a few time steps of the test point, and re-run both variants.
  • If the leakage concern is resolved and the gains persist, the same template could be tested for online streaming soft sensing, where the retrieved context naturally contains the most recent samples rather than randomly drawn historical ones.
  • The zero-shot variable selector's competitive results suggest an interesting direction: using LLMs as a prior feature-selection step before any numerical model, which could reduce dimensionality and preprocessing costs even in conventional pipelines.
Share X Bluesky LinkedIn Reddit HN

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 proposes LLM-FUESS, a two-stage soft-sensor framework built on large language models with in-context learning instead of supervised training. Stage one, LLM-ZAVS, uses retrieval-augmented generation from an industrial knowledge vector store to select auxiliary variables zero-shot. Stage two, LLM-UFSS, formats numerical samples as text prompts, uses either random few-shot contexts (FSC) or retrieval-augmented contexts from an industrial process data vector store (RAC), and produces point predictions, confidence intervals, confidence scores, and textual self-explanations. Experiments on the IndPensim penicillin fermentation simulator and a polypropylene production dataset compare the method against PCR, k-NN, MLP, and random forest regression, and report lower MAE/RMSE in several configurations, plus robustness to missing values and ablations of prompt components.

Significance. If the reported results hold, the paper would be a noteworthy demonstration that a no-training, prompt-based LLM pipeline can compete with trained regression models for industrial quality prediction, with added interpretability and uncertainty outputs. The paper has genuine strengths: it evaluates against external numerical baselines rather than only self-comparisons, tests multiple LLMs in an ablation, uses raw unnormalized inputs, and provides detailed prompt templates and JSON output formats that support reproducibility. However, the significance is currently tempered by a plausible data-leakage path in the retrieval-augmented variant, by the absence of any variance or significance analysis for the headline comparisons, and by the self-referential nature of the confidence-score validation.

major comments (4)
  1. [Experimental settings, paragraph 3 and Analysis of LLM-UFSS-RAC] The RAC results in Table 6 and Figures 8-10 rest on a retrieval protocol that may leak target information into the context. The paper states that 'the 200 samples from LLM-UFSS-FSC were used to construct the IPDVS' but never states that the test samples are excluded from this pool. Both datasets are time-series-like, and in IndPensim the penicillin concentration evolves smoothly and monotonically after 24 h, so nearest-neighbor retrieval in feature space will tend to return temporally adjacent samples whose target values are very close to the query target. The LLM could then produce predictions by copying or interpolating the retrieved labels, inflating the reported 10-shot MAE reductions of 45.93% and 16.23% over RFR. This is a load-bearing issue for the 'state-of-the-art predictive performance' claim; the authors should either demonstrate that test samples were excluded from the IPDVS or rerun the RAC experiments with a retrieval pool that provably excludes all test samples and their near-duplicates.
  2. [Tables 5-7 and Figure 16] All headline comparisons are reported as single point estimates, with no standard deviations, confidence intervals, or significance tests for either the proposed method or the baselines. Because LLM outputs are stochastic even at temperature 0 and the baselines involve hyperparameter selection, statements such as 'achieved the lowest MAE and RMSE across both datasets' are not statistically supported. The authors should report repeated-run means with variability and, where appropriate, paired significance tests across the same test samples.
  3. [Confidence score output analysis, Figure 11 and Table 8] The confidence score is generated by the same LLM that produces the prediction it is meant to assess, so the observed alignment between high confidence and low error in Figure 11b is partly self-referential. Similarly, the higher confidence scores for LLM-UFSS-RAC in Table 8 may reflect the similarity of retrieved contexts rather than calibrated uncertainty. The trustworthiness claim needs an external calibration check, for example comparing confidence scores against empirical error rates or using a held-out calibration set, rather than only showing that the model's self-reported confidence correlates with its own errors.
  4. [Table 3, IndPensim LR row for RFE] The RFE row for linear regression on IndPensim reports MAE 0.046, the best value in that column, but R² = 0.096, the worst value in that column and inconsistent with such a low MAE. This is likely a reporting error, but it directly affects the claim that LLM-ZAVS is competitive with numerical feature selection methods. The authors should verify the entries in Table 3 and correct the inconsistent value.
minor comments (5)
  1. [Experimental settings, implementation details] The paper sets temperature to 0 for consistency but then constructs 95% confidence intervals by repeated experiments; please clarify how output diversity is obtained under temperature 0 (for example API nondeterminism or sampling parameters other than temperature) and whether the same randomness affects the single-evaluation RAC results.
  2. [Figure 13b caption and Analysis of explanation results] The caption of Figure 13b describes a local explanation of 'oxygen uptake rate for melt flow rate', but oxygen uptake rate is an IndPensim variable and the surrounding text describes the variable's effect on penicillin concentration; the caption or the dataset association should be corrected.
  3. [Figure 6 and Experimental settings] The text says each set of 20 test samples in Figure 6 uses the same 20 training samples for contextualization, while the experimental settings describe 200 randomly chosen samples forming 10 contexts of 20 samples; the relationship between the figure subset and the full protocol should be stated explicitly.
  4. [Throughout] There are numerous typos that should be corrected, including 'feasubke' in the LLM-UFSS methods section, 'Performace' in Table 6, 'comparision' in Table 5, 'raitos' in Table 7, 'pediction' and 'dfferent' in Figure 16, and 'reults' in the Introduction.
  5. [Experimental settings, LLM-UFSS-RAC] The statement that each RAC test sample is evaluated only once means that per-sample confidence intervals are not available for RAC, limiting direct uncertainty comparisons between FSC and RAC; this limitation should be acknowledged where the uncertainty visualizations are discussed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the core predictive claims are checked against external baselines and the ICL pipeline involves no fitted parameters.

full rationale

The paper's main derivation chain is self-contained. LLM-ZAVS selects auxiliary variables by prompting a frozen LLM with retrieved domain knowledge; its utility is then measured by the downstream MAE and R2 of independently trained LR and SVR models (Table 3), not by a quantity defined in terms of the LLM's own output. LLM-UFSS-FSC predicts held-out test samples from fixed random context samples and is compared with PCR, KNN, MLP, and RFR (Table 5); no parameter of the LLM is fitted to the test set, so the comparison is not forced. LLM-UFSS-RAC retrieves context samples from the same 200-sample pool used for FSC, and all baselines are given the same retrieved context (Table 6); while the paper does not state a temporal split, making the RAC gains an external-validity concern, this is a data-protocol issue rather than an equation-level reduction of the prediction to its input. The confidence-score analysis is self-referential in that the LLM rates its own predictions, but the paper validates it by correlating the score with actual prediction errors, which is an empirical calibration check rather than a definitional equivalence. The authors' prior works are cited for background limitations (refs 23, 57, 58) and are not load-bearing for the core claim. I therefore find no circular step under the strict criteria.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central claim is empirical, so the ledger captures the method's hand-chosen settings and domain assumptions rather than derivation constants. No fitted physical parameters appear; the main fragility is that the protocol and self-referential validation carry several unverified assumptions. No new theoretical entities are postulated; IKVS and IPDVS are implementation artifacts rather than scientific entities.

free parameters (5)
  • context demonstration count k = 20 for FSC; 5 and 10 for RAC
    Chosen by hand; the paper shows performance varies with k, so the few-shot results depend on this choice.
  • feature selection ratio = 50% of total features
    LLM-ZAVS evaluation fixes selected variables at half the features; other proportions are not tested.
  • repetition count for uncertainty intervals = 10
    Each test sample is repeated 10 times to build 95% confidence intervals; interval quality depends on this arbitrary count.
  • temperature = 0
    Set to 0 for output consistency, which conflicts with sampling-based uncertainty intervals.
  • retrieval top-K = not specified
    The number of retrieved knowledge chunks and retrieved samples used in AVS-PT and SS-PT is not given, so RAC behavior is under-specified.
assumptions (5)
  • domain assumption Pretrained LLMs can perform accurate numeric regression on text-formatted tabular data via in-context learning.
    The entire LLM-UFSS pipeline rests on this; introduced in the Introduction and Methods without independent evidence beyond the two case studies.
  • domain assumption Similarity retrieval from IPDVS returns context samples whose target values help predict the test sample and generalize to unseen data.
    Invoked in the LLM-UFSS-RAC section; if retrieval leaks the test sample or temporally adjacent near neighbors, the reported gains are inflated.
  • domain assumption LLM self-reported confidence scores are meaningful indicators of predictive error.
    Used in the confidence score output analysis; validated only against the same model's errors, not against external calibration.
  • domain assumption Structured process data can be converted to natural-language text without losing information needed for prediction.
    Core to the prompt-based design; stated in the Methods but no ablation on formatting fidelity is provided.
  • domain assumption RAG-retrieved domain documents improve feature selection and reduce hallucination.
    Invoked in LLM-ZAVS; no quantitative ablation of IKVS retrieval quality is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Soft Sensor Method with Uncertainty-Awareness and Self-Explanation Based on Large Language Models Enhanced by Domain Knowledge Retrieval." pith.science (2026). https://pith.science/paper/V2VUBKCL

@misc{pith2026250103295,
  author       = {Pith},
  title        = {Pith review of: A Soft Sensor Method with Uncertainty-Awareness and Self-Explanation Based on Large Language Models Enhanced by Domain Knowledge Retrieval},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/V2VUBKCL}},
  note         = {Machine review of arXiv:2501.03295}
}
read the original abstract

Data-driven soft sensors are crucial in predicting key performance indicators in industrial systems. However, current methods predominantly rely on the supervised learning paradigms of parameter updating, which inherently faces challenges such as high development costs, poor robustness, training instability, and lack of interpretability. Recently, large language models (LLMs) have demonstrated significant potential across various domains, notably through In-Context Learning (ICL), which enables high-performance task execution with minimal input-label demonstrations and no prior training. This paper aims to replace supervised learning with the emerging ICL paradigm for soft sensor modeling to address existing challenges and explore new avenues for advancement. To achieve this, we propose a novel framework called the Few-shot Uncertainty-aware and self-Explaining Soft Sensor (LLM-FUESS), which includes the Zero-shot Auxiliary Variable Selector (LLM-ZAVS) and the Uncertainty-aware Few-shot Soft Sensor (LLM-UFSS). The LLM-ZAVS retrieves from the Industrial Knowledge Vector Storage to enhance LLMs' domain-specific knowledge, enabling zero-shot auxiliary variable selection. In the LLM-UFSS, we utilize text-based context demonstrations of structured data to prompt LLMs to execute ICL for predicting and propose a context sample retrieval augmentation strategy to improve performance. Additionally, we explored LLMs' AIGC and probabilistic characteristics to propose self-explanation and uncertainty quantification methods for constructing a trustworthy soft sensor. Extensive experiments demonstrate that our method achieved state-of-the-art predictive performance, strong robustness, and flexibility, effectively mitigates training instability found in traditional methods. To the best of our knowledge, this is the first work to establish soft sensor utilizing LLMs.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

61 extracted references · 59 canonical work pages

  1. [1]

    Causal variable selection for industrial process quality prediction via attention- based GRU network

    Yao L, Ge ZQ. Causal variable selection for industrial process quality prediction via attention- based GRU network. Eng Appl Artif Intell 118, 18 (2023)

    work page 2023

  2. [2]

    Machine learning for industrial sensing and control: A survey and practical perspective

    Lawrence NP, et al. Machine learning for industrial sensing and control: A survey and practical perspective. Control Eng Practice 145, 16 (2024)

    work page 2024

  3. [3]

    Dynamic transfer soft sensor for concept drift adaptation

    Zhang TM, Yan GW, Ren MF, Cheng L, Li R, Xie G. Dynamic transfer soft sensor for concept drift adaptation . J Process Control 123, 50-63 (2023)

    work page 2023

  4. [4]

    Soft Sensor Transferability: A Survey

    Curreri F, Patanè L, Xibilia MG. Soft Sensor Transferability: A Survey. Appl Sci- Basel 11, 18 (2021)

    work page 2021

  5. [5]

    Reliable Soft Sensors With an Inherent Process Graph Constraint

    Zhai RK, Zheng JH, Song ZH, Ge ZQ. Reliable Soft Sensors With an Inherent Process Graph Constraint. IEEE Trans Ind Inform 20, 8798-8806 (2024)

    work page 2024

  6. [6]

    Information Complementary Fusion Stacked Autoencoders for Soft Sensor Applications in Multimode Industrial Processes

    Zhang XM, He BC, Zhu HY, Song ZH. Information Complementary Fusion Stacked Autoencoders for Soft Sensor Applications in Multimode Industrial Processes. IEEE Trans Ind Inform 20, 106- 116 (2024)

    work page 2024

  7. [7]

    Physics-Enhanced Graph Neural Networks for Soft Sensing in Industrial Internet of Things

    Niresi KF, Bissig H, Baumann H, Fink O. Physics-Enhanced Graph Neural Networks for Soft Sensing in Industrial Internet of Things. IEEE Internet Things J 11, 34978-34990 (2024)

    work page 2024

  8. [8]

    Semi-supervised adaptive PLS soft-sensor with PCA-based drift correction method for online valuation of NOx emission in industrial water-tube boiler

    Hasnen SH, Shahid M, Zabiri H, Taqvi SAA. Semi-supervised adaptive PLS soft-sensor with PCA-based drift correction method for online valuation of NOx emission in industrial water-tube boiler. Process Saf Environ Protect 172, 787-801 (2023)

    work page 2023

Show all 61 references
  1. [9]

    Nonlinear feature extraction for soft sensor modeling based on weighted probabilistic PCA

    Yuan XF, Ye LJ, Bao L, Ge ZQ, Song ZH. Nonlinear feature extraction for soft sensor modeling based on weighted probabilistic PCA. Chemometrics Intell Lab Syst 147, 167-175 (2015)

    2015

  2. [10]

    Comparison of variable selection methods for PLS -based soft sensor modeling

    Wang ZX, He QP, Wang J. Comparison of variable selection methods for PLS -based soft sensor modeling. J Process Control 26, 56-72 (2015)

    2015

  3. [11]

    Developing variable moving window PLS models: Using case of NOx emission prediction of coal-fired power plants

    Li Z, Lee YS, Chen JH, Qian YW. Developing variable moving window PLS models: Using case of NOx emission prediction of coal-fired power plants. Fuel 296, 16 (2021). 41

    2021

  4. [12]

    A soft sensing method of billet surface temperature based on ILGSSA -LSSVM

    Liu J, et al. A soft sensing method of billet surface temperature based on ILGSSA -LSSVM. Sci Rep 12, 10 (2022)

    2022

  5. [13]

    A Novel Input Variable Selection and Structure Optimization Algorithm for Multilayer Perceptron-Based Soft Sensors

    Wang HX, Sui L, Zhang MY, Zhang FF, Ma FY, Sun K. A Novel Input Variable Selection and Structure Optimization Algorithm for Multilayer Perceptron-Based Soft Sensors. Math Probl Eng 2021, 10 (2021)

    2021

  6. [14]

    Artificial intelligence for enhanced flotation monitoring in the mining industry: A ConvLSTM-based approach

    Bendaouia A, et al. Artificial intelligence for enhanced flotation monitoring in the mining industry: A ConvLSTM-based approach. Comput Chem Eng 180, 13 (2024)

    2024

  7. [15]

    Stacked Spatial-Temporal Autoencoder for Quality Prediction in Industrial Processes

    Yan F, Yang CJ, Zhang XM. Stacked Spatial-Temporal Autoencoder for Quality Prediction in Industrial Processes. IEEE Trans Ind Inform 19, 8625-8634 (2023)

    2023

  8. [16]

    Soft sensor based on stacked auto-encoder deep neural network for air preheater rotor deformation prediction

    Wang X, Liu H. Soft sensor based on stacked auto-encoder deep neural network for air preheater rotor deformation prediction. Adv Eng Inform 36, 112-119 (2018)

    2018

  9. [17]

    A Review on Soft Sensors for Monitoring, Control, and Optimization of Industrial Processes

    Jiang YC, Yin S, Dong JW, Kaynak O. A Review on Soft Sensors for Monitoring, Control, and Optimization of Industrial Processes. IEEE Sens J 21, 12868-12881 (2021)

    2021

  10. [18]

    Review on data-driven modeling and monitoring for plant-wide industrial processes

    Ge ZQ. Review on data-driven modeling and monitoring for plant-wide industrial processes. Chemometrics Intell Lab Syst 171, 16-25 (2017)

    2017

  11. [19]

    Attention-Based Interval Aided Networks for Data Modeling of Heterogeneous Sampling Sequences With Missing Values in Process Industry

    Yuan XF, et al. Attention-Based Interval Aided Networks for Data Modeling of Heterogeneous Sampling Sequences With Missing Values in Process Industry. IEEE Trans Ind Inform 20, 5253 -5262 (2024)

    2024

  12. [20]

    A missing manufacturing process data imputation framework for nonlinear dynamic soft sensor modeling and its application

    Ma L, Wang MW, Peng KX. A missing manufacturing process data imputation framework for nonlinear dynamic soft sensor modeling and its application. Expert Syst Appl 237, 11 (2024)

    2024

  13. [21]

    Comparison of Soft-Sensor Design Methods for Industrial Plants Using Small Data Sets

    Fortuna L, Graziani S, Xibilia MG. Comparison of Soft-Sensor Design Methods for Industrial Plants Using Small Data Sets. IEEE Trans Instrum Meas 58, 2444-2451 (2009)

    2009

  14. [22]

    Novel Virtual Sample Generation Using Target-Relevant Autoencoder for Small Data-Based Soft Sensor

    Tian Y, Xu Y, Zhu QX, He YL. Novel Virtual Sample Generation Using Target-Relevant Autoencoder for Small Data-Based Soft Sensor. IEEE Trans Instrum Meas 70, 10 (2021)

    2021

  15. [23]

    A Self-Interpretable Soft Sensor Based on Deep Learning and Multiple Attention Mechanism: From Data Selection to Sensor Modeling

    Guo RY, Liu H, Xie G, Zhang YM, Liu D. A Self-Interpretable Soft Sensor Based on Deep Learning and Multiple Attention Mechanism: From Data Selection to Sensor Modeling. IEEE Trans Ind Inform 19, 6859- 6871 (2023)

    2023

  16. [24]

    Constructing a Symbolic Regression-Based Interpretable Soft Sensor for Industrial Data Analytics and Product Quality Control

    Kay H, et al. Constructing a Symbolic Regression-Based Interpretable Soft Sensor for Industrial Data Analytics and Product Quality Control. Ind Eng Chem Res 63, 4083-4092 (2024). 42

    2024

  17. [25]

    Machine Learning-Based Dynamic Modeling for Process Engineering Applications: A Guideline for Simulation and Prediction from Perceptron to Deep Learning

    Rebello CM, et al. Machine Learning-Based Dynamic Modeling for Process Engineering Applications: A Guideline for Simulation and Prediction from Perceptron to Deep Learning. Processes 10, 18 (2022)

    2022

  18. [26]

    Gpt-4 technical report

    Achiam J, et al. Gpt-4 technical report. arXiv preprint arXiv:230308774, (2023)

    2023

  19. [27]

    Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

    Team G, et al. Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context. arXiv preprint arXiv:240305530, (2024)

    2024

  20. [28]

    A survey of large language models

    Zhao WX, et al. A survey of large language models. arXiv preprint arXiv:230318223, (2023)

    2023

  21. [29]

    Pre-trained multimodal large language model enhances dermatological diagnosis using SkinGPT-4

    Zhou JX, et al. Pre-trained multimodal large language model enhances dermatological diagnosis using SkinGPT-4. Nat Commun 15, 12 (2024)

    2024

  22. [30]

    Mathprompter: Mathematical reasoning using large language models

    Imani S, Du L, Shrivastava H. Mathprompter: Mathematical reasoning using large language models. arXiv preprint arXiv:230305398, (2023)

    2023

  23. [31]

    PRISM: Patient Records Interpretation for Semantic clinical trial Matching system using large language models

    Gupta S, et al. PRISM: Patient Records Interpretation for Semantic clinical trial Matching system using large language models. npj Digit Med 7, 12 (2024)

    2024

  24. [32]

    Augmenting large language models with chemistry tools

    Bran AM, Cox S, Schilter O, Baldassari C, White AD, Schwaller P. Augmenting large language models with chemistry tools. Nat Mach Intell 6, 13 (2024)

    2024

  25. [33]

    Emergent abilities of large language models

    Wei J, et al. Emergent abilities of large language models. arXiv preprint arXiv:220607682, (2022)

    2022

  26. [34]

    Emergent analogical reasoning in large language models

    Webb T, Holyoak KJ, Lu HJ. Emergent analogical reasoning in large language models. Nat Hum Behav 7, 1526-1541 (2023)

    2023

  27. [35]

    A survey on in-context learning

    Dong Q, et al. A survey on in-context learning. In: Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (2024)

    2024

  28. [36]

    An explanation of in-context learning as implicit bayesian inference

    Xie SM, Raghunathan A, Liang P, Ma T. An explanation of in-context learning as implicit bayesian inference. arXiv preprint arXiv:211102080, (2021)

    2021

  29. [37]

    Sensitive Feature Selection for Industrial Flotation Process Soft Sensor Based on Multiswarm PSO With Collaborative Search

    Xie SW, Yu YJ, Xie YF, Tang ZH. Sensitive Feature Selection for Industrial Flotation Process Soft Sensor Based on Multiswarm PSO With Collaborative Search. IEEE Sens J 24, 17159 -17168 (2024)

    2024

  30. [38]

    Retrieval-augmented generation for knowledge-intensive nlp tasks

    Lewis P, et al. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in Neural Information Processing Systems 33, 9459-9474 (2020)

    2020

  31. [39]

    Retrieval-augmented generation for large language models: A survey

    Gao Y, et al. Retrieval-augmented generation for large language models: A survey. arXiv preprint arXiv:231210997, (2023). 43

    2023

  32. [40]

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

    Wei J, et al. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems 35, 24824-24837 (2022)

    2022

  33. [41]

    A survey of chain of thought reasoning: Advances, frontiers and future

    Chu Z, et al. A survey of chain of thought reasoning: Advances, frontiers and future. arXiv preprint arXiv:230915402, (2023)

    2023

  34. [42]

    What makes good in-context demonstrations for code intelligence tasks with llms? In: 2023 38th IEEE/ACM International Conference on Automated Software Engineering (ASE))

    Gao S, Wen X-C, Gao C, Wang W, Zhang H, Lyu MR. What makes good in-context demonstrations for code intelligence tasks with llms? In: 2023 38th IEEE/ACM International Conference on Automated Software Engineering (ASE)). IEEE (2023)

    2023

  35. [43]

    Role play with large language models

    Shanahan M, McDonell K, Reynolds L. Role play with large language models. Nature 623, 493- 498 (2023)

    2023

  36. [44]

    Large language models understand and can be enhanced by emotional stimuli

    Li C, et al. Large language models understand and can be enhanced by emotional stimuli. arXiv preprint arXiv:230711760, (2023)

    2023

  37. [45]

    The development of an industrial -scale fed-batch fermentation simulation

    Goldrick S, Stefan A, Lovett D, Montague G, Lennox B. The development of an industrial -scale fed-batch fermentation simulation. J Biotechnol 193, 70-82 (2015)

    2015

  38. [46]

    Interpretable Machine Learning Methods for Monitoring Polymer Degradation in Extrusion of Polylactic Acid

    Munir N, et al. Interpretable Machine Learning Methods for Monitoring Polymer Degradation in Extrusion of Polylactic Acid. Polymers 15, 23 (2023)

    2023

  39. [47]

    Weighted target feature regression neural networks based soft sensing for industrial process

    Guo XP, Wang QQ, Li Y. Weighted target feature regression neural networks based soft sensing for industrial process. Can J Chem Eng 102, 840-852 (2024)

    2024

  40. [48]

    Mutual Information Deep Sparse Auto-Encoding Hybrid DLSTM Prediction Network

    Li J, Huang H. Mutual Information Deep Sparse Auto-Encoding Hybrid DLSTM Prediction Network. Computer Engineering and Application 58, 277-285 (2022)

    2022

  41. [49]

    Soft measurement model for wet gas flow rate based on ultrasonic and differential pressure sensing

    Zheng DD, Shao SM, Liu AN, Wang MS, Li T. Soft measurement model for wet gas flow rate based on ultrasonic and differential pressure sensing. Meas Sci Technol 35, 13 (2024)

    2024

  42. [50]

    Soft sensor method of multimode BOF steelmaking endpoint carbon content and temperature based on vMF-WSAE dynamic deep learning

    Yang L, Liu H, Chen FG. Soft sensor method of multimode BOF steelmaking endpoint carbon content and temperature based on vMF-WSAE dynamic deep learning. High Temp Mater Process 42, 21 (2023)

    2023

  43. [51]

    Multi-objective PSO semi-supervised random forest method for dioxin soft sensor

    Xu W, Tang J, Xia H, Yu W, Qiao JF. Multi-objective PSO semi-supervised random forest method for dioxin soft sensor. Eng Appl Artif Intell 135, 16 (2024)

    2024

  44. [52]

    An evolutionary deep learning soft sensor model based on random forest feature selection technique for penicillin fermentation process

    Hua L, Zhang C, Sun W, Li YM, Xiong JL, Nazir MS. An evolutionary deep learning soft sensor model based on random forest feature selection technique for penicillin fermentation process. ISA Trans 136, 139 -151 (2023). 44

    2023

  45. [53]

    Utilization of Machine Learning Techniques i n Hot-Film Based Airflow Rate Sensors for Improving Flow Measurement

    Shin S, Baek K, Choi Y, So H. Utilization of Machine Learning Techniques i n Hot-Film Based Airflow Rate Sensors for Improving Flow Measurement. Adv Intell Syst 6, 10 (2024)

    2024

  46. [54]

    Robust probabilistic principal component regression with switching mixture Gaussian noise for soft sensing

    Sadeghian A, Jan NM, Wu O, Huang B. Robust probabilistic principal component regression with switching mixture Gaussian noise for soft sensing. Chemometrics Intell Lab Syst 222, 11 (2022)

    2022

  47. [55]

    Temperature Prediction Model for Roller Kiln by ALD-Based Double Locally Weighted Kernel Principal Component Regression

    Chen N, Dai JY, Yuan XF, Gui WH, Ren WT, Koivo HN. Temperature Prediction Model for Roller Kiln by ALD-Based Double Locally Weighted Kernel Principal Component Regression. IEEE Trans Instrum Meas 67, 2001-2010 (2018)

    2018

  48. [56]

    Uncertainty-aware soft sensor using Bayesian recurrent neural networks

    Lee M, Bae J, Kim SB. Uncertainty-aware soft sensor using Bayesian recurrent neural networks. Adv Eng Inform 50, 14 (2021)

    2021

  49. [57]

    When Deep Learning-Based Soft Sensors Encounter Reliability Challenges: A Practical Knowledge-Guided Adversarial Attack and Its Defense

    Guo RY, Liu H, Liu D. When Deep Learning-Based Soft Sensors Encounter Reliability Challenges: A Practical Knowledge-Guided Adversarial Attack and Its Defense. IEEE Trans Ind Inform 20, 2702-2714 (2024)

    2024

  50. [58]

    Adversarial Robustness Enhancement for Deep Learning- Based Soft Sensors: An Adversarial Training Strategy Using Historical Gradients and Domain Adaptation

    Guo RY, Chen QY, Liu H, Wang WQ. Adversarial Robustness Enhancement for Deep Learning- Based Soft Sensors: An Adversarial Training Strategy Using Historical Gradients and Domain Adaptation. S ensors 24, 15 (2024)

    2024

  51. [59]

    Adaptive Multi-Head Self-Attention Based supervised VAE for Industrial Soft Sensing With Missing Data

    Chen L, Xu Y, Zhu QX, He YL. Adaptive Multi-Head Self-Attention Based supervised VAE for Industrial Soft Sensing With Missing Data. IEEE Trans Autom Sci Eng 21, 3564- 3575 (2024)

    2024

  52. [60]

    Partial transfer learning network for data imputation and soft sensor under various operation conditions

    Ren JY, Chen X, Zhao CH. Partial transfer learning network for data imputation and soft sensor under various operation conditions. J Cent South Univ 30, 3395-3413 (2023)

    2023

  53. [61]

    Training language models to follow instructions with human feedback

    Ouyang L, et al. Training language models to follow instructions with human feedback. Advances in neural information processing systems 35, 27730-27744 (2022)

    2022

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.