paper_id stringlengths 10 10 | items listlengths 4 82 | gt float64 1 9.5 |
|---|---|---|
VgtpRXhxli | [
"strength: This paper studies an interesting and important problem, as realizing the Pareto front is indeed important for any practical problems.",
"strength: This paper is theoretically rigorous. The main techniques to achieve the goal, such as reduction to focus on invertible representations and solve the optim... | 6 |
gx1wHnf5Vp | [
"strength: The paper is well written and easy to follow.",
"strength: The idea of drop-upcycling is interesting although not new (see weaknesses section).",
"strength: Extensive experiments and release of all the relevant materials for reproducibility.",
"strength: The empirical evaluation section and the num... | 6.25 |
2IoFFexvuw | [
"strength: The problem of finetuning conditional flow matching models is of general interest to the community. How to preserve the generation diversity and avoid model collapse is a challenging problem.",
"strength: Combining the reward-weighted matching loss and the Wasserstein distance regulatization seems to b... | 6 |
WVIq7jYIda | [
"strength: The core idea of combining kernel rank reduced regression and manifold framework is convincing and make sense, and the introduced concept of Kendall shape space is interesting too.",
"strength: Also good to see some related works been discussed in section 2. Closed-form solution is nice too and this ad... | 3 |
KJzwUyryyl | [
"strength: This paper is well-written and easy to follow.",
"strength: By focusing on safety-relevant, non-objective questions, the benchmark differentiates itself well from existing work on interpretability evaluations.",
"strength: The focus on distribution shift also makes the evaluation more realistic than ... | 5 |
fBkdjUnymd | [
"strength: The method proposed in this paper is successfully applied in a white-box + black box framework, and shows its effectiveness in reasoning, planning, and personalization tasks compared with black-box LLMs.",
"strength: The paper is well-written and easy to follow.",
"strength: The proposed method is si... | 5 |
zavLQJ1XjB | [
"strength: This paper is well motivated. The studied point is interesting. It is empirically found that temperature scaling is good for calibrating deep models. However, as pointed out in this paper, it may harm calibration under some conditions.",
"strength: The empirical study shows very supportive results for ... | 6.666667 |
Bb1ddVX8rL | [
"strength: The Legendre-KAN achieves lower test set losses on a set of symbolic expressions compared to the standard KAN.",
"strength: The authors provide an extensive overview of the method.",
"strength: The shift from B-splines to Legendre polynomials appears well-motivated, and the results convincingly show ... | 3.5 |
9I6UOIfbwf | [
"strength: Innovative Data Generation Pipeline: The authors designed a comprehensive pipeline for generating a synthetic dataset specifically for model training in video face re-aging. This pipeline addresses the challenge of obtaining paired video data with consistent identities and varying ages, thereby enhancing... | 3.5 |
GqGoa44obw | [
"strength: A novel algorithm RLHF-IMAF is proposed.",
"strength: Inconsistent feedback for RLHF is indeed a topic that is understudied and worth more attention due to the nature of human feedback.",
"strength: The problem is well motivated.",
"strength: The reasoning behind proposed ideas are well described a... | 4.5 |
SQrHpTllXa | [
"strength: The model achieves the state-of-the-art performance on the WikiTQ, FetaQA and WikiSQL using orders of magnitude fewer parameters when compared of LLMs such as Codex.",
"strength: Furthermore, the model is more robust on the perturbation than other counterparts. Also for large tables, the CABINET also s... | 8 |
owEQ0FTfVj | [
"strength: **Having a clear as well as straightforward motivation and contribution**. The benchmark is highly relevant for advancing glycan-related machine learning research, offering a platform to assess various models on tasks that are both scientifically and practically significant in fields like immunology and ... | 6.666667 |
2NwHLAffZZ | [
"strength: Understanding the behaviour of overparameterized neural networks is a very important and interesting question.",
"strength: The overall problem is interesting, and the local structure of the paper is reasonable (no typos, definitions are always provided, etc.). Any substantial contribution to understan... | 2.333333 |
u6vC7KaFel | [
"strength: **1. Flexibility and efficiency with minimal overhead** -- The main strength of the method is that it allows for good generalization to new tasks without adding much extra in terms of training. It is also very flexible because it can adapt to unseen prompts during test time. This can help make LLMs a lot... | 4.75 |
GZYCCQacwH | [
"strength: Compared to previous learning algorithms for solving bilevel optimization, the proposed algorithm is developed under weaker assumptions of upper- and lower-level objectives, and hessian-free, giving an advantage in communication and computational efficiency in Federated Learning.",
"strength: The theor... | 4.75 |
mMh4W72Hhe | [
"strength: The paper is sound and the topic of the paper is of high interest to the research community.",
"strength: The paper does a good job at introducing technical details and makes the paper easy to follow.",
"strength: As far as I know, the paper is the first to provide ways to efficiently reduce the bran... | 6 |
TWVMVPx2wO | [
"strength: Although the proposed method is simple, the proposed idea of introducing visual prompt tuning to improve proxy-based deep metric learning is well motivated and valid.",
"strength: The contribution is solid. As far as I know, this may be the first work to explore the parameter-efficient (PEFT) methods s... | 6.333333 |
AcVpLS86RT | [
"strength: The paper tackles a practical issue in autonomous driving, where dataset diversity can restrict the ability to scale training. By addressing how to integrate datasets with different characteristics, the research directly benefits the community by opening up avenues for broader dataset utilization, potent... | 7 |
lf8QQ2KMgv | [
"strength: This is good science, the paper doesn't go into a lot of depth other than the experiments it does and the reasoning for them, but it paints a complete picture, and rectifies past work.",
"strength: The paper is well-structured, offering a clear foundational context for the issues tackled.",
"strength... | 3.75 |
aMBSY2ebPw | [
"strength: wide coverage of materials, models and approaches",
"strength: presented clearly (except for the complexity of results in Table 2) and convincingly",
"strength: The paper showed a novel study that whether grammar books alone support translation for extremely low-resource (XLR) languages or not. By is... | 7.333333 |
cJs4oE4m9Q | [
"strength: Originality: The paper presents a problem that may lead to suboptimal performance in the field of anomaly detection and provides a solution, offering a novel approach to enhance the performance of anomaly detection algorithms.",
"strength: Quality: The research is of high quality, marked by a well-stru... | 8 |
EBT0oymkZb | [
"strength: The paper tackles an important problem, generalization in RL with offline datasets, and takes a theoretical perspective. From a look, the derivations seems sound and generally, enough detail is included to understand.",
"strength: Also, although the paper has a theoretical focus, it is nice to see some... | 5.5 |
h71cSd2loX | [
"strength: Overall, I like the paper’s motivation and proposed approach: DPO is very prevalent method to perform preference optimization, and the inability to handle ties is a definitely a drawback.",
"strength: The proposed variations are well motivated and the theoretical discussion on their differences was qui... | 5.5 |
qgsXsqahMq | [
"strength: As reported in the experiments, the combination of structural information and a mixture of calibrators trained on different sets of features seems to clearly outperform previously proposed calibration techniques for node classifiers.",
"strength: This combination of ideas is well motivated by the paper... | 7.5 |
hKMPz3wkPV | [
"strength: I think this is a generally well-written and thought-provoking paper. Compositionality and compositional generalization is an important topic across a number of different fields and I agree that a better formal understanding of what we mean by it is important. The authors present a conceptual framework t... | 6.75 |
n7iwmPacDt | [
"strength: I think the speedups reported are quite impressive.",
"strength: The limitations section is very well written and highlights the problems with implementation.",
"strength: The empirical results are strong. The paper conducted experiments on 3 different models under a variety of settings, and showed 3... | 3 |
sr0My6yDNu | [
"strength: The author proposed a novel gradient projection technique to maintain the adversarial robustness in a continuous learning setting.",
"strength: Experimental results on several benchmarks demonstrate the superiority of DGP over existing approaches.",
"strength: It would be useful in the scenario that ... | 3.25 |
TeVAZXr3yv | [
"strength: The topic of this paper is very important — currently, most audio LLMs are benchmarked using different evals, and researchers must test other models for comparison, which is tedious and impractical for many. A unified benchmark is common in text and vision, such as MMLU, MMMU, and DocVQA, and has already... | 7.5 |
zkNCWtw2fd | [
"strength: Addresses a relevant challenge in multilingual information retrieval.",
"strength: Provides comprehensive experimental validation across multiple benchmark datasets (XQuAD-R, MLQA-R, MIRACL).",
"strength: The paper shows that two standard batching strategies are complementary for information retrieva... | 3 |
Cy5v64DqEF | [
"strength: This paper is the first to study the relation between the idempotence and perceptual image compression methods.",
"strength: The findings are very interesting and motivating, the high correlation between the idempotence and perceptual image compression may benefit both above compression areas.",
"str... | 7.5 |
E8gYIrbP00 | [
"strength: The paper is well structured and presented and is very clear and easy to follow and read.",
"strength: The paper clearly shows the issue related to relying on a high correlation between human and machine-generated outputs, giving cases where the uncertainty in human annotations is high; these correlati... | 6.75 |
UUwrBhhsxT | [
"strength: The paper is pretty novel to leverage decision transformers to model thescenario generation problems. The integration of context-aware scenario generation improves the model's generalization capabilities, allowing it to adapt better to novel situations. The authors also demonstrate the scalability of the... | 5.25 |
kJFIH23hXb | [
"strength: The paper introduces a novel approach called FoldFlow for generative modeling of protein structures. This approach is based on the flow-matching paradigm over rigid motions. The authors claim that incorporation of Riemannian optimal transport and simulation-free training techniques improves the stability... | 8 |
VLFhbOCz5D | [
"strength: Efficient Jacobian-Vector Product Calculation: The authors demonstrate that the Jacobian-Vector Product resulting from linearization can be efficiently computed in a single forward pass. This reduces the training and inference costs of linearized transformers to a similar order of magnitude as the origin... | 6 |
2LhCPowI6i | [
"strength: There is ample contextualization and motivation of the work.",
"strength: The experiments across architectures are thorough and represent an interesting set of architectures.",
"strength: There is some good discussion of generalization and why the method seems to hurt at lower numbers of tasks.",
"... | 2.333333 |
urcEYsZOBz | [
"strength: This paper's strength lies in demonstrating that data assimilation in the latent space can effectively solve the problem of vanishing gradients.",
"strength: The Latent-EnSF method presented in this paper addresses key challenges in data assimilation for high-dimensional systems with sparse observation... | 6.166667 |
LnN5UdhcjT | [
"strength: The paper is clearly written and easy to follow.",
"strength: The proposed design of the region-wise motion controller, along with the training strategy used to build the training data, appears well-thought-out. The provided examples demonstrate high-quality results compared to baseline methods.",
"s... | 5 |
w6nlcS8Kkn | [
"strength: CoT is a popular and often over-used strategy in LLM based applications. This paper dives deep into where it actually is useful and where it isn't - contributing to a better overall understanding of the technique and lay foundations for future improvements.",
"strength: The paper is well-written, detai... | 6.666667 |
W48CPXEpXR | [
"strength: The motivation behind this work is interesting and is a necessary step toward better leveraging LLMs.",
"strength: The evaluation is extensive and includes several datasets across three different types of reasoning.",
"strength: Interesting Premise: The idea of reframing hallucinations, typically vie... | 5 |
SUEXRbzq9l | [
"strength: This paper addresses an important problem by attempting to develop an FPTAS for estimating statistical similarity between product distributions, which is a #P-hard problem.",
"strength: The introduction of the multidimensional #MASKEDKNAPSACK problem looks like a novel extension of the classic knapsack... | 4.6 |
KgKN7F0PyQ | [
"strength: Solving complex real-world planning tasks is very important and relevant to AI research.",
"strength: This paper incorporates the idea of tree search into the ReAct algorithm and achieves improved performance.",
"strength: Decomposing a complex task into a tree structure makes a lot of sense to me. H... | 4.5 |
XMaPp8CIXq | [
"strength: The experiments encompass a variety of datasets, such as CIFAR-10, CIFAR-100, and ImageNet, while utilizing different backbone architectures like ResNet, ViT, and VGG. Furthermore, they conduct extensive comparisons with numerous baseline methods.",
"strength: The approach for search of pruned/active c... | 3 |
g1fE8fSOm5 | [
"strength: Several theoretical results about two-layer GCNs are obtained to compare with CNNs.",
"strength: The whole paper is well-structured and problem is well formulated, especially both the training phases and generalization ability are discussed.",
"strength: Detailed and quantitative explanation are enta... | 5 |
a7eIuzEh2R | [
"strength: The paper does a good work of curating a benchmark for text games based navigation and mapping.",
"strength: The metrics of evaluation for two identified tasks, namely DF and RF questions (destination, route) gives an head start to reuse existing datasets posing them for a different problem.",
"stren... | 6.25 |
MLShfiJ3CB | [
"strength: This paper is easy to understand.",
"strength: The article observes the use of different optimizers for training Convolutional Neural Networks (CNNs) and ViTs, suggesting a potential overstatement of ViTs' vulnerability to attacks with defense.",
"strength: The CAT attack seems effective in attacking... | 4.5 |
BL4WBIfyrz | [
"strength: LiMAC combines a small Action Transformer (AcT) with a fine-tuned vision-language model (VLM). This hybrid approach is tailored to the computational constraints of mobile devices, achieving efficient and accurate control without relying on large, resource-intensive models. The AcT independently handles c... | 7.5 |
gye2U9uNXx | [
"strength: The paper is well-written and experimental design choices are remarkably well-documented, which facilitates reproducibility experiments and future work to easily extend this work to different settings.",
"strength: The thesaurus based method is interesting and creative",
"strength: I find the author’... | 7.5 |
p4eG8rCa0b | [
"strength: The paper is well-written and easy to follow",
"strength: The proposed DDT and soft guidance modules are novel and effective, demonstrated by both qualitative and quantitative results",
"strength: Evaluations are comprehensive and showcase better results than existing SOTA methods",
"strength: The ... | 6.666667 |
kGvXIlIVLM | [
"strength: The problem of sampling schemes for autoregressive models is fundamental and should be explored.",
"strength: Detailed illustrations are provided. For example, the authors review preliminaries, give formal derivation, and also the practical implementation.",
"strength: It is interesting to see that C... | 7 |
LS1VuhkReU | [
"strength: The paper is well-written, clear and easy to read with methodologies explained well and experimental set-up very clear",
"strength: The analysis contains broad coverage over six different prompt recovery methodologies",
"strength: The provided qualitative analysis is very thorough for the given examp... | 3 |
xriJVaTh4C | [
"strength: The paradox of certified training is an important subject to study.",
"strength: The proposed method (GLS) is well-motivated and backed up by theoretical results.",
"strength: The empirical results on small perturbation settings in Table 3 are quite impressive.",
"strength: GLS introduces novelty i... | 3.333333 |
TyFrPOKYXw | [
"strength: With safety being an important aspect in LLMs, this paper tackles an important question -- how to do value alignment under both the safety and usefulness axes.",
"strength: The paper is well written and explains the methodology involved clearly.",
"strength: Even if the techniques to accommodate safe... | 7.5 |
cA8iQJFioL | [
"strength: This paper appears to be one of the first studies (to the best of my knowledge) focusing on the analysis of citation contexts and intents within philosophical documents.",
"strength: The dataset curated and utilized in this study offers a resource for philosophers, enabling them to comprehend the inter... | 2.5 |
VTG68CNUCf | [
"strength: Curation of prompt dataset for fine-tuning the model is quite interesting",
"strength: Novelty: The paper presents a unique approach to APE by focusing on prompt complementation rather than rewriting. This potentially preserves the user's original intent, a key advantage over rewriting-based methods li... | 4 |
4qFIkOhq24 | [
"strength: The theory is presented clearly: the assumptions are presented well and the theorems explained nicely.",
"strength: The potential impact of the work is quite large: this work presents fundamental limits on the ability of models to be correctly aligned. If current trends continue and large models contin... | 6.5 |
jZw0CWXuDc | [
"strength: The authors of this paper have really found one of the central problems in TDA methods and addressed them in a neat way.",
"strength: The paper provides sufficient background for gradient-based TDA methods and is really well-written.",
"strength: The key trick seems to be taking the projection operat... | 5.5 |
pE6gWrASQm | [
"strength: The paper addresses the relevant problem of the generalization of the acquired robustness to unseen classes, examples, and tasks.",
"strength: The paper is well-written and easy to grasp.",
"strength: The experimental protocol is well described and allows to reproduce the study.",
"strength: The pa... | 6.5 |
qPx3i9sMxv | [
"strength: Innovative Approach: The paper proposes a unique one-stage generation framework that integrates multimodal inputs (text and images) with azimuth-aware guidance, enabling controllable stereo audio synthesis.",
"strength: Significant Dataset Contribution: The BEWO-1M dataset is a notable contribution, pr... | 7.5 |
wXIncJRlK0 | [
"strength: After applyinig the bound operation, the algorithm is empirically observed to converge faster and achieve higher scores.",
"strength: The clip operation is easy to implement.",
"strength: The originality of the paper is one of its strengths. Although the technique of bounding the log density term in ... | 5.75 |
LSq9ef8ANs | [
"strength: The idea of adaptively sampling video frames is well motivated and presented. It has practical values for large multimodal models when processing extrememly long videos.",
"strength: The paper is well written and implementation details are sufficently presented.",
"strength: The language prompt for e... | 4.6 |
kNpSUN0uCc | [
"strength: The paper delivers rigorous theoretical results.",
"strength: The approach is novel to the best of my knowledge.",
"strength: The paper is overall well organized, making it relatively easy to follow.",
"strength: The work is well motivated and the introduction / background are concise yet provide e... | 7.333333 |
6PGT9OJX5N | [
"strength: Introduces an innovative NLI-Score for identifying noisy samples by leveraging neighboring sample consistency.",
"strength: Combines feature and label propagation effectively to reduce selection bias in noisy data scenarios.",
"strength: It proposes a novel two-stage noisy data pruning method.",
"s... | 3 |
OuYCW4ACDl | [
"strength: Structured Learner-Expert Framework: GSIIRL´s learner-expert setup, where the learner infers the expert´s reward and adjusts its own policy, provides a unique structure within multi-agent IRL. This interaction-focused approach enables the learner to balance its intrinsic rewards with its understanding of... | 4.5 |
5zjsZiYEnr | [
"strength: Research problem is challenging and necessary for this direction and various domain applications.",
"strength: A new dataset is proposed for multimodal long document understanding.",
"strength: A LLM-based auto-evaluating framework is introduced.",
"strength: Useful new eval dataset, training datas... | 6 |
ff2V3UR9sC | [
"strength: The paper is well-motivated, highlighting the importance of leveraging user behavior data for content-understanding tasks.",
"strength: It contributes a large-scale dataset BLIFT, consisting of 400k images and 330k videos, along with their receiver behavior, which is beneficial for research about visio... | 6.75 |
3OyaXFQuDl | [
"strength: Introduces a novel compute-matched sampling framework with clear mathematical foundations",
"strength: Proposes a new \"weak-to-strong improvement\" training paradigm that challenges conventional wisdom",
"strength: Provides a fresh perspective on the compute-quality trade-off in synthetic data gener... | 7 |
LxCPyLREX5 | [
"strength: The paper studied a relevant problem in federated learning: a special type of data heterogeneity with client label shift.",
"strength: The paper is generally well-written and easy to follow.",
"strength: This paper pioneers the exploration of label shift challenges within federated learning, introduc... | 4.666667 |
Dr4qD9bzZd | [
"strength: The paper is easy to read.",
"strength: The reported performance is good.",
"strength: The method is simple.",
"strength: The author(s) proposed a new architecture for protein design w/ detailed introduction.",
"strength: The curated datasets have good domain significance for studying computation... | 4 |
wixDdL0vj8 | [
"strength: Novel Approach: The proposed method introduces a parallel framework for collaborative data optimization, addressing the high time complexity associated with sequential optimization for the first time.",
"strength: Efficiency: Experimental results demonstrate that the CoOpt method significantly outperfo... | 3.8 |
GlgD9o9bl4 | [
"strength: The proposed method is technically sound in general.",
"strength: The proposed approach could outperform baselines in general, and it has a strong performance in the cold-start scenarios.",
"strength: The presentation is clear in general.",
"strength: The paper introduces attention-aware augmentati... | 3.5 |
AjunxrcKa2 | [
"strength: The presented qualitative results are on par or slightly better than baselines.",
"strength: Originality: The idea of using conditional diffusion models to synthesize LoRA parameters for fine-tuning is novel, providing an innovative method for model adaptation. To the best of my knowledge, this is the ... | 3.4 |
DrhZneqz4n | [
"strength: The proposed method is the first work that applies diffusion to single motion learning.",
"strength: The paper is well-written and details are clearly presented.",
"strength: The methodology is well-designed in the way to reduce the range of receptive field to prevent acquiring global context and ove... | 7.5 |
0yTf37PXcH | [
"strength: The paper is well written and structured",
"strength: The identified problem of the lack of strong visual capabilities (e.g. detection, localization, color-comprehension etc.) in current vision language models is interesting and worth studying",
"strength: It's also interesting to see the need for mo... | 5.4 |
ffcHGwb4KF | [
"strength: The authors' proposed method, SPADE, is straightforward to understand and makes use of recent innovations in related bodies of work (e.g., sparsity solvers) in a clever way. The authors introduce SPADE as a general and customizable approach, and clearly outline different design decisions that one could ... | 5 |
pcnq7fZs4t | [
"strength: The proposed method is innovative in using a dual dictionary approach to improve class prediction accuracy.",
"strength: The paper is well-structured, making it relatively easy to follow the methodology and experimental setup.",
"strength: The dual-layer embedding mechanism is introduced to improve t... | 3 |
XrsOu4KgDE | [
"strength: This paper selects a good research problem, which is to understand how the frequencies of certain culture related concepts or entities in the pre-training data influence the model performance, in particular from the perspective of generalization and memorization.",
"strength: The high-level ideas to di... | 7 |
qTlcbLSm4p | [
"strength: For originality, this paper first combines a low-resolution ordinary diffusion model with a high-resolution blurring diffusion model. To some extent, this solves the problem of noise schedule in high-resolution diffusion. Besides, the analysis of the noise gap between low-resolution and high-resolution i... | 7 |
kFsWpSxkFz | [
"strength: Design is robust",
"strength: Procedural generation system clearly enables extensive domain generation",
"strength: Hierarchical layout generation, scalable obstacle retrieval, and cohabitant populating are convincingly argued to be crucial and effective for generating diverse scenes",
"strength: B... | 7.5 |
H3UayAQWoE | [
"strength: The work presents a strong motivation for the research problem along with a solid case for psychometrics as \"the how\", gives a compelling case for the selected scales for the benchmark, and ultimately interesting empirical results testing several modern LLMs. In my opinion, this is an under-explored ar... | 6.666667 |
opSPgPIwAD | [
"strength: The most interesting thing I found about this paper is the application of theoretical machine learning into the recourse problem. This is definitely something new that I have not seen in previous papers on recourse generation",
"strength: The paper builds up on existing recourse literature and provide ... | 3 |
dGH4kHFKFj | [
"strength: I really like this paper.",
"strength: I think it addresses an important, general problem by bringing together a bunch of relevant ideas in a well-justified way: using generative models for discriminative problems, physically-inspired regularization, etc.",
"strength: The core insights are simple and... | 6.75 |
TmAmuMXkFc | [
"strength: The study is novel and the presentation is good.",
"strength: The observed losing dimensions phenomenon during memorization is very meaningful and opens many interesting research questions.",
"strength: I found sections 5.2 and 5.3 enjoyable to read.",
"strength: The effective sample size going fro... | 4.25 |
nclyFUZpX9 | [
"strength: The proposed method is meaningful for extending SSM to multivariate time series modeling, especially in the context that most current models including the standard SSM only have univariate modeling capabilities and channel independence performs better.",
"strength: The problem of modeling high-dimensio... | 4 |
EjIKerYk1O | [
"strength: The paper tackles an important problem in aircraft traffic management and control.",
"strength: Clearly, vision-based automated schemes for detecting, identifying, and tracking air traffic in and around an airport is of immense value.",
"strength: The paper cogently argues the need for such a system.... | 2.333333 |
2dnO3LLiJ1 | [
"strength: The paper identifies an interesting phenomenon observed in the popular transformer models (DINO). By removing this artifact, the authors demonstrates the improved models have clear attention maps that could be used for downstream analysis such as object localization.",
"strength: The step-by-step inves... | 8 |
an3kPpce6b | [
"strength: This work focuses on the challenging and poorly explored problem of generating 3D molecules from the low-data domains. The proposed model shows promising results across several tasks.",
"strength: The authors provide comprehensive analysis on the generation results.",
"strength: The authors also prop... | 5.25 |
MVe2dnWPCu | [
"strength: The probabalistic models introduced are simple and make sense. Using such modeling to avoid the expensive counterfactual of evaluating a path is a logical approach.",
"strength: The paper is well written and I found the appendix helpful.",
"strength: The evaluations seem consistent with prior works"... | 7.5 |
cTR17xl89h | [
"strength: Overall, the paper proposes several sound components that address their proposed problems (reward mismatch and using morphology information).",
"strength: Modifying the training objective for the design policy makes intuitive sense, at leas twith how the author's describe it.",
"strength: Self-attent... | 7.5 |
tp1QiTH2aa | [
"strength: This paper addresses the critical challenge of generating synthetic data to enhance large language models (LLMs) in a cost-effective manner.",
"strength: The proposed algorithm, QWICK, is technically robust, employing a multi-armed bandit formulation with budget constraints.",
"strength: QWICK’s desi... | 4.5 |
FR8mMMiu2L | [
"strength: The paper is written clearly, and the math is rigorous and easy to follow.",
"strength: The results show that in distortion metrics the approximated mean does better than the baselines (especially DPS).",
"strength: Paper is written very clearly and is easy to read.",
"strength: Presenting a toy ex... | 4.25 |
wfLuiDjQ0u | [
"strength: This is a very straight-forward paper, it is easy to read, it presents compelling, though not jaw-dropping results.",
"strength: This method does appear to slightly improve the performance of text embedding models even at the 7B scale.",
"strength: Novelty: In-context learning has been shown to be ef... | 7 |
bWcnvZ3qMb | [
"strength: The experiment result is surprisingly good considering the tiny footprint of the model. The standard deviation of the error is very small which may indicate that the model is very stable because of its simplicity.",
"strength: Authors provide comprehensive ablation analysis to show light-weightness of ... | 8 |
1V1QQYARmd | [
"strength: The application of the k-NN radius to differentiate OOD (Out-Of-Distribution) data seems like a novel idea to me.",
"strength: In the theoretical part, I read the statements and found the results they proved to be reasonable.",
"strength: The authors compare their method with many baselines, and impr... | 5.5 |
aTFPO9FHL3 | [
"strength: The paper is clearly written and easy to follow.",
"strength: The results of different downstream task prove the validity and superiority of model, especially the results on large scale datasets.",
"strength: This paper presents a graph transformer method on the dynamic graph, with enough experiments... | 4.25 |
63eIAvrWk4 | [
"strength: This research pioneers a new direction in defense strategies for ITR, highlighting the inadequacies of conventional image classification defense methods.",
"strength: The introduction of multimodal adversarial training significantly improves the robustness of ITR systems.",
"strength: This study offe... | 4.5 |
i28ZjVxl81 | [
"strength: Multiple Dataset and Models",
"strength: Impressing Experiments and Results",
"strength: The manuscript tackles an important challenge in machine learning by focusing on out-of-distribution (OOD) data handling for tabular datasets.",
"strength: The paper introduces Tabular Contrast Learning (TCL) a... | 2.5 |
kaAtQwhnM2 | [
"strength: The paper presents its findings in a clear and structured manner, offering clear explanations of the energy discrepancy loss and provide useful insights on its connection to prior loss functions.",
"strength: The paper focuses on the modeling of discrete distributions using Energy-Based Models (EBMs), ... | 5.4 |
MbtUctg3KW | [
"strength: The paper clearly describes the misalignment augmentation problem.",
"strength: The proposed method is effective.",
"strength: The proposed method appears interesting. Leveraging a pretrained model like CLIP to capture invariant information could be effective in certain cases. However, the approach p... | 2.5 |
1EEst6oDU7 | [
"strength: The writing is overall good and easy to follow.",
"strength: The idea of translating natural language advice to RLang and using RLang to generate synthetic transitions makes sense.",
"strength: The writing flow in the experiment is great – sec 4.1 and 4.2 present two effective cases with assumptions ... | 5.5 |
yi3QcCGfP1 | [
"strength: The paper is well written. The motivation is clear and the contribution BRO is well stated.",
"strength: The BRO method leverage FFT for convolution in a similar fashion as the Caley approach",
"strength: The authors have performed an extensive set of experiments to demonstrate their results (compara... | 4.8 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.