Split (2)

train · 2.31k rows

paper_id stringlengths 10 10	items listlengths 4 82	gt float64 1 9.5
mtJSMcF3ek	[ "strength: The research focus of this paper is commendable, as the GV-Gap aids in determining whether a model possesses self-improvement capabilities at any stage (Pre-/Post-training).", "strength: The overall layout are well-organized.", "strength: The experiments appear to be extensive and comprehensive. (Res...	7
FhBT596F1X	[ "strength: The proposed method is basis-independent.", "strength: The method shows good results on structural extrapolation and data efficiency.", "strength: Originality: GG-XC offers a novel methodology by integrating GNNs with DFT, particularly beneficial for non-local XC functional design. This approach adva...	7.5
BCeock53nt	[ "strength: Existing works using KANs in-place of MLP modules in various models (such as feedforward NNs, CNNs, Transformers etc) have had limited success due to various issues. This work goes a step further and proposes an alternative to the original KAN architecture called Group Rational (GR-KAN) solving the issue...	6.8
faSfhqDpZP	[ "strength: The work proposes a novel solution to alleviate difficulties in surgical simulation pipelines.", "strength: The current approach use mesh-based tissue modeling, which although are realistic, require manual modeling.", "strength: The authors combine several existing technologies in a novel way to cons...	4.75
5Dwqu5urzs	[ "strength: Considers safety information and known system dynamics in the final policy.", "strength: Introduces novel scheme for neural network policy and action-value function editing using the known dynamics of the environment.", "strength: Addressing Safety in DRL: The paper introduces a DRL framework specifi...	7.5
LnxviiZ1xi	[ "strength: The authors show interesting performance gain on inter-layer link predictions.", "strength: The paper is clearly written.", "strength: The paper is well written and the experiments show improvements in predicting inter-layer connections.", "strength: S1. The setup of multiplex graph are from real-w...	3
zPxlHOLxmh	[ "strength: The integration of LLMs, specifically GPT-4o Mini, with PlantUML for generating mind maps from qualitative counseling data is a creative and original contribution. This approach moves beyond traditional text-based summaries by providing a visual representation of complex counseling sessions, which is und...	2
9rzEPbs4Wg	[ "strength: The proposed method does not require any changes to optimization, as it operates only in the input layer", "strength: The experimental evaluation is performed on multiple datasets and architecture complexity", "strength: The paper is well written", "strength: The study of anchoring-based training a...	6.166667
fszrlQ2DuP	[ "strength: Evaluate UDA models in an unsupervised manner is very important, but there is a lack of relevant research in the current community", "strength: Experiments demonstrate the effectiveness of “Transfer Score” in method selection, hyperparameter tuning, and checkpoint selection.", "strength: The paper is...	6
c0PnZCNY2N	[ "strength: ROBULT successfully maintains synergy while explicitly preserving the unique information of each modality, even when labels are missing during training and modalities are absent during testing. This is supported by experimental results and ablation studies.", "strength: The supplementary materials prov...	4.75
tmBKIecDE9	[ "strength: Figure 1 provides a nice clean bird's eye view of the overall approach and helps with readability.", "strength: The evaluation of the agent for not just the game score but also other dimensions provides a helpful qualitative assessment of the proposed approach and baselines through the spider graph in ...	7.25
9L9j5bQPIY	[ "strength: The concept Idea is interesting", "strength: The paper is well-written, with few formatting issues.", "strength: I like the idea of encoding model weights without using the training dataset, enabling work on models where access to the training data is not possible.", "strength: The experiments are ...	2.5
o0DQDGaVIY	[ "strength: introduces a novel perspective on noise in deep learning by categorizing it into positive noise (PN) and harmful noise (HN) based on its impact on task complexity", "strength: offers a fresh approach to understanding noise in the context of deep learning", "strength: presents impressive experimental ...	4.25
Iyrtb9EJBp	[ "strength: The paper is an original and significant contribution to the field of RAG in LLMs as it helps to mitigate the problems of citation hallucination or groundedness of claims.", "strength: The paper clearly articulates an extended experimental protocol, providing the rationale for its methodology. The find...	8
M9p2SIq0Oj	[ "strength: The problem that the paper considers is very important in continual reinforcement learning, especially in cases when skills are built and added to the agent’s base action set in a continual manner (although this is not explored in the paper).", "strength: Albeit not novel, the idea of learning action r...	3.75
orG37FHN4b	[ "strength: This paper addresses an important research area with practical applications.", "strength: Focuses on preserving weight orientation after quantization by minimizing angle deviation, a unique approach not seen in other baselines.", "strength: Data calibration-free, making it suitable for privacy-sensit...	3
uqYjAQ5diD	[ "strength: The main idea of the paper is to speed-up the RGB-based mapping process by using key ideas published recently – (1) A factorized radiance field representation that was proposed in by Chen et al 2022 to make radiance field learning more efficient, and (2) leveraging the concept of learning a depth covaria...	3
OcXsdBo6vK	[ "strength: The paper is well-written and easy to follow.", "strength: The authors provide a comprehensive set of experiments across different datasets and models.", "strength: The figures and tables are well-organized, clear and easy to understand.", "strength: The method is relatively lightweight and easy to...	4
AecVG5CXdp	[ "strength: This feature, which models continuous passenger arrivals, creates a learning environment for the RL agent that mirrors real-life complexities.", "strength: The paper's approach is designed to avoid combinatorial complexity, ensuring efficient decision-making through a well-structured action space. This...	3
wD2sfTDy1W	[ "strength: Expert involvement during development: The construction of the dataset involves heavy expert involvement (from annotating images to deciding classes), ensuring the practicality of the benchmark.", "strength: Focus on low-altitude imagery: As described in the related work, prior datasets on low-altitude...	3.75
u2QdCiOgwA	[ "strength: The paper's main strength lies in its analysis of the importance of different modules for various speech processing tasks.", "strength: The findings indicate that, within the encoder-decoder framework, pruning the decoder is more effective for speech translation (ST) and automatic speech recognition (A...	7
6jr94SCjH6	[ "strength: The problem motivating this work is important, and to the best of my knowledge this algorithm seems like a novel contribution", "strength: Many parts of the paper that are nicely written, including the motivations outline in section 1, and discussion in section 3.", "strength: The math discussed in t...	4.6
IEs29RYxfK	[ "strength: The authors provide extensive experimental results across 43 TSF benchmarks, demonstrating VISIONTS's effectiveness in zero-shot and few-shot settings.", "strength: The paper offers a theoretical analysis of the similarities between images and time series, providing a solid foundation for the proposed ...	5.333333
pK7V0glCdj	[ "strength: Multi-objective optimization is an interesting topic in Bayesian optimization.", "strength: The proposed method based on multivariate ranking seems interesting.", "strength: This paper focuses on improving the computational efficiency of the acquisition function, which is an important issue in multi-...	4.25
Tn8EQIFIMQ	[ "strength: I found the results of the paper very surprising, I think that the approach is creative, and to the best of my knowledge, may perhaps offer a minimal explanation for why LLMs replicate human behavior in some cognitive biases - namely, something about the next word prediction mechanism and mathematical da...	7
U1DjXQeJRx	[ "strength: The integration of physics-based principles is always welcome in the dynamical systems world.", "strength: The authors have a very nice contribution to make here potentially as the integration of physics principles into neural networks is very important.", "strength: As the reviewer summarizes above,...	6.6
rcmhydaEJp	[ "strength: The paper explores an underexplored technique of data imputation using normalizing flows, making it a novel application.", "strength: The explanation of the methods is clear and informative.", "strength: The examples listed in lines 31/32 are quite specific and niche, but not irrelevant.", "strengt...	3
ec9hJPn59o	[ "strength: The proposed BiEnhancer module presents a novel approach to enhancing the representation of low-light images by optimizing the loss function of high-level tasks, which is a unique perspective compared to previous methods that focus on image enhancement for human visualization.", "strength: The authors ...	3.4
fnnDtyMxcX	[ "strength: The proposed method is easy to understand.", "strength: It addresses an important problem.", "strength: Novelty: Although the overall idea of learning a similarity matrix for heterophilous graph learning is not new, this paper proposed an interesting framework by observing the advantage of RMK framew...	3.25
5nldnvvHfw	[ "strength: The topic is relevant to the theme of ICLR.", "strength: The authors consider replacing the constant coefficients with a sequence of coefficients, which can help the algorithm perform better. Meanwhile, the authors give a theoretical analysis of the proposed algorithm.", "strength: The idea that usin...	2.5
mXZ98iNFw2	[ "strength: The paper proposes Validation and Iterative Refinement modules that, depending on the specific context, selectively optimize either the box or the text, thereby further enhancing the model's accuracy.", "strength: The model outperforms previous works on the UICrit dataset.", "strength: Building compu...	4.75
YPpkFqMX6V	[ "strength: How to design an effective HGNN architecture plays a vital role in the graph learning community. The implications of this work could be significant for many downstream applications.", "strength: Extensive experiments are conducted to evaluate the proposed approach on many benchmark datasets and models....	5.25
VZTFUtldbC	[ "strength: The overall paper focuses on an important problem of learning modular controllers such that they can be transferred from one morphology to another.", "strength: The authors introduce a very important problem for actual robotics, that of transferring learned policies between robots (with different morph...	4.75
4g0PUEAHg0	[ "strength: There has been a lot of growing interest in theoretically studying whether transformers can learn certain structures [1, 2, 3]. The problem this work studies, whether transformers learn bayesian networks, is very interesting and relevant to the ICLR community.", "strength: The general problem of learni...	3.5
1pXzC30ry5	[ "strength: The model achieves multi-purpose segmentation through an efficient structure and unified training approach.", "strength: The paper is well-written and easy to follow.", "strength: The experiments on panoptic segmentation, interactive segmentation, and video segmentation are solid, comprehensive, and ...	7.5
vmiV4Z99lK	[ "strength: Quantization of neural networks is an important problem, provided the significant research impetus that is going into this domain in order to make modern large neural networks (which can easily consist of billions of parameters) deployable on memory-constrained devices.", "strength: One of the strength...	4.25
xPxHQHDH2u	[ "strength: By extracting explicit geometry, the paper addresses the inter-reflection issue in Gaussian splatting, which is important in realistic PBR. Experimental results showcase its SOTA performance in novel view synthesis and decomposition on reflective cases.", "strength: Instead of per-Gaussian PBR (like Re...	6.5
M2MinWsyjC	[ "strength: Robustness to Irregular Sampling: KRNO effectively addresses the challenge of irregularly sampled data, making it applicable in real‑world scenarios where regular sampling isn’t guaranteed.", "strength: Computational Efficiency: The model achieves almost linear computational costs while allowing for ex...	5
ajxAJ8GUX4	[ "strength: The study tackles a significant challenge in TAMP arising from the bottleneck in geometric planning within complex 3D environments and proposes a novel, though incremental, architecture to predict action and grasp feasibility to reduce the need for costly geometric planning queries.", "strength: The mo...	6.5
TCSaLeANpN	[ "strength: SYNBUILD-3D fills a significant gap by providing detailed interior and exterior building structures with semantic annotations, which is crucial for training AI models capable of generating realistic and complex 3D buildings.", "strength: The dataset benefits from a robust generation pipeline that combi...	3
hiZPVlbGsI	[ "strength: The paper addresses a relatively underexplored area: the recognition of handwritten invoices with stamps, contributing a preliminary exploration in this domain.", "strength: The primary contribution and strength of the paper is in its context-splitting architecture, which uses XTab-PAN-based backbone a...	2.6
ja4rpheN2n	[ "strength: The authors present a novel method for creating gene-gene interactions that are aware of disease subtypes. They also present a new combination architecture and a novel method for integrating knowledge graphs and gene expression data, as well as a shared integrated dataset.", "strength: The paper is ver...	8
9LdJDU7E91	[ "strength: IRIS demonstrated the improved results over CodeQL", "strength: IRIS found unknown vulnerabilities", "strength: This paper is well written and contains details.", "strength: It's interesting to learn that the LLMs can infer taint specifications with 70% accuracy", "strength: Originality: The pape...	5.75
FR2WQcwjG4	[ "strength: Innovative Approach to OOD Generation: The paper proposes a novel method for generating out-of-distribution (OOD) samples by intervening in core features of in-distribution (ID) samples. This method leverages saliency maps to identify core regions, making it uniquely capable of creating OOD samples that ...	5.6
Y93F5eNmZG	[ "strength: The paper provides an in-depth explanation of the model architectures and training procedures for both M-LNN and P-LNN, as well as the noise structures used to augment synthetic data.", "strength: The introduction is well-written, and the motivation behind LPPLS is clearly laid out.", "strength: The ...	3
EQAHilKZ8D	[ "strength: The paper tackles a challenging open problem in visual understanding.", "strength: A new metric is introduced.", "strength: The core idea of selecting feature dimensions related to the effect of glass or mirrors has some potential, but should be worked out in a more clever way.", "strength: A new m...	2.2
QLKgDBUXTR	[ "strength: The paper states an important idea that a model's prediction relies on multiple features/views, and the experimental result (Table 1) provides a strong evidence", "strength: I find the finding in Table 1 interesting (previous point)", "strength: I like the idea that this method can be used to select ...	4.75
nnAPWDt4hn	[ "strength: The paper is well-structured and easy to follow.", "strength: The approach of assessing the geospatial reasoning abilities of advanced foundation models through map services is interesting. By leveraging real-world contexts and results provided by Google Maps, the evaluation becomes both practical and ...	4.5
5UKrnKuspb	[ "strength: The unique advantage of the system is the association of geometry and semantic features for plane reconstruction problem, which requires both geometry-aware perception and semantic-aware grouping.", "strength: The paper leverages foundation models to achieve plane segment initializaing and utilize geom...	8
y5einmJ0Yx	[ "strength: No OOD Data Required: Synthesizes pseudo-OOD data through adversarial training, alleviating the need for real OOD samples.", "strength: Implementation Flexibility: Supports both LDM and VAE variants, offering trade-offs between performance and computational efficiency.", "strength: *Strong Empiri...	7.5
PA3MWNDD6O	[ "strength: The paper combines ideas from MAB literature with tools from blockchains/smart contracts which is an interesting innovation.", "strength: The authors provide detailed theoretical analyses of the regret.", "strength: The paper leverages blockchains in cybersecurity to complement existing robust multi-...	3.5
rRRgj3iIHR	[ "strength: The effects of bidding were incorporated, and improvements were made to Deep Montecaro.", "strength: Stronger performance is achieved against DouZero and its variants.", "strength: The AlphaDou framework is straightforward and likely easy to implement.", "strength: The bid model is a novel artifact...	3
WK6hQoAtgx	[ "strength: Fine-grained visual and OCR-free document understanding is a good point for evaluating LVLMs.", "strength: The evaluated models are representative and SOTA.", "strength: This paper is well-written and clear.", "strength: The grounding-like task is challenging and the authors explore further in the ...	4.25
j8hdRqOUhN	[ "strength: The paper is clearly written and easy to follow", "strength: The proposed stochastic resampling is novel and theoretically solid.", "strength: The performance of the proposed method outperforms baseline methods, especially simple extension of DPs.", "strength: The authors propose a novel hard data ...	7.5
esYrEndGsr	[ "strength: A theoretical framework is shown to unify influence functions, Journey-TRAK, and D-TRAK for diffusion models. This provides some clarity to the field of data attribution for diffusion models, which currently seems rather empirical.", "strength: The theoretical framework motivates the design choices for...	8
M6XWoEdmwf	[ "strength: By redesigning the off-policy in-context approach and using sequence models, AMAGO overcomes previous bottlenecks in memory capacity, planning horizon, and model size.", "strength: The introduction of a hindsight relabeling scheme broadens the applicability and scalability of this approach to open-worl...	7.5
QN97ubU1HH	[ "strength: The important task of efficiency of LLM alignment, and the interesting idea of extrapolating the LLM weights along the direction of the initial model toward the partially-trained one", "strength: Experiments are done on multimple open-source LLMs and bechmarks", "strength: The paper provides theoreti...	4.5
7suavRDxe8	[ "strength: The idea of using a large language model to provide plausible deniability is highly original and provides an interesting example of the combination of recent advances in machine learning with cryptography. More precisely, using the compression capability of LLMs seem interesting approach to be able to ac...	4.8
bIlnpVM4bc	[ "strength: The proposed architecture shows great results on common benchmark.", "strength: It seems also to be better at long contexts handling.", "strength: Shows that hybrid models can be competitive alternatives to transformers.", "strength: The investigation of SSM-attention hybrids is well motivated and ...	6.666667
CSAfU7J8Gw	[ "strength: The proposed idea sounds and improves multiple baseline models.", "strength: The paper is clearly written, especially Figure 2 is very informative.", "strength: Grad-CAM visualization and the ablation study show the benefit of the proposed method.", "strength: By introducing a specialized assistant...	5
sK2A7Ve2co	[ "strength: The paper proposes a sampler that samples weights via traversing the loss landscape of a pre-trained deep neural network via a series of normal distributions. The approach is evaluated on a series of classification and out-of-distribution detection tasks.", "strength: The studied topic of sampling on m...	2.5
FsgGBhNIt4	[ "strength: This paper introduces the unsupervised approach to facial attribute analysis, reducing dependence on labeled data by fully exploring StyleGAN's feature extraction capabilities. It also introduces a novel mutual information-based feature refinement method, enhancing representation quality for unsupervised...	3
d4uL2MSe0z	[ "strength: Large transformer models have demonstrated excellent performance, but they are often very expensive in terms of computational resources. The paper addresses a very timely problem: ensuring transformer models can be applied in a practical setting without excessive cost.", "strength: The proposed method,...	4.5
v3DwQlyGbv	[ "strength: The paper is well-written.", "strength: I appreciate the background and the description of how the model is trained.", "strength: The idea of targeting mathematics is important and building LLMs specializing in math (at least some part of it) is important.", "strength: The dataset is an important c...	2.333333
ZGNWW7xZ6Q	[ "strength: The approach is novel and interesting.", "strength: The experiments show strong results and improvements over SOTA.", "strength: The paper is well written though the organization of the approach description can be improved.", "strength: Originality: This paper presents a solid concept for addressin...	7.5
8oUF3uGIVo	[ "strength: This work introduces an interesting approach to leveraging virtual nodes to improve the capability of graph transformers to capture graph high-order information, for node-level tasks. Prior works leverage virtual nodes to solve graph-level tasks (e.g., graph classification).", "strength: The designed a...	4
0ZUKLCxwBo	[ "strength: The paper's presentation is clear and to the point, and the construction of the weights is succinctly explained.", "strength: The experimental evidence is convincing.", "strength: Mechanistic interpretability is also a highly interesting direction overall.", "strength: Grokking is an interesting an...	6
Ocg3XIymmp	[ "strength: Novel combination of generative design and 2D substructure similarity search", "strength: Innovative voxel captioning approach for SMILES generation", "strength: Creative solution to address traditional pharmacophore screening limitations", "strength: This work builds upon 3D ligand-based virtual s...	3.5
j511LaqEeP	[ "strength: The paper connects two modern techniques in conformal prediction, and is thus very relevant for the community.", "strength: It is well-written and easy to follow as an expert.", "strength: The writing is simple, and I expect the paper will also be easy-to-follow for readers unfamiliar with conformal ...	6
qW5f8TAZ4J	[ "strength: The authors propose a methodology for skin image generation based on diffusion models that account for fairness during image generation. To address the imbalance of skin lesions, they investigate two data sampling strategies, a balanced diffusion model training by introducing a class diversity loss, and ...	3
cselR6Jne3	[ "strength: This paper propose a new dataset, LatentQA, aiming to improve interpretability and control in language models (LLMs) by transforming activations into language responses.", "strength: This paper includes comprehensive experiments and anlysis", "strength: This paper provides some insights from explaini...	5.25
pLyjsv1KWH	[ "strength: The FedCDD framework improves the generalization ability of the model in different datasets by effectively managing the consistency and difference in parameter updates.", "strength: Gaussian noise masking reduces computational cost and communication overhead by selectively updating important parameters...	3
ZxQD6oYIOm	[ "strength: The authors study an interesting problem setting: transforming one recipe to another cuisine", "strength: The authors perform extensive experiments", "strength: The paper presented one approach for the recipe transformation from one regional cuisine to another by combining previously pretrained model...	3
yRd4loGAhJ	[ "strength: This paper is well-organized and easy to read.", "strength: The proposed method presents a reasonable approach for long-context retrieval by identifying the key components of Transformer architecture to boost retrieval performance.", "strength: The approach is practical and has the potential for broa...	5.25
5y3QbuK6HD	[ "strength: The paper studies an interesting problem in average-reward RL, leverages a structural property specific to average-reward MDPs, introduces a generic framework, applies it to CVaR RL, and removes the need to solve bi-level optimization problems explicitly.", "strength: The paper is mostly well-organized...	4.5
7kRFnSFN89	[ "strength: The idea of learning a tokenization for sign language video and combining it with text tokens is compelling, and the use of a VQ-VAE-based approach for this seems sensible.", "strength: This work represents the first use of RAG for sign language translation, as far as I know.", "strength: This paper ...	5
ZPdZLlNXSm	[ "strength: The idea of introducing unique losses based on the theory of statistical physics looks interesting and novel. No prior research has taken on this particular task.", "strength: The proposed method is evaluated on several popular DML benchmarks. The authors evaluate their method on advanced MLRC metrics,...	5.4
jBpEsliki9	[ "strength: The proposed method uses hypergraphs to capture higher-order interconnections between variables, which is a novel approach in addressing missing data.", "strength: The use of realistic datasets, such as MIMIC-III, distinguishes this study from existing studies that primarily relies on benchmark dataset...	2.5
EeDSMy5Ruj	[ "strength: Well-motived objective: augmenting existing formal corpus synthetically is surely of immerse interest, given the limited amount of formal corpus available.", "strength: The paper is relatively well-written and easy to follow, and with good amount of implementation details in the appendix for reproducib...	5
mHkbi3XM58	[ "strength: The main contribution of this paper is mostly conceptual. The key ideas of (1) employing a direct regression approach for learning score-based models and (2) the sampling strategy from the learned denoising function are both intriguing. It is perhaps a bit surprising to see how they can work even on some...	3.25
JXvEzl8YkS	[ "strength: Inclusion of multiple penalty terms in the proposed method.", "strength: The authors identify a relevant problem in Jump Models for financial time series. With more development, this work could offer useful insights into this type of models and new ways to improve model regularization.", "strength: T...	2
aS1IhKdLPP	[ "strength: The theoretical characterization of sub-optimality is reasonable.", "strength: The proposed method can be a potential solution for the gap between beam search and greedy decoding.", "strength: Novelty in technique: Utilizing a fast-slow pointer for reflection and generation is quite a technically int...	4.75
5t57omGVMw	[ "strength: The problem is well-motivated (even though the algorithm looks slightly restrictive).", "strength: The critical observation is that the SOR iteration is bounded by a convex surrogate function with the parameter.", "strength: The observation is non-trivial and thus the paper shows a new and interestin...	8
YrycTjllL0	[ "strength: The paper makes a valuable contribution to code generation benchmarks by incorporating diverse function calls and complex instructions, offering a novel perspective on evaluating LLM capabilities in library and tool use. Results indicate that even top-performing LLMs face challenges with these tasks, hig...	9
6uReXuDWrw	[ "strength: Authors are strongly encouraged to thoroughly investigate BCI and EEG issues prior to proposing any universal solutions.", "strength: Moreover, enhancing the clarity of manuscript writing is imperative to ensure that readers can comprehend and replicate the research findings.", "strength: Importing m...	2
ySmovxuDMi	[ "strength: It provides comprehensive code for both evaluation and training.", "strength: The task design is well-structured, and the benchmark’s support for image input is both rare and crucial for advancing multi-agent reinforcement learning (MARL) development.", "strength: This work uses the Unity engine to c...	4.25
VJDFhkwQg6	[ "strength: The paper provides a contribution to the community interested in generative modeling, enabling GFlowNets in federated settings, introducing a novel training scheme for centralized settings, and demonstrating the potential of GFlowNets on a new application domain.", "strength: The authors introduce a ne...	5
86uYj8DcfK	[ "strength: The dataset encompasses a diverse range of image modification types from various sources, which increases the dataset’s variability in modification categories.", "strength: DiffTell may have practical applications in detecting manipulated images and generating descriptive captions for them, may contrib...	4.75
zUlK1qMIcE	[ "strength: The claims of the paper are easy to understand (though I dont quite believe them, see below)", "strength: The experimental results one of the datasets was interesting to read", "strength: Interesting new paradigm. Even though it's similar to the ideas of mixture of experts which are well-studies in c...	3
3Pn24GOcQ1	[ "strength: Comparing learning with data augmentation vs an invariant architecture is an interesting problem and linear networks may be a good place to start for a theoretical study.", "strength: The paper is written rather well but the global structure of the paper (presentation order) is confusing (see Weaknesse...	5.8
7BLXhmWvwF	[ "strength: The paper is well-written and presents its ideas clearly with solid formulation.", "strength: The proposed approach of modeling robotic manipulation problems as heterogeneous graph learning problems is well-motivated and clever to unify the structure for both rigid and deformable object tasks using sub...	8
yIN4yDCcmo	[ "strength: Comprehensive Benchmark: The paper presents INS-MMBench, which is the first comprehensive benchmark tailored for evaluating LVLMs in the insurance domain. This benchmark is extensive, covering 8,856 multiple-choice visual questions across 12 meta-tasks and 22 fundamental tasks, providing a robust framewo...	5
B8aHIDSi7E	[ "strength: The proposed method is novel and insightful", "strength: The proposed method is well-motivated and has the potential to make a broader impact.", "strength: The experiment results are promising.", "strength: The paper proposes a method to compress the rotation matrix Q introduced by SliceGPT using t...	6
AXC9KydyZq	[ "strength: A sound problem, to try to study jointly matching and clustering for graphs", "strength: The theoretical parts, often postponed in the supplementary sections, are well written. Yet they are coming straight from classical results.", "strength: The numerical experiments are well conducted.", "strengt...	7
XHTirKsQV6	[ "strength: Originality: ProtoSnap's use of deep diffusion features and skeleton - based prototypes for unsupervised cuneiform sign alignment is novel.", "strength: Quality: The overall flow of the methodology section is sound and logical.", "strength: Clarity: This paper is clearer on the whole, from the introd...	6.25
C6a0Obrp3o	[ "strength: The method allows more flexible ediitng for the inversed concepts from a single image, surpassing its baselines.", "strength: The method presents a novel way to regularize the Image-to-Text Inversion process with predicted distributions by the original model.", "strength: The paper presents an ediitn...	4.333333
ech9J3xl9X	[ "strength: The paper is written with a coherent story-line, which leads one through the entire paper, and does a very good job in describing its experiments as well as the exact setup used. The reviewer is highly confident that one would be able to reproduce the results from their description in the paper.", "str...	2.5
FUaDMRVrbS	[ "strength: This is a very well-written paper addressing an important research question in the literature. The theory is solid, especially Theorem 3.4, can be very useful for researchers in related areas.", "strength: The motivation for studying identifiability conditions for kernels of GP is strong, as it enables...	7
gJeYtRuguR	[ "strength: This work has adeptly identified and proposed solutions for a problem in the literature of token pruning method of ViT.", "strength: To address the inconsistency between [CLS] attention and token significance at the early blocks, the proposed method that incorporates multi-exit into ViT) is novel and i...	7.5
1ThYY28HXg	[ "strength: Although the masked latent conditioning is not new, the architectural modification upon SVD is sensible and allows joint training on 3D and 4D data.", "strength: The proposed model is capable of 3D and 4D generation of both object-centric and scene-level videos, which is more general than most prior me...	6.25

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.