| paper_id,Summary,Questions,Limitations,Ethical Concerns,Soundness,Presentation,Contribution,Overall,Confidence,Strengths,Weaknesses,Originality,Quality,Clarity,Significance,Decision | |
| mode_transition,"The paper proposes a modification to diffusion models to facilitate mode transition between different modes in data distributions. The authors evaluate their approach using Frechet Inception Distance (FID) and Inception Score (IS) on several 2D datasets, showing improved performance over a baseline model without mode connectivity mechanisms.","['Can the authors provide a more detailed explanation of the mode transition mechanism?', 'What is the qualitative performance of the proposed method? Can you provide visualizations of the generated samples?', 'How does the proposed method compare to other relevant baselines in terms of performance?', 'Can you provide a more comprehensive ablation study exploring different hyperparameters and settings?', 'What specific modifications were made to the diffusion model architecture?', 'How do the proposed changes facilitate mode transition?', 'Have you conducted any statistical tests to confirm the significance of the differences in FID scores?', 'How are the reconstruction loss and mode transition loss combined during training?', 'Why were only simplistic 2D datasets used for evaluation?', 'How does the proposed model compare with other state-of-the-art generative models on more complex datasets?', 'What are the potential limitations of the proposed method?', 'Can the authors present more comprehensive experimental results and analyses?', 'What are the potential limitations and negative societal impacts of this work?']","['The paper should provide more details on the proposed method and improve the clarity of the explanation.', 'The experimental results section should include more qualitative analysis and visualizations to support the claims.', 'A comprehensive comparison with other relevant baselines should be included.', 'The ablation study should explore different hyperparameters and settings in more detail.', 'The paper does not sufficiently address the limitations of the proposed method, especially in terms of its applicability to more complex datasets and scalability.', 'There is no discussion on the possible negative societal impacts of the work.']",False,2,2,2,3,4,"['Addresses an important problem in generative modeling, i.e., mode transition.', 'Proposes a novel modification to diffusion models to enable mode transition.', 'Uses standard evaluation metrics like FID and IS for performance assessment.']","['The method is not clearly explained. The mechanism for mode transition is vaguely described, and there are no concrete details on how it is implemented.', 'The experimental results section is weak. The paper provides FID scores but lacks qualitative analysis and visualizations to support the claims.', 'The related work section mentions various works but does not compare the proposed method to all relevant baselines comprehensively.', 'The ablation study on the mode transition rate is limited and does not explore other hyperparameters or settings in sufficient detail.', 'The clarity and organization of the paper are poor, making it hard to follow the methodology and results.', 'The novelty of the proposed method is questionable; it seems like a minor modification of existing diffusion models.', 'No discussion on the limitations of the proposed method or potential negative societal impacts.', ""The datasets used for evaluation are simplistic and do not demonstrate the method's effectiveness on more complex, real-world data.""]",2,2,2,2,Reject | |
| noise_schedule_signatures,"The paper introduces the concept of 'Noise Schedule Signatures' in diffusion processes, suggesting that these signatures can uniquely identify and characterize different datasets. The method involves analyzing noise schedules during training and inference to discover unique patterns that serve as dataset-specific fingerprints.","['How are the noise schedule signatures specifically computed and validated?', 'Can the authors provide more complex and real-world datasets for evaluation?', 'What are the practical benefits of using noise schedule signatures over traditional methods?', 'Can you provide more details on the process of signature creation?', 'What specific machine learning techniques were used to create the signatures?', 'How do the proposed noise schedule signatures improve upon existing approaches?', 'Can you provide more in-depth analysis and discussion of the experimental results?', 'How does your approach compare to existing methods in terms of performance and accuracy?', 'What are the potential limitations and negative societal impacts of your proposed method?', 'How do the proposed noise schedule signatures compare to other potential methods for dataset characterization?', 'What are the implications of using more complex datasets for evaluating the proposed method?']","['The work does not address the limitations of the proposed method or potential negative societal impacts.', 'The paper lacks a thorough discussion of limitations and potential negative societal impacts. It is important for the authors to address these aspects to provide a more balanced view of their work.']",False,2,2,2,3,4,"['The idea of using noise schedules to generate dataset-specific signatures is novel.', 'Potential implications for improving the sophistication of diffusion models.']","['The conceptual novelty is not sufficiently justified in terms of practical benefits.', 'Insufficient implementation details on computing noise schedule signatures.', 'The experimental setup uses very simplistic datasets, which may not demonstrate the method’s robustness.', 'Lack of detailed results and thorough evaluation.', 'Incomplete related work section, lacking context and comparison with existing methods.', 'The methodology section is underdeveloped and lacks detail.', 'The evaluation metrics and experimental setup are not sufficiently detailed.', 'The experimental results section lacks in-depth analysis and discussion.', 'The significance of the findings is not well demonstrated.']",2,2,2,2,Reject | |
| mode_signatures,"The paper introduces the concept of mode signatures to capture the unique behaviors exhibited by different diffusion models. The proposed method aims to identify specific mode signatures characteristic of each model, providing valuable insights into their behavior. The authors verify their approach through experiments on synthetic and real-world datasets.","['Can you provide more details on how mode signatures are computed? Specifically, what metrics are used and what is the process for identifying mode signatures?', 'Can you complete the experimental results section and provide a comparative analysis with existing methods?', 'What are the limitations of your approach?', 'Are there any potential ethical concerns with your approach?']","['The paper does not discuss the limitations of the proposed method. It is important to address this to provide a balanced view of the contributions.', 'Potential ethical concerns are not addressed. This is important to ensure the responsible application of the proposed method.']",False,2,2,2,3,4,"['The concept of mode signatures is novel and provides a new perspective on understanding diffusion models.', 'The paper addresses an important problem in the field of deep generative models, which is understanding the unique behaviors of different models.']","['The methodology section lacks clarity and does not provide sufficient details on how mode signatures are computed. For example, the paper does not explain the specific metrics used or the process of identifying mode signatures.', 'The experimental results section is incomplete, making it difficult to evaluate the effectiveness of the proposed method. The paper should include detailed results and comparisons with existing methods.', 'There is a lack of comparative analysis with existing methods. The paper should compare the proposed method with other state-of-the-art approaches to highlight its advantages and disadvantages.', 'The paper does not discuss the limitations or potential ethical concerns of their approach. It is important to address these aspects to provide a balanced view of the contributions.', 'The writing is somewhat disorganized, making it difficult to follow the flow of the paper. The authors should improve the organization and clarity of the paper.']",2,2,2,2,Reject | |
| dataset_distribution_fingerprinting,"The paper proposes a novel method for fingerprinting dataset distributions using diffusion models. The approach leverages diffusion models to learn a probabilistic representation of the data, enabling analysis of generated samples in different regions of the data space. Experiments on various 2D datasets are conducted to demonstrate the effectiveness of the method.","['Can the authors provide more detailed explanations of the methodology and the specific steps involved?', 'How does this method compare to other state-of-the-art methods in more complex datasets?', 'Can the authors provide more comprehensive experimental results and analyses?', 'Can the authors provide a detailed related work section to contextualize their contributions?', 'What is the detailed methodology for training the diffusion models, including the noise scheduler and training pipeline?', 'Can the authors include experiments on real-world datasets to demonstrate the robustness of their method?', 'Can the authors provide complete figures and captions in the results section?', 'Can the authors improve the organization and clarity of the paper?', 'What are the specific architectural details and training processes used in your method?', 'How does your method perform on more complex, high-dimensional datasets?', 'What are the potential limitations and negative societal impacts of your approach?']","['The paper does not sufficiently address the limitations and potential negative societal impacts of the work. More detailed discussions are needed.', 'The paper currently lacks a detailed comparison with existing methods, and the experiments are limited to synthetic 2D datasets.', 'The paper is limited by its experimental scope, only addressing simple 2D datasets.', 'There is a lack of thorough evaluation and comparison with more complex baselines.', 'The clarity and organization of the paper need significant improvement.']",False,2,2,2,3,4,"['Addresses an important problem in understanding dataset distributions.', 'The idea of using diffusion models for dataset distribution fingerprinting is novel.', 'The approach has potential applications in data augmentation, anomaly detection, and generative modeling.']","['Lack of detail in the methodology section, making it difficult to understand the exact process.', 'Contributions are not clearly articulated and seem incremental rather than groundbreaking.', 'The experimental setup and results analysis are underdeveloped and lack depth.', 'The paper is not well-organized, affecting clarity and readability.', 'No substantial improvements or novel insights over existing methods are convincingly demonstrated.', 'The related work section is missing, making it difficult to understand how the proposed method compares with existing approaches.', 'Experiments are only conducted on synthetic 2D datasets, which are not sufficient to demonstrate the robustness of the method.', 'The results section is incomplete, with missing figures and captions.', 'There is no thorough analysis or discussion of the results.', 'Certain sections, such as conclusions, are incomplete, indicating a lack of thoroughness.', 'Does not discuss potential limitations or negative societal impacts.']",2,2,2,2,Reject | |
| diffusion_perturbations,"The paper proposes a novel diffusion model architecture for generating high-quality 2D data, incorporating a sinusoidal embedding scheme and a residual block-based network. The goal is to improve the quality and diversity of generated 2D data, such as images and shapes.","['Can the authors provide more detailed explanations and theoretical justifications for the sinusoidal embedding scheme and the residual block configurations?', 'How does the proposed model compare to state-of-the-art methods in terms of quantitative and qualitative results?', 'Can the authors address the placeholders and incomplete sections in the paper?', 'What specific datasets were used in the experiments, and how were they selected?', 'What are the quantitative results of the proposed model compared to existing methods?', 'Can the authors clarify the evaluation metrics used in the experiments?']",['The paper does not adequately address the limitations of the proposed method or any potential negative societal impacts.'],False,1,1,2,2,4,"['Addresses a relevant problem in the field of generative modeling.', 'Incorporates sinusoidal embeddings and residual blocks, which could potentially improve model performance.']","['The novelty of the approach is limited, as diffusion models, sinusoidal embeddings, and residual blocks are well-known techniques.', 'The paper lacks detailed explanations and theoretical justifications for its methodological choices.', 'The experimental section is underdeveloped, with insufficient quantitative and qualitative results.', 'The paper is poorly organized and contains placeholders, indicating it is incomplete.', 'No discussion on potential negative societal impacts or ethical considerations.']",2,1,1,2,Reject | |
| learning_rate_schedule,"The paper proposes using diffusion models for generating high-quality 2D data, evaluated on simple datasets like circle, dino, line, and moons. The results show competitive performance in terms of training time, evaluation loss, and inference time.","['Can the authors provide more details on the noise schedule and the choice of hyperparameters?', 'How does the proposed approach compare with other state-of-the-art methods in generating high-quality 2D data?', 'What are the potential limitations of the proposed method, and how can it be extended to more complex datasets?', 'Can the authors provide more details on the architecture of the neural network used in the reverse process?', 'Can the authors provide more detailed results from the ablation study?', 'What specific hyperparameters were used, and how were they tuned?', 'Can the authors provide a more comprehensive analysis of the experimental results?', 'What are the theoretical advantages of the proposed method over existing diffusion models?', 'How does the proposed method compare with existing methods such as GANs and VAEs in terms of both quality and computational efficiency?', 'Can the authors include additional datasets in their experiments to validate the generalizability of the proposed method?']","['The paper lacks a thorough discussion on the limitations and potential extensions of the proposed method.', 'The model requires a large amount of training data to achieve good performance.', 'The model may not be suitable for datasets with complex structures or patterns.', 'Potential negative societal impacts of the work are not discussed.']",False,2,2,2,3,4,"['The paper addresses a relevant problem in generating high-quality 2D data.', 'Diffusion models are a promising and relatively novel approach for generative tasks.', 'The methodology is implemented and evaluated on several datasets, showing competitive results.']","[""The datasets used for evaluation are too simple and do not convincingly demonstrate the model's ability to handle more complex real-world data."", 'The methodology section lacks sufficient detail, especially regarding the architecture of the neural network used in the reverse process.', 'The paper lacks comparisons with other state-of-the-art methods, making it hard to gauge the effectiveness of the proposed approach.', 'The description of the forward and reverse processes is superficial.', 'The paper mentions an ablation study but does not provide sufficient details or results from these studies.', 'The performance metrics used are standard but lack comprehensive comparison with baseline methods.', 'The paper could benefit from a more thorough discussion of limitations and potential negative societal impacts.']",2,2,2,2,Reject | |
| resolution_conditioned_diffusion,The paper proposes a resolution-conditioned diffusion model to generate high-quality samples at various resolutions using a single model. The key contribution is a conditioning mechanism that allows the model to adapt to different resolutions.,"['Can you provide more details on the conditioning mechanism used to adapt to different resolutions?', 'What specific datasets were used in the experiments, and what were the results?', 'How does the computational cost of your approach compare to training multiple models for different resolutions?', 'Can you include more ablation studies to analyze the impact of different components of your model?', 'Can the authors provide more detailed descriptions of the model architecture and training procedure?', 'What are the specific advantages of the proposed model compared to existing methods?', 'Can the authors include comprehensive experimental results and comparisons with baseline models?', 'How does the proposed model handle computational costs and scalability issues?']","['The paper does not adequately address the limitations and potential negative societal impact of the proposed work. More discussion on these aspects is needed.', 'The paper mentions that the method can be computationally expensive and may not work well for very low or very high resolutions. However, these limitations are not empirically validated due to the absence of experimental results.']",False,1,1,1,2,4,"['The idea of generating samples at multiple resolutions with a single model is novel and addresses a practical limitation of existing diffusion models.', 'The proposed conditioning mechanism could offer flexibility in applications requiring variable-resolution outputs.']","['The paper is incomplete, with missing sections such as experimental setup, results, and some references.', 'There is insufficient detail in the methodology section, making it difficult to understand the exact workings of the proposed model.', 'The paper lacks clarity and organization, with some sections missing captions and proper citations.', 'The experimental validation is minimal and does not convincingly demonstrate the effectiveness of the proposed approach.', 'The related work section is missing, making it difficult to place the work within the context of existing research.', 'No discussion on the limitations and potential negative societal impacts of the proposed model.']",2,1,2,2,Reject | |
| noise_schedule_interpolation,"The paper proposes a novel diffusion model architecture for generating 2D data by combining linear and quadratic noise schedules. The approach aims to improve the quality and diversity of generated 2D data. The authors claim improvements in sample quality and diversity, demonstrated through experiments on several 2D datasets.","['Can the authors provide more detailed explanations of the methodology and experimental setup?', 'Can the authors include a more comprehensive analysis of the results and compare them with state-of-the-art methods?', 'Can the authors provide more details on the denoising model architecture and how the linear and quadratic noise schedules are combined?', 'What are the specific results and comparisons with state-of-the-art models?', 'Have you considered the limitations and potential negative societal impacts of your work?', 'What are the specific training and inference times compared to other methods?', 'Can the authors complete the related work section to better position their contributions within the existing literature?']","['The paper does not adequately address the limitations and potential negative societal impact of the work.', 'The scalability and applicability of the model to more complex datasets and real-world tasks remain unclear.']",False,2,2,2,3,4,"['Addresses an important problem in 2D data generation, which has numerous applications in computer vision, robotics, and scientific simulations.', 'The proposed combination of linear and quadratic noise schedules is novel and has the potential to improve the quality of generated data.']","['The methodology section lacks sufficient detail on the denoising model and the aggregation of noise schedules.', 'The results section is underdeveloped and does not provide a comprehensive analysis of the results.', 'Some important citations and references are missing.', 'The paper does not compare the proposed method with state-of-the-art methods rigorously.', 'Incomplete Background section, leaving out important related work and theoretical context.', ""Formatting issues and placeholders (e.g., 'Figure ??') suggest that the paper was rushed and not thoroughly proofread."", 'No discussion on the limitations or potential negative societal impacts of the proposed method.']",2,2,2,2,Reject | |
| noise_schedule_analysis,"The paper investigates the relationship between noise schedules, generated samples, and mode coverage in diffusion models. It proposes a novel approach to designing noise schedules that balance sample quality and diversity. The paper claims to improve mode coverage and sample quality through extensive experiments.","['Can you provide a detailed description of the proposed method for optimizing noise schedules?', 'What datasets and evaluation metrics were used in the experiments?', 'How does your approach compare to existing methods in terms of performance?', 'Can you include more detailed results and analysis to support your claims?', 'Can the authors provide a comprehensive related work section?', 'Can the authors detail the methodology and experimental setup?', 'What specific experiments were conducted, and what were the outcomes?', 'What are the limitations and potential negative societal impacts of this work?']",['The paper does not adequately address its limitations or the potential negative societal impact of its work.'],False,1,1,1,2,4,"['The paper addresses a relevant and important issue in the field of generative models, particularly focusing on the impact of noise schedules on sample quality and mode coverage.', 'The proposed approach to designing noise schedules is novel and aims to balance competing objectives of sample quality and diversity.']","['The methodology section is incomplete, lacking essential details about the proposed approach and how it optimizes noise schedules.', 'The experimental setup section is missing, failing to provide information about datasets, evaluation metrics, and experimental procedures.', 'The results section is not adequately developed, with insufficient analysis and discussion of the findings.', ""The paper's overall clarity and organization are poor, making it difficult to understand the contributions and significance of the work."", 'Key related works are mentioned but not adequately discussed or compared against the proposed method.', 'The paper does not adequately address its limitations or the potential negative societal impact of its work.']",2,1,2,2,Reject | |
| diffusion_model_temporal_adaptation,"The paper proposes a novel approach to adapt diffusion models to changes in data distribution over time. The method aims to leverage the flexibility of diffusion models to learn from new data while preserving knowledge from previous data. However, the paper lacks detailed content in critical sections such as Related Work, Background, Method, Experimental Setup, and Results, making it impossible to evaluate the validity and impact of the proposed approach.","['Please provide detailed content for the missing sections, including a thorough literature review, detailed methodology, experimental setup, and results.', 'How does the proposed method specifically adapt to changes in data distribution over time?', 'What datasets were used for the experiments, and what were the results?', 'How does your approach compare to existing methods in terms of performance?']","['The paper is incomplete, and thus the limitations of the proposed method cannot be assessed.', 'The paper does not discuss potential limitations or negative societal impacts of the proposed approach.']",False,1,1,1,1,5,"['Addresses a relevant and important problem of temporal adaptation in diffusion models.', ""The proposed idea of leveraging diffusion models' flexibility to adapt to new data while preserving previous knowledge is promising.""]","['The paper is incomplete, with major sections such as Related Work, Background, Method, Experimental Setup, Results, and Conclusions missing content.', 'The methodology is not described in detail, and there are no explanations or equations provided.', 'No experimental results are presented, making it impossible to verify the claims.', 'Insufficient references and lack of a thorough literature review.', 'The overall clarity and presentation are poor due to the missing sections.']",1,1,1,2,Reject | |
| diffusion_model_masking,The paper proposes a novel approach to generating diverse and novel samples using diffusion models with a masking mechanism. The idea is to selectively mask input data dimensions during training to encourage the model to produce varied samples.,"['Can the authors provide detailed descriptions and justifications for each section (related work, background, methods, experimental setup, results, and conclusions)?', 'Can concrete experimental results, figures, and tables be included to substantiate the claims made in the paper?', 'How does the proposed method compare to existing methods in terms of performance and diversity of generated samples?', 'Can you provide a detailed description of the masking mechanism and how it is implemented in the diffusion model?', 'What datasets were used for the experiments, and what were the exact experimental setups?', 'Could you provide quantitative results comparing your method with existing approaches to demonstrate its effectiveness?', 'Please include the theoretical analysis mentioned in the abstract.', 'Can you address potential ethical concerns, limitations, and societal impacts of your work?']","['The primary limitation is the overall incompleteness and lack of detailed experimental validation of the proposed approach.', 'The paper does not address potential limitations and negative societal impacts of the proposed method.']",False,1,1,2,2,5,"['The concept of using a masking mechanism in diffusion models to promote diversity is intriguing and potentially impactful.', 'The problem of mode collapse in generative models is significant, and tackling it could have substantial impact.']","['The paper has several incomplete sections, including related work, background, methods, experimental setup, results, and conclusions.', 'There are placeholders instead of actual content in multiple parts of the paper, indicating it is unfinished.', 'The proposed method is not backed by sufficient experimental evidence, as no concrete results, figures, or tables are provided.', 'The novelty claims are not substantiated with detailed comparisons to existing methods.', 'The clarity and organization of the paper are poor, making it difficult to understand the approach and its impact.', 'The theoretical foundation and empirical validation of the proposed method are inadequately discussed.', 'The paper lacks proper citations and references to related work, which is essential for contextualizing the contributions.', 'Ethical concerns, limitations, and potential negative societal impacts are not addressed.']",2,1,1,2,Reject | |
| hybrid_embedding,"The paper introduces a hybrid embedding approach for diffusion models, combining sinusoidal and Fourier features to improve performance on 2D datasets. The goal is to capture high-frequency patterns and relationships in the data more effectively. The paper includes experiments on simple 2D datasets such as circle, dino, line, and moons.","['Can you provide more details about the implementation of the hybrid embedding layer and how the sinusoidal and Fourier features are combined?', 'Have you tested the proposed method on more complex or higher-dimensional datasets? If so, what were the results?', 'Can you provide a comparison with more existing methods to highlight the novelty and advantages of your approach?', 'Can you provide more details on the related work and how your method compares to a broader set of existing techniques?', 'How does the choice of hyperparameters impact the performance of your method?', 'What are the limitations of your approach, and how might it impact society negatively?', 'Can the authors provide a more detailed theoretical justification for the hybrid embedding approach?', 'Can the authors include additional ablation studies to further validate the effectiveness of combining sinusoidal and Fourier features?']","['The paper does not address the limitations of the proposed method in detail. For instance, it mentions that the method may not be suitable for high-dimensional data distributions, but does not provide any further analysis or discussion on this point.', 'Potential negative societal impacts of the method should be considered and discussed.']",False,2,2,2,3,4,"[""The idea of combining sinusoidal and Fourier features for diffusion models is interesting and has the potential to improve the model's ability to capture complex patterns."", 'The paper addresses an important challenge in diffusion models, which is capturing high-frequency patterns and relationships in the data.']","['The novelty of the approach is limited, as combining different types of embeddings is not new.', ""The experiments are limited to simple 2D datasets and do not provide sufficient evidence of the method's generalizability to more complex or higher-dimensional datasets."", 'The paper lacks detailed descriptions of the implementation and experimental setup, making it difficult to fully understand and reproduce the results.', 'The results on simple 2D datasets are not sufficient to demonstrate the effectiveness of the proposed method.', 'The related work section is not well-developed, lacking thorough comparison and discussion of previous methods.', 'There is no discussion of the limitations and potential negative societal impacts of the proposed method.']",2,2,2,2,Reject | |
| diffusion_trajectory_quality,"The paper investigates the impact of noise schedules on sample quality in diffusion models and proposes a novel approach to analyze the diffusion trajectory. However, the paper lacks sufficient details, is poorly organized, and does not provide adequate experimental validation.","['Can you provide a detailed description of the proposed method?', 'What are the quantitative results and how do they compare with existing methods?', 'How does your method improve sample quality compared to traditional approaches?', 'Have you considered any potential ethical implications or limitations of your work?']","['The methodology is not clearly explained, making it difficult to assess its limitations.', 'No discussion on potential negative societal impacts or ethical considerations.']",False,1,1,1,1,5,['Addresses an important and less explored aspect of diffusion models: noise schedules.'],"['The paper is poorly organized and lacks clarity in presenting its methodology and results.', 'No detailed description of the proposed method, making it difficult to understand and reproduce the work.', 'The experimental setup and results are not adequately discussed. There are no quantitative results or comparisons with existing methods.', 'No substantial evidence or theoretical analysis to support its claims.', 'Ethical considerations, limitations, and potential societal impacts are not addressed.']",2,1,1,2,Reject | |
| mode_specific_diffusion,"The paper proposes a mode-specific latent variable denoising diffusion model to improve sample quality and mode coverage in generative models. The approach introduces mode-specific latent variables into the denoising diffusion model framework, showing promising results on various datasets.","['Can the authors provide more details on the methodology, particularly the training process and the architecture of the proposed model?', 'What datasets were used in the experiments, and what are the specific evaluation metrics?', 'How does the proposed model compare with state-of-the-art methods in terms of sample quality and mode coverage?', 'Can you complete the related work section and provide a comprehensive review of related literature?']","['The lack of detailed explanations and clarity in the methodology and experimental setup sections makes it difficult to fully understand and evaluate the proposed model.', 'The paper does not discuss any limitations or potential negative societal impacts of the proposed method. Addressing these points would be beneficial.']",False,1,1,2,2,5,"['The paper addresses a critical challenge in generative modeling: balancing sample quality and mode coverage.', 'The introduction of mode-specific latent variables into the denoising diffusion model is a novel idea.']","['The paper is incomplete, with key sections such as methodology, experimental setup, and results missing or containing placeholders.', 'The related work section is not fully developed, making it difficult to assess the novelty and contextualize the contributions.', ""No experimental results or figures are provided, which prevents any assessment of the model's performance."", 'The clarity and organization of the paper are poor due to the missing content and vague descriptions.']",2,1,1,2,Reject | |
| diffusion_confidence,"The paper proposes a novel diffusion-based approach for generating 2D data, including images and shapes. It leverages diffusion-based image synthesis to iteratively refine a random noise signal to converge to a target data distribution. The paper aims to extend the application of diffusion models to 2D data and demonstrates its potential for generating diverse data types through experiments on various 2D datasets.","['Can the authors provide a detailed explanation of the proposed diffusion-based architecture?', 'How does the approach differ from existing diffusion-based image synthesis methods?', 'Can the authors include more comprehensive experiments and comparisons to state-of-the-art methods?', 'What are the specific results obtained from the experiments, and how do they compare to existing methods?', 'How does the proposed approach specifically improve upon the limitations of current methods like GANs and VAEs?', 'Can you provide a detailed description of the proposed diffusion-based methodology?', 'What is the experimental setup and how were the results obtained?', 'How does this work differ from existing diffusion-based approaches?', 'What are the limitations and potential negative societal impacts of this work?', 'Can you provide more details on the model architecture and training procedure?', 'What datasets were used in the experiments, and what baselines were compared against?', 'Can you include visual examples of the generated 2D data?', 'Can you elaborate on how your approach improves over existing diffusion models?']","['The paper does not adequately address the limitations of the proposed method or its potential negative societal impact.', 'The paper is incomplete, with missing sections and insufficient details, making it challenging to assess limitations and potential negative societal impacts.']",False,1,1,1,2,4,"['The idea of extending diffusion models to generate a broad range of 2D data is interesting and has potential significance.', 'The topic of diffusion models for 2D data generation is of high relevance and interest in the field.', 'The proposed extension of diffusion models to a broader range of 2D data types is a novel idea that could have significant implications.']","['The methodology section is incomplete, with placeholders instead of detailed explanations.', 'The experimental setup and results sections lack depth and critical comparisons to existing methods.', 'There is insufficient clarity on how the proposed approach differs from or improves upon existing diffusion-based methods.', 'The paper is poorly organized, making it difficult to follow and assess the technical contributions.', 'The related work section is minimal and does not adequately situate the proposed work within the context of existing research.', 'The paper is incomplete, with critical sections such as methodology, related work, and detailed results missing.', 'The contributions are vaguely described and lack concrete details.', 'The experimental results are briefly mentioned without sufficient analysis or context.', 'The paper does not discuss its limitations or potential negative societal impacts, which is crucial for evaluating the broader implications of the work.']",2,1,1,2,Reject | |
| geometric_constraints,"The paper proposes a novel diffusion model for generating high-quality 2D data by using a sinusoidal embedding and a denoising network. The model is evaluated on several 2D datasets, demonstrating its ability to generate realistic data samples. The results showcase the effectiveness of the model in terms of training time, evaluation loss, inference time, and KL divergence.","['Can you provide more details on the geometric constraint and its role in the model?', 'How does the proposed method compare to state-of-the-art methods in 2D data generation?', 'Can you clarify the architecture of the denoising network and its training process?', 'Can the authors provide more details on the sinusoidal embedding and its theoretical justification?', 'What are the specific configurations and parameters used in the experiments?', 'How does the proposed model compare with other state-of-the-art methods in more detail?', 'How does your proposed method significantly differ from existing diffusion models?', 'Can you provide more detailed theoretical analysis or experimental validation to support your claims?', 'What specific improvements does your model offer over existing methods in 2D data generation?', 'Can the authors include more qualitative results and visualizations to support their claims?']","['The model assumes data continuity and boundedness, which might not hold for all datasets. This limitation is not adequately discussed.', 'The paper does not provide sufficient ablation studies to justify the choice of sinusoidal embedding and other architectural components.', 'The paper assumes that the data is continuous and can be represented by a probability density function, which may not hold for all datasets. This limitation is acknowledged but not sufficiently addressed.', 'The paper does not discuss potential negative societal impacts or ethical concerns related to the generation of synthetic data.', 'The assumption that data is continuous and can be represented by a probability density function may not hold for all datasets.', 'The lack of detailed ablation studies makes it unclear which components of the proposed model contribute most to its performance.', 'The paper does not adequately address the limitations of the proposed method, especially in terms of the simplicity of datasets used.', 'Potential negative societal impacts are not discussed.']",False,2,2,2,3,4,"['Addresses the significant problem of generating 2D data, which has unique challenges.', 'The use of sinusoidal embeddings and a denoising network is an interesting adaptation of existing methods.']","['The paper lacks detailed explanation and motivation for key components, such as the geometric constraint and the specific architecture of the denoising network. For example, the choice of an ellipse with a major axis of 1.5 and a minor axis of 0.5 is not justified.', 'Figures and results are not well-explained. For instance, Figure 1 and Figure 3 lack captions, and the results in Table 1 are not thoroughly discussed. This makes it difficult to understand the significance of the findings.', 'The evaluation section does not provide a comprehensive comparison with state-of-the-art methods, which is essential to establish the significance of the proposed approach. Including comparisons with methods like GANs or VAEs would strengthen the evaluation.', 'The assumptions made by the model (e.g., data continuity and boundedness) are not thoroughly justified or discussed in the context of practical applications. This limits the generalizability of the approach.', 'The originality of the contributions is questionable given the existing literature on diffusion models and their applications. The proposed approach does not significantly advance the state-of-the-art.', 'There is insufficient theoretical justification for the use of sinusoidal embeddings and the specific architecture of the denoising network. More detailed theoretical analysis is needed.', 'The experimental setup is not described in sufficient detail to ensure reproducibility. Key parameters and configurations, such as the noise schedule and the architecture of the residual blocks, are missing.', 'The results section is relatively sparse, and the authors do not provide a thorough comparison with other state-of-the-art methods. More detailed quantitative and qualitative comparisons are needed.', 'Lacks novelty and relies heavily on existing techniques without significant innovation.', 'Insufficient theoretical analysis and experimental validation to support the claims.', 'Poor clarity and organization; missing captions and detailed descriptions hinder understanding.', 'Does not convincingly demonstrate significant improvements over existing methods.']",2,2,2,2,Reject | |
| dataset_agnostic_diffusion,The paper proposes a method for adapting diffusion models to new datasets through fine-tuning. The aim is to enable diffusion models trained on a source dataset to adapt to the distribution of a target dataset with minimal fine-tuning. The paper demonstrates its approach through experiments on various simplistic datasets.,"['Can you provide more details on the experimental setup and how the results were obtained?', 'How does your method compare to existing methods in the literature?', 'Can you provide more explanations and background on the fine-tuning strategy used?', 'What specific metrics were used to evaluate the performance of the adapted models?', 'Can the authors provide more comprehensive results, including experiments on more complex and realistic datasets?']","['The paper lacks detailed explanations and background information, making it difficult to assess its contributions accurately.', 'The experimental results are not well-explained, and there is no comparison with existing methods.', 'The paper does not adequately address the limitations of the proposed method.', 'There is no discussion on the potential negative societal impact of the proposed approach.']",False,2,1,2,2,4,"['The problem of adapting diffusion models to new datasets is relevant and important.', 'The idea of fine-tuning diffusion models to adapt to new datasets is novel and has practical implications.']","['The paper is not well-structured and lacks thorough explanations in multiple sections.', 'The related work section is missing, making it difficult to understand how this work fits within existing research.', 'The experimental setup and results are not described in detail, making it challenging to evaluate the effectiveness of the proposed method.', 'There is a lack of clarity in the presentation, and the significance of the results is not well-explained.', 'The methodology is not clearly described, and there is a lack of detailed explanation of the novel strategy for fine-tuning.', ""The experimental results are based on simplistic datasets, which do not convincingly demonstrate the model's ability to generalize to more complex and realistic datasets."", 'There is no discussion on the limitations or potential negative societal impacts of the proposed approach.']",2,2,1,2,Reject | |
| diffusion_trajectory_control,"The paper proposes a novel method to control the diffusion trajectory in diffusion models to enhance sample quality and diversity. The approach introduces a control term to guide the diffusion process, aiming to generate better samples. However, the paper is incomplete, lacking detailed sections on the method, experimental setup, results, related work, and background.","['Can you provide a detailed description of the proposed method, including mathematical formulations and implementation details?', 'What are the specific experimental setups and datasets used to validate your approach?', 'How does your method compare to existing techniques in terms of performance metrics?', 'Can you include a comprehensive review of related work to contextualize your contributions?', 'What are the potential limitations of the proposed method?', 'Can the authors discuss any potential negative societal impacts of their work?']","['The paper is currently incomplete, which is a significant limitation. The authors need to provide a complete and detailed description of their approach, related work, and experimental results to allow for a thorough evaluation.', 'The paper does not discuss any limitations or potential negative societal impacts. Addressing these aspects would provide a more balanced and thorough evaluation of the proposed method.']",False,1,1,1,2,4,"['Addresses a relevant and important problem in diffusion models.', 'Proposes a novel approach that could potentially improve sample quality and diversity.']","['The paper is incomplete, with critical sections such as the method description, experimental setup, and results missing or placeholders.', 'The lack of related work and background information makes it difficult to understand the context and novelty of the proposed approach.', 'Without detailed experimental results, it is impossible to evaluate the effectiveness of the proposed method.', 'The clarity and organization of the paper are severely compromised due to the incomplete sections.']",2,1,1,2,Reject | |
| noise_schedule_symmetry,The paper aims to explore the effectiveness of Denoising Diffusion Probabilistic Models (DDPMs) in generating high-quality samples and analyzes the impact of various hyperparameters. The study includes a comparison between symmetric and asymmetric noise schedules.,"['Please provide the complete content for the sections: introduction, related work, background, method, experimental setup, and results.', 'Can you include a thorough discussion of the theoretical foundations and practical implications of your proposed methods?', 'Please provide detailed experimental results and analyses to support your claims.']","['The primary limitation is the incomplete nature of the paper, which prevents a thorough review of its contributions and significance.']",False,1,1,2,2,4,"['The topic is relevant and timely, focusing on the performance of DDPMs, which are a popular class of generative models.']","['The paper is notably incomplete, with several sections such as the introduction, related work, background, method, experimental setup, and results missing content.', 'Without the actual content, it is impossible to evaluate the originality, quality, and significance of the work.', 'There is no comprehensive discussion of the theoretical foundations or practical implications of the proposed methods.', 'The experimental results are mentioned in the abstract but are not supported by detailed discussions or analyses in the main text.']",2,1,1,2,Reject | |
| symmetry_invariant_loss,"The paper introduces a novel regularization technique for diffusion models using symmetry-invariant loss functions, specifically focusing on rotational symmetry. The goal is to improve sample quality and reduce training steps by promoting rotation-invariant sample generation through a rotational symmetry-invariant mean squared error loss function.","['Can you provide detailed experimental results and proper evaluation metrics to support your claims?', 'What are the specific contributions of your work relative to existing diffusion models and regularization techniques?', 'Can you clarify the mathematical formulation of your symmetry-invariant loss function and its implementation details?', 'How does the proposed method compare quantitatively and qualitatively with other existing methods?', 'What are the architectural details and hyperparameters used in the diffusion models?', 'Can the authors provide a detailed related work section to contextualize their contributions?', 'What are the limitations of the proposed method, and how might they be addressed in future work?', 'Are there any potential negative societal impacts or ethical considerations associated with your work?']","['The paper does not adequately discuss potential limitations of the proposed approach, such as its applicability to datasets or domains where rotational symmetry is not a natural property.', 'The paper does not address the limitations of the proposed method or provide a discussion on potential negative societal impacts or ethical considerations.']",False,1,1,2,2,4,"['The idea of using symmetry-invariant loss functions to address mode collapse in diffusion models is novel and potentially impactful.', 'Placing an emphasis on rotational symmetry could be beneficial in applications where rotational invariance is desired.']","['The paper suffers from lack of clarity and detailed explanations in several key sections, including the method section and experimental setup.', 'Important references are missing, denoted by placeholders (e.g., ?), which undermines the credibility and completeness of the literature review.', ""The experimental results section is absent ('RESULTS HERE'), making it impossible to verify the claims made about improved sample quality and reduced training steps."", 'The background and related work sections are very general and do not adequately contextualize the proposed method within existing research.', 'There is no theoretical justification or in-depth discussion explaining why the proposed loss function should theoretically address mode collapse better than existing methods.', 'The paper lacks sufficient detail for reproducibility. Key parameters and architectural details of the diffusion models used are not provided.', ""The conclusion section is also missing, reducing the paper's overall coherence and leaving the reader without a summary of findings and contributions."", 'There is no discussion of the limitations and potential negative societal impacts of the work.']",2,1,2,2,Reject | |