从故事生成程序到大语言模型:人工智能编剧创作的历史与实践

人工智能生成内容(AI Generated Content,简称AIGC)电影包括内涵丰富的多种维度,从编剧角度可以分为AI生成故事(AI Generated Storyline,简称AIGSL)与AI生成剧本(AI Generated Screenplay,简称AIGSP)。AIGSL研究始于20世纪60年代初,按照创作模式可以分为基于结构模型的故事生成、基于规则的故事生成和基于机器学习的故事生成三类程序。2022年底以来,以Chat GPT为代表的大语言模型将AIGC的内容从简单的AIGSL拓展到了AIGSP领域,同时催生了基于大语言模型的专用故事与剧本生成工具。大语言模型的故事和剧本生产存在事实错误、陈词滥调以及主观偏见等天然缺陷,但AIGSL与AIGSP已经开始实现产业化。目前的人工智能编剧创作绝大多数并不完全由AI创作,属于弱人工智能领域。中国电影工作者与软件工程师有必要利用机器学习等AI技术,学习中国传统叙事文本并总结其叙事风格,为具有中国叙事特色的编剧创作提供智能助力。

Recent Progress in Reinforcement Learning and Adaptive Dynamic Programming for Advanced Control Applications

Reinforcement learning(RL) has roots in dynamic programming and it is called adaptive/approximate dynamic programming(ADP) within the control community. This paper reviews recent developments in ADP along with RL and its applications to various advanced control fields. First, the background of the development of ADP is described, emphasizing the significance of regulation and tracking control problems. Some effective offline and online algorithms for ADP/adaptive critic control are displayed, where the main results towards discrete-time systems and continuous-time systems are surveyed, respectively.Then, the research progress on adaptive critic control based on the event-triggered framework and under uncertain environment is discussed, respectively, where event-based design, robust stabilization, and game design are reviewed. Moreover, the extensions of ADP for addressing control problems under complex environment attract enormous attention. The ADP architecture is revisited under the perspective of data-driven and RL frameworks,showing how they promote ADP formulation significantly.Finally, several typical control applications with respect to RL and ADP are summarized, particularly in the fields of wastewater treatment processes and power systems, followed by some general prospects for future research. Overall, the comprehensive survey on ADP and RL for advanced control applications has d emonstrated its remarkable potential within the artificial intelligence era. In addition, it also plays a vital role in promoting environmental protection and industrial intelligence.

Crosstalk among mitophagy,pyroptosis,ferroptosis,and necroptosis in central nervous system injuries

Central nervous system injuries have a high rate of resulting in disability and mortality;however,at present,effective treatments are lacking.Programmed cell death,which is a genetically determined fo rm of active and ordered cell death with many types,has recently attra cted increasing attention due to its functions in determining the fate of cell survival.A growing number of studies have suggested that programmed cell death is involved in central nervous system injuries and plays an important role in the progression of brain damage.In this review,we provide an ove rview of the role of programmed cell death in central nervous system injuries,including the pathways involved in mitophagy,pyroptosis,ferroptosis,and necroptosis,and the underlying mechanisms by which mitophagy regulates pyroptosis,ferroptosis,and necro ptosis.We also discuss the new direction of therapeutic strategies to rgeting mitophagy for the treatment of central nervous system injuries,with the aim to determine the connection between programmed cell death and central nervous system injuries and to identify new therapies to modulate programmed cell death following central nervous system injury.In conclusion,based on these properties and effects,interventions targeting programmed cell death could be developed as potential therapeutic agents for central nervous system injury patients.

Hepatic arterial infusion chemotherapy with anti-angiogenesis agents and immune checkpoint inhibitors for unresectable hepatocellular carcinoma and meta-analysis

BACKGROUND Hepatic arterial infusion chemotherapy(HAIC)has been proven to be an ideal choice for treating unresectable hepatocellular carcinoma(uHCC).HAIC-based treatment showed great potential for treating uHCC.However,large-scale studies on HAIC-based treatments and meta-analyses of first-line treatments for uHCC are lacking.AIM To investigate better first-line treatment options for uHCC and to assess the safety and efficacy of HAIC combined with angiogenesis inhibitors,programmed cell death of protein 1(PD-1)and its ligand(PD-L1)blockers(triple therapy)under real-world conditions.METHODS Several electronic databases were searched to identify eligible randomized controlled trials for this meta-analysis.Study-level pooled analyses of hazard ratios(HRs)and odds ratios(ORs)were performed.This was a retrospective single-center study involving 442 patients with uHCC who received triple therapy or angiogenesis inhibitors plus PD-1/PD-L1 blockades(AIPB)at Sun Yat-sen University Cancer Center from January 2018 to April 2023.Propensity score matching(PSM)was performed to balance the bias between the groups.The Kaplan-Meier method and cox regression were used to analyse the survival data,and the log-rank test was used to compare the suvival time between the groups.RESULTS A total of 13 randomized controlled trials were included.HAIC alone and in combination with sorafenib were found to be effective treatments(P values for ORs:HAIC,0.95;for HRs:HAIC+sorafenib,0.04).After PSM,176 HCC patients were included in the analysis.The triple therapy group(n=88)had a longer median overall survival than the AIPB group(n=88)(31.6 months vs 14.6 months,P<0.001)and a greater incidence of adverse events(94.3%vs 75.4%,P<0.001).CONCLUSION This meta-analysis suggests that HAIC-based treatments are likely to be the best choice for uHCC.Our findings confirm that triple therapy is more effective for uHCC patients than AIPB.

Smart prediction of liquefaction-induced lateral spreading

The prediction of liquefaction-induced lateral spreading/displacement(Dh)is a challenging task for civil/geotechnical engineers.In this study,a new approach is proposed to predict Dh using gene expression programming(GEP).Based on statistical reasoning,individual models were developed for two topographies:free-face and gently sloping ground.Along with a comparison with conventional approaches for predicting the Dh,four additional regression-based soft computing models,i.e.Gaussian process regression(GPR),relevance vector machine(RVM),sequential minimal optimization regression(SMOR),and M5-tree,were developed and compared with the GEP model.The results indicate that the GEP models predict Dh with less bias,as evidenced by the root mean square error(RMSE)and mean absolute error(MAE)for training(i.e.1.092 and 0.815;and 0.643 and 0.526)and for testing(i.e.0.89 and 0.705;and 0.773 and 0.573)in free-face and gently sloping ground topographies,respectively.The overall performance for the free-face topology was ranked as follows:GEP>RVM>M5-tree>GPR>SMOR,with a total score of 40,32,24,15,and 10,respectively.For the gently sloping condition,the performance was ranked as follows:GEP>RVM>GPR>M5-tree>SMOR with a total score of 40,32,21,19,and 8,respectively.Finally,the results of the sensitivity analysis showed that for both free-face and gently sloping ground,the liquefiable layer thickness(T_(15))was the major parameter with percentage deterioration(%D)value of 99.15 and 90.72,respectively.

A stable implicit nodal integration-based particle finite element method(N-PFEM)for modelling saturated soil dynamics

In this study,we present a novel nodal integration-based particle finite element method(N-PFEM)designed for the dynamic analysis of saturated soils.Our approach incorporates the nodal integration technique into a generalised Hellinger-Reissner(HR)variational principle,creating an implicit PFEM formulation.To mitigate the volumetric locking issue in low-order elements,we employ a node-based strain smoothing technique.By discretising field variables at the centre of smoothing cells,we achieve nodal integration over cells,eliminating the need for sophisticated mapping operations after re-meshing in the PFEM.We express the discretised governing equations as a min-max optimisation problem,which is further reformulated as a standard second-order cone programming(SOCP)problem.Stresses,pore water pressure,and displacements are simultaneously determined using the advanced primal-dual interior point method.Consequently,our numerical model offers improved accuracy for stresses and pore water pressure compared to the displacement-based PFEM formulation.Numerical experiments demonstrate that the N-PFEM efficiently captures both transient and long-term hydro-mechanical behaviour of saturated soils with high accuracy,obviating the need for stabilisation or regularisation techniques commonly employed in other nodal integration-based PFEM approaches.This work holds significant implications for the development of robust and accurate numerical tools for studying saturated soil dynamics.

Rehabilitation care for pain in elderly knee replacement patients

BACKGROUND Total knee arthroplasty(TKA)is recognized as the most effective surgical intervention for relieving pain and improving joint mobility and deformity in patients with knee osteoarthritis and other synovial diseases.The application of accelerated postoperative rehabilitation(enhanced recovery after surgery)has demonstrated its efficacy in improving patient outcomes,and early postoperative joint function exercise has become a key prognostic factor in knee replacement.The unexpected appearance of limb pain and swelling hindered the patient’s tendency for early mobilization,leading in prolonged hospitalization,delayed functional recovery and negative psychological responses.A retrospective analysis was conducted on a cohort of 116 patients who under-went TKA at our hospital between July 2019 and July 2021.The patients were divided into two groups:A control group(n=58)receiving programmatic nur-sing,and an observed group(n=58)receiving programmed nursing combined with a collaborative nursing model.A pain management team consisting of attending physicians,head nurses,and responsible nurses was established.Outcome measures included visual analogue scale(VAS)scores,activities of daily living(ADL)scale scores,and functional scores.The ADL scores of patients in both groups exhibited a continuous increase.However,there was no statistically significant difference in the ADL scores between the two groups at 48 h and the 7th d post-surgery(P>0.05).Upon reexamination at the 3rd mo,the observation group demonstrated higher ADL scores compared to the control group(67.48±14.69 vs 59.40±16.06,P<0.05).The VAS scores of both groups significantly decreased,with no significant difference observed between the groups at each time point(P>0.05).The functional status of patients in both groups exhibited a gradual increase prior to intervention and at the 1st,2nd,and 3rd month following discharge(P<0.05).There was no statistically significant difference in knee joint function scores between the two groups at the 1st month after discharge(47.52 vs 45.81,P>0.05).However,the knee joint function scores of patients in the observation group were significantly higher than those in the control group at the 2nd(59.38 vs 53.19,P<0.05)and 3rd month(71.92 vs 64.34,P<0.05)following discharge.CONCLUSION The utilization of programmed pain nursing in conjunction with collaborative nursing for out-of-hospital care of TKA patients has demonstrated favorable outcomes,encompassing pain reduction,enhanced prognosis,and improved nursing quality for patients.

Immune signature of small bowel adenocarcinoma and the role of tumor microenvironment

In this editorial we comment on the article published“Clinical significance of programmed cell death-ligand expression in small bowel adenocarcinoma is determined by the tumor microenvironment”.Small bowel adenocarcinoma(SBA)is a rare gastrointestinal neoplasm and despite the small intestine's significant surface area,SBA accounts for less than 3%of such tumors.Early detection is challenging and the reason arises from its asymptomatic nature,often leading to late-stage discovery and poor prognosis.Treatment involves platinum-based chemotherapy with a 5-fluorouracil combination,but the lack of effective chemotherapy contributes to a generally poor prognosis.SBAs are linked to genetic disorders and risk factors,including chronic inflammatory conditions.The unique characteristics of the small bowel,such as rapid cell renewal and an active immune system,contributes to the rarity of these tumors as well as the high intratumoral infiltration of immune cells is associated with a favorable prognosis.Programmed cell death-ligand 1(PD-L1)expression varies across different cancers,with potential discrepancies in its prognostic value.Microsatellite instability(MSI)in SBA is associated with a high tumor mutational burden,affecting the prognosis and response to immunotherapy.The presence of PD-L1 and programmed cell death 1,along with tumor-infiltrating lymphocytes,plays a crucial role in the complex microenvironment of SBA and contributes to a more favorable prognosis,especially in the context of high MSI tumors.Stromal tumor-infiltrating lymphocytes are identified as independent prognostic indicators and the association between MSI status and a favorable prognosis,emphasizes the importance of evaluating the immune status of tumors for treatment decisions.

Quercetin Alleviates Lipopolysaccharide-Induced Cardiac Inflammation via Inhibiting Autophagy and Programmed Cell Death

Objective The aim of this study is to explore the potential modulatory role of quercetin against Endotoxin or lipopolysaccharide(LPS)induced septic cardiac dysfunction.Methods Specific pathogen-free chicken embryos(n=120)were allocated untreated control,phosphate buffer solution(PBS)vehicle,PBS with ethanol vehicle,LPS(500 ng/egg),LPS with quercetin treatment(10,20,or 40 nmol/egg,respectively),Quercetin groups(10,20,or 40 nmol/egg).Fifteenday-old embryonated eggs were inoculated with abovementioned solutions via the allantoic cavity.At embryonic day 19,the hearts of the embryos were collected for histopathological examination,RNA extraction,real-time polymerase chain reaction,immunohistochemical investigations,and Western blotting.Results They demonstrated that the heart presented inflammatory responses after LPS induction.The LPS-induced higher mRNA expressions of inflammation-related factors(TLR4,TNFα,MYD88,NF-κB1,IFNγ,IL-1β,IL-8,IL-6,IL-10,p38,MMP3,and MMP9)were blocked by quercetin with three dosages.Quercetin significantly decreased immunopositivity to TLR4 and MMP9 in the treatment group when compared with the LPS group.Quercetin significantly decreased protein expressions of TLR4,IFNγ,MMP3,and MMP9 when compared with the LPS group.Quercetin treatment prevented LPS-induced increase in the mRNA expression of Claudin 1 and ZO-1,and significantly decreased protein expression of claudin 1 when compared with the LPS group.Quercetin significantly downregulated autophagyrelated gene expressions(PPARα,SGLT1,APOA4,AMPKα1,AMPKα2,ATG5,ATG7,Beclin-1,and LC3B)and programmed cell death(Fas,Bcl-2,CASP1,CASP12,CASP3,and RIPK1)after LPS induction.Quercetin significantly decreased immunopositivity to APOA4,AMPKα2,and LC3-II/LC3-I in the treatment group when compared with the LPS group.Quercetin significantly decreased protein expressions of AMPKα1,LC3-I,and LC3-II.Quercetin significantly decreased the protein expression to CASP1 and CASP3 by immunohistochemical investigation or Western blotting in treatment group when compared with LPS group.Conclusion Quercetin alleviates cardiac inflammation induced by LPS through modulating autophagy,programmed cell death,and myocardiocytes permeability.

In vivo astrocyte reprogramming following spinal cord injury

Harmful and helpful roles of astrocytes in spinal cord injury(SCI):SCI induce gradable sensory,motor and autonomic impairments that correlate with the lesion severity and the rostro-caudal location of the injury site.The absence of spontaneous axonal regeneration after injury results from neuron-intrinsic and neuron-extrinsic parameters.Indeed,not only adult neurons display limited capability to regrow axons but also the injury environment contains inhibitors to axonal regeneration and a lack of growth-promoting factors.Amongst other cell populations that respond to the lesion,reactive astrocytes were first considered as only detrimental to spontaneous axonal regeneration.Indeed,astrocytes.

Multi-Timescale Distributed Approach to Generalized-Nash-Equilibrium Seeking in Noncooperative Nonconvex Games

Dear Editor,The distributed generalized-Nash-equilibrium(GNE)seeking in noncooperative games with nonconvexity is the topic of this letter.Inspired by the sequential quadratic programming(SQP)method,a multi-timescale multi-agent system(MAS)is developed,and its convergence to a critical point of the game is proven.To illustrate the qualities and efficacy of the theoretical findings,a numerical example is elaborated.

Transcriptional reprogramming during human osteoclast differentiation identifies regulators of osteoclast activity

Enhanced osteoclastogenesis and osteoclast activity contribute to the development of osteoporosis,which is characterized by increased bone resorption and inadequate bone formation.As novel antiosteoporotic therapeutics are needed,understanding the genetic regulation of human osteoclastogenesis could help identify potential treatment targets.This study aimed to provide an overview of transcriptional reprogramming during human osteoclast differentiation.Osteoclasts were differentiated from CD14+monocytes from eight female donors.RNA sequencing during differentiation revealed 8980 differentially expressed genes grouped into eight temporal patterns conserved across donors.These patterns revealed distinct molecular functions associated with postmenopausal osteoporosis susceptibility genes based on RNA from iliac crest biopsies and bone mineral density SNPs.Network analyses revealed mutual dependencies between temporal expression patterns and provided insight into subtype-specific transcriptional networks.The donor-specific expression patterns revealed genes at the monocyte stage,such as filamin B(FLNB)and oxidized low-density lipoprotein receptor 1(OLR1,encoding LOX-1),that are predictive of the resorptive activity of mature osteoclasts.The expression of differentially expressed G-protein coupled receptors was strong during osteoclast differentiation,and these receptors are associated with bone mineral density SNPs,suggesting that they play a pivotal role in osteoclast differentiation and activity.The regulatory effects of three differentially expressed G-protein coupled receptors were exemplified by in vitro pharmacological modulation of complement 5 A receptor 1(C5AR1),somatostatin receptor 2(SSTR2),and free fatty acid receptor 4(FFAR4/GPR120).Activating C5AR1 enhanced osteoclast formation,while activating SSTR2 decreased the resorptive activity of mature osteoclasts,and activating FFAR4 decreased both the number and resorptive activity of mature osteoclasts.In conclusion,we report the occurrence of transcriptional reprogramming during human osteoclast differentiation and identified SSTR2 and FFAR4 as antiresorptive G-protein coupled receptors and FLNB and LOX-1 as potential molecular markers of osteoclast activity.These data can help future investigations identify molecular regulators of osteoclast differentiation and activity and provide the basis for novel antiosteoporotic targets.

Glia-to-neuron reprogramming to the rescue?

Over the last two decades,the dogma that cell fate is immutable has been increasingly challenged,with important implications for regenerative medicine.The brea kth rough discovery that induced pluripotent stem cells could be generated from adult mouse fibroblasts is powerful proof that cell fate can be changed.An exciting extension of the discovery of cell fate impermanence is the direct cellular reprogram ming hypothesis-that terminally differentiated cells can be reprogrammed into other adult cell fates without first passing through a stem cell state.

Uncertainty quantification of inverse analysis for geomaterials using probabilistic programming

Uncertainty is an essentially challenging for safe construction and long-term stability of geotechnical engineering.The inverse analysis is commonly utilized to determine the physico-mechanical parameters.However,conventional inverse analysis cannot deal with uncertainty in geotechnical and geological systems.In this study,a framework was developed to evaluate and quantify uncertainty in inverse analysis based on the reduced-order model(ROM)and probabilistic programming.The ROM was utilized to capture the mechanical and deformation properties of surrounding rock mass in geomechanical problems.Probabilistic programming was employed to evaluate uncertainty during construction in geotechnical engineering.A circular tunnel was then used to illustrate the proposed framework using analytical and numerical solution.The results show that the geomechanical parameters and associated uncertainty can be properly obtained and the proposed framework can capture the mechanical behaviors under uncertainty.Then,a slope case was employed to demonstrate the performance of the developed framework.The results prove that the proposed framework provides a scientific,feasible,and effective tool to characterize the properties and physical mechanism of geomaterials under uncertainty in geotechnical engineering problems.

Combining reinforcement learning with mathematical programming:An approach for optimal design of heat exchanger networks

Heat integration is important for energy-saving in the process industry.It is linked to the persistently challenging task of optimal design of heat exchanger networks(HEN).Due to the inherent highly nonconvex nonlinear and combinatorial nature of the HEN problem,it is not easy to find solutions of high quality for large-scale problems.The reinforcement learning(RL)method,which learns strategies through ongoing exploration and exploitation,reveals advantages in such area.However,due to the complexity of the HEN design problem,the RL method for HEN should be dedicated and designed.A hybrid strategy combining RL with mathematical programming is proposed to take better advantage of both methods.An insightful state representation of the HEN structure as well as a customized reward function is introduced.A Q-learning algorithm is applied to update the HEN structure using theε-greedy strategy.Better results are obtained from three literature cases of different scales.

Effect of eccentric and inclined loading on the bearing capacity of strip footing placed on rock mass

This paper deals with the bearing capacity determination of strip footing on a rock mass in hilly area by considering the influence of inclined and eccentric loading. Applying the generalized HoekBrown failure criterion, the failure behavior of the rock mass is modeled with the help of the power cone programming in the lower bound finite element limit analysis framework. Using bearing capacity factor(Ns), the change in bearing capacity of the strip footing due to the occurrence of eccentrically inclined loading is presented. The variations of the magnitude of Ns are obtained by examining the effects of the Hoek-Brown rock mass strength parameters(uniaxial compressive strength(sci), disturbance factor(D), rock parameter(mi), and Geological Strength Index(GSI)) in the presence of different magnitudes of eccentricity(e) and inclination angle(λ) with respect to the vertical plane, and presented as design charts. Both the inclined loading modes, i.e., inclination towards the center of strip footing(+λ) and inclination away from the center of strip footing(-λ), are adopted to perform the investigation. In addition, the correlation between the input parameters and the corresponding output is developed by utilizing the artificial neural network(ANN). Additionally, from sensitivity analysis, it is observed that inclination angle(λ) is the most sensitive parameter. For practicing engineers, the obtained design equation and design charts can be beneficial to understand the bearing capacity variation in the existence of eccentrically inclined loading in mountain areas.

Spatiotemporal transformable nano-assembly for on-demand drug delivery to enhance anti-tumor immunotherapy

Induction of tumor cell senescence has become a promising strategy for anti-tumor immunotherapy,but fibrotic matrix severely blocks senescence inducers penetration and immune cells infiltration.Herein,we designed a cancer-associated fibroblasts(CAFs)triggered structure-transformable nano-assembly(HSD-P@V),which can directionally deliver valsartan(Val,CAFs regulator)and doxorubicin(DOX,senescence inducer)to the specific targets.In detail,DOX is conjugated with hyaluronic acid(HA)via diselenide bonds(Se-Se)to form HSD micelles,while CAFs-sensitive peptide is grafted onto the HSD to form a hydrophilic polymer,which is coated on Val nanocrystals(VNs)surface for improving the stability and achieving responsive release.Once arriving at tumor microenvironment and touching CAFs,HSD-P@V disintegrates into VNs and HSD micelles due to sensitive peptide detachment.VNs can degrade the extracellularmatrix,leading to the enhanced penetration of HSD.HSD targets tumor cells,releases DOX to induce senescence,and recruits effector immune cells.Furthermore,senescent cells are cleared by the recruited immune cells to finish the integrated anti-tumor therapy.In vitro and in vivo results show that the nanoassembly remarkably inhibits tumor growth as well as lungmetastasis,and extends tumorbearing mice survival.This work provides a promising paradigm of programmed delivering multi-site nanomedicine for cancer immunotherapy.

Ethical Decision-Making Framework Based on Incremental ILP Considering Conflicts

Humans are experiencing the inclusion of artificial agents in their lives,such as unmanned vehicles,service robots,voice assistants,and intelligent medical care.If the artificial agents cannot align with social values or make ethical decisions,they may not meet the expectations of humans.Traditionally,an ethical decision-making framework is constructed by rule-based or statistical approaches.In this paper,we propose an ethical decision-making framework based on incremental ILP(Inductive Logic Programming),which can overcome the brittleness of rule-based approaches and little interpretability of statistical approaches.As the current incremental ILP makes it difficult to solve conflicts,we propose a novel ethical decision-making framework considering conflicts in this paper,which adopts our proposed incremental ILP system.The framework consists of two processes:the learning process and the deduction process.The first process records bottom clauses with their score functions and learns rules guided by the entailment and the score function.The second process obtains an ethical decision based on the rules.In an ethical scenario about chatbots for teenagers’mental health,we verify that our framework can learn ethical rules and make ethical decisions.Besides,we extract incremental ILP from the framework and compare it with the state-of-the-art ILP systems based on ASP(Answer Set Programming)focusing on conflict resolution.The results of comparisons show that our proposed system can generate better-quality rules than most other systems.

Value Iteration-Based Cooperative Adaptive Optimal Control for Multi-Player Differential Games With Incomplete Information

This paper presents a novel cooperative value iteration(VI)-based adaptive dynamic programming method for multi-player differential game models with a convergence proof.The players are divided into two groups in the learning process and adapt their policies sequentially.Our method removes the dependence of admissible initial policies,which is one of the main drawbacks of the PI-based frameworks.Furthermore,this algorithm enables the players to adapt their control policies without full knowledge of others’ system parameters or control laws.The efficacy of our method is illustrated by three examples.

Improved Unit Commitment with Accurate Dynamic Scenarios Clustering Based on Multi-Parametric Programming and Benders Decomposition

Stochastic unit commitment is one of the most powerful methods to address uncertainty. However, the existingscenario clustering technique for stochastic unit commitment cannot accurately select representative scenarios,which threatens the robustness of stochastic unit commitment and hinders its application. This paper providesa stochastic unit commitment with dynamic scenario clustering based on multi-parametric programming andBenders decomposition. The stochastic unit commitment is solved via the Benders decomposition, which decouplesthe primal problem into the master problem and two types of subproblems. In the master problem, the committedgenerator is determined, while the feasibility and optimality of generator output are checked in these twosubproblems. Scenarios are dynamically clustered during the subproblem solution process through the multiparametric programming with respect to the solution of the master problem. In other words, multiple scenariosare clustered into several representative scenarios after the subproblem is solved, and the Benders cut obtainedby the representative scenario is generated for the master problem. Different from the conventional stochasticunit commitment, the proposed approach integrates scenario clustering into the Benders decomposition solutionprocess. Such a clustering approach could accurately cluster representative scenarios that have impacts on theunit commitment. The proposed method is tested on a 6-bus system and the modified IEEE 118-bus system.Numerical results illustrate the effectiveness of the proposed method in clustering scenarios. Compared withthe conventional clustering method, the proposed method can accurately select representative scenarios whilemitigating computational burden, thus guaranteeing the robustness of unit commitment.