paper_id string | arxiv_id string | title string | markdown dict | reviews list | scores dict | metadata dict | meta_review dict | decision dict |
|---|---|---|---|---|---|---|---|---|
00SnKBGTsz | 2410.06215v1 | DataEnvGym: Data Generation Agents in Teacher Environments with Student Feedback | {
"content": "## Abstract\n\nAbstract The process of creating training data to teach models is currently driven by humans, who manually analyze model weaknesses and plan how to create data that improves a student model.\nRecent approaches using large language models (LLMs) as annotators reduce human annotation effort... | [
{
"id": "r8ZflFk3T7",
"initial_rating": 8,
"confidence": 4,
"soundness": 4,
"contribution": 4,
"presentation": 3,
"summary": "This paper introduces Gym environments for data synthesis, framing the problem as sequential decision-making. In these environments, actions correspond to data-ge... | {
"rating": "5;6;6;8",
"rating_avg": 6.25,
"confidence": "4;3;4;4",
"confidence_avg": 3.75,
"soundness": "2;2;4;3",
"soundness_avg": 2.75,
"contribution": "3;3;4;3",
"contribution_avg": 3.25,
"presentation": "3;3;2;2",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.543857"
} | {
"id": "zpboemkkjR",
"metareview": "The paper frames the problem of automatic data generation (to improve a ML model) as a sequential decision making task, and provides Gym environments as well as LLM-based agents that are effective for them. The resulting datasets are shown to be effective for ML models in math r... | {
"decision": "Accept (Spotlight)"
} |
00ezkB2iZf | 2406.06573v2 | MedFuzz: Exploring the Robustness of Large Language Models in Medical Question Answering | {
"content": "## Abstract\n\nAbstract Large language models (LLM) have achieved impressive performance on medical question-answering benchmarks.\nHowever, high benchmark accuracy does not imply that the performance generalizes to real-world clinical settings.\nMedical question-answering benchmarks rely on assumptions... | [
{
"id": "TeO25XUwES",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper investigates the robustness of large language models in handling medical QA tasks by introducing a new evaluation method, MedFuzz. For each multiple-ch... | {
"rating": "3;3;5;6",
"rating_avg": 4.25,
"confidence": "4;4;5;3",
"confidence_avg": 4,
"soundness": "3;2;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;4;3",
"contribution_avg": 2.75,
"presentation": "3;2;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.544972"
} | {
"id": "Se1GK2iVy4",
"metareview": "In this paper, the authors propose MedFuzz, an LLM-based technique to provide challenging medical questions. The technique allows to test medical LLMs at scale on a multichoice QA dataset, by modifying factors deemed as 'irrelevant' to the final diagnosis.\n\nWhile all reviewers... | {
"decision": "Reject"
} |
02Od16GFRW | 2410.01452v1 | Ensembles provably learn equivariance through data augmentation | {
"content": "## Abstract\n\nAbstract Recently, it was proved that group equivariance emerges in ensembles of neural networks as the result of full augmentation in the limit of infinitely wide neural networks (neural tangent kernel limit). In this paper, we extend this result significantly. We provide a proof that th... | [
{
"id": "HKJJNQ1JKw",
"initial_rating": 3,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper shows that an ensemble of models when trained with data augmentation leads to emergence of equivariance properties naturally. The results generalize ov... | {
"rating": "3;6;6",
"rating_avg": 5,
"confidence": "4;3;3",
"confidence_avg": 3.3333333333333335,
"soundness": "3;3;3",
"soundness_avg": 3,
"contribution": "2;3;2",
"contribution_avg": 2.3333333333333335,
"presentation": "3;3;2",
"presentation_avg": 2.6666666666666665
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.545938"
} | {
"id": "I3eiVnCVfC",
"metareview": "This paper studies how equivariance emerges in ensembles of neural networks trained with data augmentation. The authors extend prior theoretical results by showing that equivariance holds under weaker conditions that exist in prior work (e.g. without requiring the NTK limit).\nT... | {
"decision": "Reject"
} |
02haSpO453 | 2409.04429v2 | VILA-U: a Unified Foundation Model Integrating Visual Understanding and Generation | {
"content": "## Abstract\n\nAbstract VILA-U is a U nified foundation model that integrates V ideo, I mage, La nguage understanding and generation. Traditional visual language models (VLMs) use separate modules for understanding and generating visual content, which can lead to misalignment and increased complexity. I... | [
{
"id": "cGas6kZlaM",
"initial_rating": 6,
"confidence": 4,
"soundness": 4,
"contribution": 3,
"presentation": 3,
"summary": "- The paper presents VILA-U, a unified model for language, image and video understanding + generation\n- The model is trained with an autoregressive next token pr... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "4;5;3;4",
"confidence_avg": 4,
"soundness": "3;2;2;4",
"soundness_avg": 2.75,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "3;2;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.546904"
} | {
"id": "gg4i7pnNPQ",
"metareview": "VILA-U presents a unified foundation model that integrates video, image, and language understanding and generation within a single autoregressive next-token prediction framework. Unlike traditional visual language models that use separate modules for understanding and generating... | {
"decision": "Accept (Poster)"
} |
02kZwCo0C3 | 2406.15567v1 | SAIL: Self-improving Efficient Online Alignment of Large Language Models | {
"content": "## Abstract\n\nAbstract Reinforcement Learning from Human Feedback (RLHF) is a key method for aligning large language models (LLMs) with human preferences. However, current offline alignment approaches like DPO, IPO, and SLiC rely heavily on fixed preference datasets, which can lead to sub-optimal perfo... | [
{
"id": "BU6la6v4Ci",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 4,
"presentation": 2,
"summary": "Compared to offline RLHF methods, online RLHF methods empirically show stronger performance, yet is computationally expensive, vulnerable to distribution shifts a... | {
"rating": "3;6;6;8",
"rating_avg": 5.75,
"confidence": "4;3;4;4",
"confidence_avg": 3.75,
"soundness": "3;3;3;4",
"soundness_avg": 3.25,
"contribution": "3;3;4;3",
"contribution_avg": 3.25,
"presentation": "2;4;2;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.547632"
} | {
"id": "E3XYUMBr1C",
"metareview": "The paper introduces SAIL, a self-improving online RLHF approach for aligning large language models (LLMs). SAIL frames online alignment as a bilevel optimization problem, reducing it to a computationally efficient single-level method. The framework enables continuous improvemen... | {
"decision": "Reject"
} |
03EkqSCKuO | 2405.17163v1 | Port-Hamiltonian Architectural Bias for Long-Range Propagation in Deep Graph Networks | {
"content": "## Abstract\n\nAbstract The dynamics of information diffusion within graphs is a critical open issue that heavily influences graph representation learning, especially when considering long-range propagation. This calls for principled approaches that control and regulate the degree of propagation and dis... | [
{
"id": "d6JJf0KjwN",
"initial_rating": 6,
"confidence": 3,
"soundness": 4,
"contribution": 2,
"presentation": 3,
"summary": "This paper introduces port-Hamiltonian Deep Graph Networks (PH-DGN), a new framework for graph neural networks that addresses the challenge of long-range informat... | {
"rating": "5;6;8",
"rating_avg": 6.333333333333333,
"confidence": "2;3;3",
"confidence_avg": 2.6666666666666665,
"soundness": "2;4;4",
"soundness_avg": 3.3333333333333335,
"contribution": "2;2;3",
"contribution_avg": 2.3333333333333335,
"presentation": "3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.548475"
} | {
"id": "H9PCwNoHeA",
"metareview": "**(a) Scientific Claims and Findings:**\nThe paper introduces a novel framework called port-Hamiltonian Deep Graph Networks (pH-DGNs). This framework models neural information flow in graphs by leveraging principles from Hamiltonian dynamical systems, aiming to address challenge... | {
"decision": "Accept (Poster)"
} |
03u7pbpyeN | 2409.17972v2 | BEATS: Optimizing LLM Mathematical Capabilities with BackVerify and Adaptive Disambiguate based Efficient Tree Search | {
"content": "## Abstract\n\nAbstract Large Language Models (LLMs) have exhibited exceptional performance across a broad range of tasks and domains. However, they still encounter difficulties in solving mathematical problems due to the rigorous and logical nature of mathematics. Previous studies have employed techniq... | [
{
"id": "0Md90Alr6g",
"initial_rating": 3,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "This paper presents BEATS, a framework that enhances mathematical problem-solving in language models by introducing targeted prompting strategies that guide the m... | {
"rating": "3;3;5;6",
"rating_avg": 4.25,
"confidence": "5;3;4;4",
"confidence_avg": 4,
"soundness": "3;3;4;3",
"soundness_avg": 3.25,
"contribution": "1;2;2;2",
"contribution_avg": 1.75,
"presentation": "4;2;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.549167"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
06GH83hDIv | 2410.01871v1 | Auction-Based Regulation for Artificial Intelligence | {
"content": "## Abstract\n\nAbstract In an era of “moving fast and breaking things” , regulators have moved slowly to pick up the safety, bias, and legal pieces left in the wake of broken Artificial Intelligence (AI) deployment.\nSince AI models, such as large language models, are able to push misinformation and sto... | [
{
"id": "54x9fXc8nh",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper proposes a novel framework of auction-based regulatory mechanism as an asymmetric and incomplete all-pay auction. The mechanism is described mathematic... | {
"rating": "3;5;5;8",
"rating_avg": 5.25,
"confidence": "3;4;3;3",
"confidence_avg": 3.25,
"soundness": "2;2;3;4",
"soundness_avg": 2.75,
"contribution": "2;2;2;3",
"contribution_avg": 2.25,
"presentation": "3;2;4;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.549803"
} | {
"id": "J6U2HHgjTm",
"metareview": "This paper looks at the problem of regulating AI models, specifically for safety. This is approached through theoretical framework, through an all-pay auction with companies and a regulator. The authors find Nash equilibria and have theoretical results. \n\nReviewers agree that ... | {
"decision": "Reject"
} |
07yvxWDSla | 2409.07431v2 | Synthetic continued pretraining | {
"content": "## Abstract\n\nAbstract Pretraining on large-scale, unstructured internet text enables language models to acquire a significant amount of world knowledge.\nHowever, this knowledge acquisition is data-inefficient —to learn a given fact, models must be trained on hundreds to thousands of diverse represent... | [
{
"id": "5lDBbDNEH3",
"initial_rating": 8,
"confidence": 4,
"soundness": 4,
"contribution": 3,
"presentation": 4,
"summary": "This paper addresses the problem of data inefficiency in pretraining language models. Current pretraining corpora may not generalize effectively and models may be... | {
"rating": "6;6;8;8",
"rating_avg": 7,
"confidence": "3;4;3;4",
"confidence_avg": 3.5,
"soundness": "3;2;4;4",
"soundness_avg": 3.25,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "2;3;3;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Oral",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.550881"
} | {
"id": "eEZ1z1zkVr",
"metareview": "The paper proposes a data-synthesizing method for continued pretraining for adapting an LLM to a specific domain, where the proposed approach is based on EntiGraph including entities and their relations. In addition, the authors derived bounds for their scaling trends.\n\nAll re... | {
"decision": "Accept (Oral)"
} |
09FiNmvNMw | 2410.08047v1 | Divide and Translate: Compositional First-Order Logic Translation and Verification for Complex Logical Reasoning | {
"content": "## Abstract\n\nAbstract Complex logical reasoning tasks require a long sequence of reasoning, which a large language model (LLM) with chain-of-thought prompting still falls short. To alleviate this issue, neurosymbolic approaches incorporate a symbolic solver. Specifically, an LLM only translates a natu... | [
{
"id": "o6kdM9TOlc",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "The paper introduces CLOVER, an approach designed to enhance the translation of natural language logical problems into logical code, thereby improving the perform... | {
"rating": "5;5;5",
"rating_avg": 5,
"confidence": "4;3;4",
"confidence_avg": 3.6666666666666665,
"soundness": "2;3;3",
"soundness_avg": 2.6666666666666665,
"contribution": "3;2;2",
"contribution_avg": 2.3333333333333335,
"presentation": "2;1;2",
"presentation_avg": 1.6666666666666667
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.552006"
} | {
"id": "yQMwgvFDdg",
"metareview": "The reviewers generally saw the merits of the proposal. There was general interest in the paper. The authors have responded to several issues raised by the reviewers. On reading the paper at a fairly high-level, it does appear interesting and novel. The authors in the rebuttal ... | {
"decision": "Accept (Poster)"
} |
09LEjbLcZW | 2410.20424v3 | AutoKaggle: A Multi-Agent Framework for Autonomous Data Science Competitions | {
"content": "## Abstract\n\nAbstract Data science tasks involving tabular data present complex challenges that require sophisticated problem-solving approaches.\nWe propose AutoKaggle, a powerful and user-centric framework that assists data scientists in completing daily data pipelines through a collaborative multi-... | [
{
"id": "51LdCo1AcX",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper introduces AutoKaggle, a pipeline to automatically solve Kaggle Competitions. The authors use 5 subparts in a row: a reader, a planner, a developer, a ... | {
"rating": "5;5;5",
"rating_avg": 5,
"confidence": "4;4;3",
"confidence_avg": 3.6666666666666665,
"soundness": "2;2;3",
"soundness_avg": 2.3333333333333335,
"contribution": "3;2;2",
"contribution_avg": 2.3333333333333335,
"presentation": "3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.552678"
} | {
"id": "xvdXj2Rjbw",
"metareview": "The paper presents AutoKaggle, an LLM creating a multi-agent system together with a library of hand-crafted ML tools in order to solve Kaggle problems. In my opinion, the reviewers did a very thorough job and have presented salient arguments about the suitability of this paper f... | {
"decision": "Reject"
} |
0ASCZrVzSX | 2408.06996v1 | Blessing of Dimensionality for Approximating Sobolev Classes on Manifolds | {
"content": "## Abstract\n\nAbstract The manifold hypothesis says that natural high-dimensional data is actually supported on or around a low-dimensional manifold. Recent success of statistical and learning-based methods empirically supports this hypothesis, due to outperforming classical statistical intuition in ve... | [
{
"id": "uVX8uvjgic",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper studies the complexity of Sobolev function class on Riemannian manifolds. Specifically, the paper derives lower bound of the approximation error of a S... | {
"rating": "5;5;6",
"rating_avg": 5.333333333333333,
"confidence": "3;3;3",
"confidence_avg": 3,
"soundness": "4;2;3",
"soundness_avg": 3,
"contribution": "2;2;2",
"contribution_avg": 2,
"presentation": "3;2;3",
"presentation_avg": 2.6666666666666665
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.553461"
} | {
"id": "C0fqvyw2to",
"metareview": "The paper studies the complexity of Sobolev spaces on manifolds. Its main result is a lower bound on the error incurred by approximating Sobolev functions in Lq with subspaces of functions learnable from n samples; the bound scales as roughly n^{-1/d} where d is the dimension of... | {
"decision": "Reject"
} |
0Ag8FQ5Rr3 | 2411.07191v1 | The Super Weight in Large Language Models | {
"content": "## Abstract\n\nAbstract Recent works have shown a surprising result: a small fraction of Large Language Model (LLM) parameter outliers are disproportionately important to the quality of the model.\nLLMs contain billions of parameters, so these small fractions, such as 0.01%, translate to hundreds of tho... | [
{
"id": "92gomx5hZI",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper focuses on the impact of outlier weights in large language models (LLMs), specifically larger weights, which the authors term superweights and superact... | {
"rating": "1;5;5;5;5",
"rating_avg": 4.2,
"confidence": "5;4;4;3;3",
"confidence_avg": 3.8,
"soundness": "1;2;3;3;3",
"soundness_avg": 2.4,
"contribution": "1;2;2;2;2",
"contribution_avg": 1.8,
"presentation": "1;3;3;3;3",
"presentation_avg": 2.6
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.554199"
} | {
"id": "9tCo14UeDJ",
"metareview": "This paper explores the impact of superweights—defined as weights with larger magnitudes—on the performance of large language models (LLMs). The authors analyze the influence of these superweights on LLMs’ performance and propose specialized quantization methods tailored to supe... | {
"decision": "Reject"
} |
0CtIt485ew | 2410.05899v1 | Brain-inspired continual pre-trained learner via silent synaptic consolidation | {
"content": "## Abstract\n\nAbstract Pre-trained models have demonstrated impressive generalization capabilities, yet they remain vulnerable to catastrophic forgetting when incrementally trained on new tasks. Existing architecture-based strategies encounter two primary challenges: 1) Integrating a pre-trained networ... | [
{
"id": "dNk03ucBaT",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The paper introduces the Artsy framework, which enhances continual learning in pre-trained models by mimicking the activation mechanisms of silent synapses via sp... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "4;4;4;3",
"confidence_avg": 3.75,
"soundness": "2;2;3;2",
"soundness_avg": 2.25,
"contribution": "2;3;2;2",
"contribution_avg": 2.25,
"presentation": "2;2;3;2",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.554849"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0DZEs8NpUH | 2408.11779v1 | Personality Alignment of Large Language Models | {
"content": "## Abstract\n\nAbstract Current methods for aligning large language models (LLMs) typically aim to reflect general human values and behaviors, but they often fail to capture the unique characteristics and preferences of individual users. To address this gap, we introduce the concept of Personality Align... | [
{
"id": "UoRUU86feH",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper proposes the concept of Personality Alignment for tailoring LLMs to match the preferences and behaviors of individual users or groups. The authors cre... | {
"rating": "5;5;6",
"rating_avg": 5.333333333333333,
"confidence": "4;4;5",
"confidence_avg": 4.333333333333333,
"soundness": "3;3;4",
"soundness_avg": 3.3333333333333335,
"contribution": "2;2;3",
"contribution_avg": 2.3333333333333335,
"presentation": "3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.555698"
} | {
"id": "NKPkZOp6rR",
"metareview": "We recommend the paper to be accepted for Poster. \n\nThe paper can be of interest to the wide community at ICLR working on LLM and it introduces a relatively novel methodology that seems to be more efficient than baseline methods. \n\nBelow a more detailed description of the pa... | {
"decision": "Accept (Poster)"
} |
0EP01yhDlg | 2410.17765v1 | Faster Language Models with Better Multi-Token Prediction Using Tensor Decomposition | {
"content": "## Abstract\n\nAbstract We propose a new model for multi-token prediction in transformers, aiming to enhance sampling efficiency without compromising accuracy. Motivated by recent work that predicts the probabilities of subsequent tokens using multiple heads, we connect this approach to rank- 1 1 1 1 ca... | [
{
"id": "dzupEXEHLz",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper borrows key idea from Gloecke et al. [1] to train multi-token predictors instead of single next word predictor. This work identifies a key flaw in [1] ... | {
"rating": "3;5;5;5",
"rating_avg": 4.5,
"confidence": "4;4;3;4",
"confidence_avg": 3.75,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "2;3;2;3",
"contribution_avg": 2.5,
"presentation": "2;3;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.556528"
} | {
"id": "1cx9Eaufbb",
"metareview": "This paper aims to speed up the generation process of language models. Motivated by the observation that predicting multiple future tokens in parallel does not consider inter-token dependencies, and that it can be viewed as rank-1 tensor decomposition, this work proposes to exte... | {
"decision": "Reject"
} |
0GzqVqCKns | 2410.13770v1 | Probing the Latent Hierarchical Structure of Data via Diffusion Models | {
"content": "## Abstract\n\nAbstract High-dimensional data must be highly structured to be learnable. Although the compositional and hierarchical nature of data is often put forward to explain learnability, quantitative measurements establishing these properties are scarce. Likewise, accessing the latent variables u... | [
{
"id": "tcYvMFpF6C",
"initial_rating": 5,
"confidence": 2,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "The paper examines the hierarchical correlation structures among input tokens using a dynamic correlation function and dynamical susceptibility within a forward-b... | {
"rating": "5;5;6;6",
"rating_avg": 5.5,
"confidence": "3;2;3;3",
"confidence_avg": 2.75,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;3;3;3",
"contribution_avg": 2.75,
"presentation": "2;2;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.557532"
} | {
"id": "zCU8lv6FkO",
"metareview": "This paper investigates the hierarchical structure of data using forward-backward experiments with diffusion models. The authors propose that changes in data occur in correlated chunks, with a characteristic correlation length that diverges at a critical noise level associated w... | {
"decision": "Accept (Poster)"
} |
0IhoIn0jJ3 | 2406.16552v1 | Inference of Sequential Patterns for Neural Message Passing in Temporal Graphs | {
"content": "## Abstract\n\nAbstract The modelling of temporal patterns in dynamic graphs is an important current research issue in the development of time-aware Graph Neural Networks (GNNs).\nHowever, whether or not a specific sequence of events in a temporal graph constitutes a temporal pattern not only depends on... | [
{
"id": "u3xrl8trVh",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "This paper studies how to model temporal patterns in dynamic graphs and proposes to use statistical graph inference to identify sequence anomalies for graph augme... | {
"rating": "3;5;5;5",
"rating_avg": 4.5,
"confidence": "3;4;3;4",
"confidence_avg": 3.5,
"soundness": "2;3;2;3",
"soundness_avg": 2.5,
"contribution": "1;1;3;2",
"contribution_avg": 1.75,
"presentation": "1;2;2;2",
"presentation_avg": 1.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.558340"
} | {
"id": "8ojtKOXZOL",
"metareview": "## Summary\nThe paper proposes HYPA-DBGNN, a method for static node classification on temporal graphs. The model uses De Bruijn graphs to encode sequential patterns and employs a null model correction via hypergeometric testing to identify and adjust for anomalous temporal patte... | {
"decision": "Reject"
} |
0JOhLEf2bX | 2410.17708v1 | Proteome-wide prediction of mode of inheritance and molecular mechanism underlying genetic diseases using structural interactomics | {
"content": "## Abstract\n\nAbstract Genetic diseases can be classified according to their modes of inheritance and their underlying molecular mechanisms. Autosomal dominant disorders often result from DNA variants that cause loss-of-function, gain-of-function, or dominant-negative effects, while autosomal recessive... | [
{
"id": "jeWway56oy",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "The authors present a methodology able to detect both mode of inheritance (MOI) of proteins encoded by autosomal genes and the functional effects of gene variants... | {
"rating": "3;3;5",
"rating_avg": 3.6666666666666665,
"confidence": "3;4;4",
"confidence_avg": 3.6666666666666665,
"soundness": "2;2;2",
"soundness_avg": 2,
"contribution": "1;1;2",
"contribution_avg": 1.3333333333333333,
"presentation": "2;1;3",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.559032"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0JcPJ0CLbx | 2410.23132v1 | Revisiting MAE pre-training for 3D medical image segmentation | {
"content": "## Abstract\n\nAbstract Self-Supervised Learning (SSL) presents an exciting opportunity to unlock the potential of vast, untapped clinical datasets, for various downstream applications that suffer from the scarcity of labeled data.\nWhile SSL has revolutionized fields like natural language processing an... | [
{
"id": "iF7oMf3Z38",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper proposes a SSL framework, called nnSSL, for 3D medical image segmentation based on a MAE strategy and thorough evaluation of various design choices. The... | {
"rating": "3;3;3;6",
"rating_avg": 3.75,
"confidence": "4;5;4;4",
"confidence_avg": 4.25,
"soundness": "2;2;2;3",
"soundness_avg": 2.25,
"contribution": "2;1;2;3",
"contribution_avg": 2,
"presentation": "1;1;3;3",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.559659"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0JjsZC0w8x | 2410.09675v1 | COrAL: Order-Agnostic Language Modeling for Efficient Iterative Refinement | {
"content": "## Abstract\n\nAbstract Iterative refinement has emerged as an effective paradigm for enhancing the capabilities of large language models (LLMs) on complex tasks. However, existing approaches typically implement iterative refinement at the application or prompting level, relying on autoregressive (AR) m... | [
{
"id": "hJjYbcLqPe",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "This paper proposes Context-Wise Order-Agnostic Language Modeling (COrAL), which incorporates iterative refinement directly into the LLM architecture while mainta... | {
"rating": "3;6;6;6",
"rating_avg": 5.25,
"confidence": "4;2;3;5",
"confidence_avg": 3.5,
"soundness": "2;4;3;4",
"soundness_avg": 3.25,
"contribution": "2;3;3;2",
"contribution_avg": 2.5,
"presentation": "3;2;2;4",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.560417"
} | {
"id": "kA3zV7zzrp",
"metareview": "**Summary:** \n\nThe paper introduces COrAL, a framework designed to integrate iterative refinement directly into LLMs while maintaining computational efficiency. COrAL addresses limitations of autoregressive models, such as high inference latency and sequential dependency, by m... | {
"decision": "Reject"
} |
0K1OaL6XuK | 2410.12112v1 | Planning Anything with Rigor: General-Purpose Zero-Shot Planning with LLM-based Formalized Programming | {
"content": "## Abstract\n\nAbstract While large language models (LLMs) have recently demonstrated strong potential in solving planning problems, there is a trade-off between flexibility and complexity. LLMs, as zero-shot planners themselves, are still not capable of directly generating valid plans for complex plann... | [
{
"id": "mhaMSvVr6p",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper addresses the problem of solving planning problems that are given in natural language. The proposed algorithm they propose – LLMFP - is a workflow of ... | {
"rating": "1;6;6;6",
"rating_avg": 4.75,
"confidence": "4;4;3;3",
"confidence_avg": 3.5,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "1;2;3;3",
"contribution_avg": 2.25,
"presentation": "2;3;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.561211"
} | {
"id": "ks3sjYFZoZ",
"metareview": "The paper presents LLM-Based Formalized Programming (LLMFP), a framework for incorporating LLM's to solve natural language planning tasks. The framework uses an LLM iteratively with external planning tools to create a viable solution. The LLM is used to extract variables and con... | {
"decision": "Accept (Poster)"
} |
0L8wZ9WRah | 2406.13474v1 | Attention-aware Post-training Quantization without Backpropagation | {
"content": "## Abstract\n\nAbstract Quantization is a promising solution for deploying large-scale language models (LLMs) on resource-constrained devices.\nExisting quantization approaches, however, rely on gradient-based optimization, regardless of it being post-training quantization (PTQ) or quantization-aware tr... | [
{
"id": "p0sO5RRM3p",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper presents a novel post-training quantization (PTQ) method, termed BOA (Backpropagation-free Optimization for Attention-aware PTQ), targeting large langu... | {
"rating": "3;3;3;5",
"rating_avg": 3.5,
"confidence": "5;3;5;3",
"confidence_avg": 4,
"soundness": "3;2;2;3",
"soundness_avg": 2.5,
"contribution": "2;2;2;2",
"contribution_avg": 2,
"presentation": "2;3;1;3",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.561993"
} | {
"id": "xUqRjFQNgV",
"metareview": "This paper introduces BOA, a post-training quantization (PTQ) method for large language models (LLMs) that avoids backpropagation by leveraging attention-aware Hessian matrices to capture inter-layer dependencies within the attention module. BOA demonstrates improved quantizatio... | {
"decision": "Reject"
} |
0Lpz2o6NDE | 2410.10821v2 | Tex4D: Zero-shot 4D Scene Texturing with Video Diffusion Models | {
"content": "## Abstract\n\nAbstract 3D meshes are widely used in computer vision and graphics because of their efficiency in animation and minimal memory footprint.\nThey are extensively employed in movies, games, AR, and VR, leading to the creation of a vast number of mesh sequences.\nHowever, creating temporally ... | [
{
"id": "uLy8pSbxVO",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "This paper introduces a novel framework for generating textures for mesh sequences. The authors utilize a depth-conditioned video diffusion model to ensure tempor... | {
"rating": "3;5;5;5",
"rating_avg": 4.5,
"confidence": "4;4;3;4",
"confidence_avg": 3.75,
"soundness": "2;2;2;3",
"soundness_avg": 2.25,
"contribution": "2;2;2;2",
"contribution_avg": 2,
"presentation": "3;2;3;2",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.562783"
} | {
"id": "Yhr8RwLm4c",
"metareview": "This paper introduces a method for generating temporal and multi-view consistent textures for a mesh sequence in a training-free manner using a pretrained depth-conditioned video diffusion model. The proposed method builds upon previous 3D texture generation methods such as Sync... | {
"decision": "Reject"
} |
0N8yq8QwkD | 2405.17811v1 | Mani-GS: Gaussian Splatting Manipulation with Triangular Mesh | {
"content": "## Abstract\n\nAbstract Neural 3D representations such as Neural Radiance Fields (NeRF), excel at producing photo-realistic rendering results but lack the flexibility for manipulation and editing which is crucial for content creation. Previous works have attempted to address this issue by deforming a Ne... | [
{
"id": "kmvAuxnIuY",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 1,
"summary": "The authors improve the methodology of manipulating renderings generated by 3D Gaussian Splatting (3DGS). To achieve this, they propose the use of triangular mesh... | {
"rating": "5;5;5;5;5",
"rating_avg": 5,
"confidence": "5;4;5;5;4",
"confidence_avg": 4.6,
"soundness": "2;3;3;3;3",
"soundness_avg": 2.8,
"contribution": "2;2;1;2;2",
"contribution_avg": 1.8,
"presentation": "2;3;3;3;1",
"presentation_avg": 2.4
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.563510"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0NvSMb7xgC | 2306.13064v1 | Auditing Predictive Models for Intersectional Biases | {
"content": "## Abstract\n\nAbstract Predictive models that satisfy group fairness criteria in aggregate for members of a protected class, but do not guarantee subgroup fairness, could produce biased predictions for individuals at the intersection of two or more protected classes. To address this risk, we propose Co... | [
{
"id": "ECYPXCEe6S",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This study develops a new statistical test for identifying bias in prediction models across four different axes based on both the probabilistic outputs and the bi... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "4;2;3;3",
"confidence_avg": 3,
"soundness": "2;2;2;3",
"soundness_avg": 2.25,
"contribution": "2;3;2;3",
"contribution_avg": 2.5,
"presentation": "1;2;2;3",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.564349"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0OB3RVmTXE | 2410.08074v1 | Unstable Unlearning: The Hidden Risk of Concept Resurgence in Diffusion Models | {
"content": "## Abstract\n\nAbstract Text-to-image diffusion models rely on massive, web-scale datasets.\nTraining them from scratch is computationally expensive, and as a result, developers often prefer to make incremental updates to existing models. These updates often compose fine-tuning steps (to learn new conce... | [
{
"id": "0upDQD9vrB",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The paper examines a significant vulnerability in text-to-image diffusion models regarding the unlearning of unwanted concepts, termed \"concept resurgence.\" It ... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "4;4;4;3",
"confidence_avg": 3.75,
"soundness": "1;2;2;2",
"soundness_avg": 1.75,
"contribution": "3;2;2;3",
"contribution_avg": 2.5,
"presentation": "3;2;3;2",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.565343"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0QZcoGdmtJ | 2410.22235v1 | Auditing $f$-Differential Privacy in One Run | {
"content": "## Abstract\n\nAbstract Empirical auditing has emerged as a means of catching some of the flaws in the implementation of privacy-preserving algorithms. Existing auditing mechanisms, however, are either computationally inefficient – requiring multiple runs of the machine learning algorithms —- or subopti... | [
{
"id": "0Ixn0cWe8d",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper proposes an approach for auditing the guarantees of a differentially-private algorithm, which in contrast to other existing auditing schemes, does not ... | {
"rating": "3;6;8",
"rating_avg": 5.666666666666667,
"confidence": "5;3;3",
"confidence_avg": 3.6666666666666665,
"soundness": "3;3;3",
"soundness_avg": 3,
"contribution": "3;3;3",
"contribution_avg": 3,
"presentation": "1;2;2",
"presentation_avg": 1.6666666666666667
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.566724"
} | {
"id": "TtNX4Ijlhd",
"metareview": "## Summary of Contributions\n\nThis paper studies auditing of differential privacy (DP) with one run of the algorithm. Previous work (e.g. (Steinke et al., 2023)) studies the setting where we wish to test whether the algorithm satisfies $(\\epsilon, \\delta)$-DP for a single pai... | {
"decision": "Reject"
} |
0QePvFoqY6 | 2410.08107v2 | IncEventGS: Pose-Free Gaussian Splatting from a Single Event Camera | {
"content": "## Abstract\n\nAbstract Implicit neural representation and explicit 3D Gaussian Splatting (3D-GS) for novel view synthesis have achieved remarkable progress with frame-based camera ( e.g . RGB and RGB-D cameras) recently. Compared to frame-based camera, a novel type of bio-inspired visual sensor, i.e . ... | [
{
"id": "sirug4K9S7",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "This paper proposes IncEventGS, an incremental dense 3D reconstruction method using a single event camera. To incrementally recover the 3D scene, IncEventGS lever... | {
"rating": "3;5;5;5",
"rating_avg": 4.5,
"confidence": "5;4;4;4",
"confidence_avg": 4.25,
"soundness": "3;4;2;2",
"soundness_avg": 2.75,
"contribution": "3;3;3;2",
"contribution_avg": 2.75,
"presentation": "2;2;2;3",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.567504"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0R8JUzjSdq | 2406.05375v2 | LEMMA-RCA: A Large Multi-modal Multi-domain Dataset for Root Cause Analysis | {
"content": "## Abstract\n\nAbstract Root cause analysis (RCA) is crucial for enhancing the reliability and performance of complex systems. However, progress in this field has been hindered by the lack of large-scale, open-source datasets tailored for RCA. To bridge this gap, we introduce LEMMA-RCA, a large dataset ... | [
{
"id": "4tPtzY5wVb",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "This paper introduces Lemma-RCA, a dataset designed for root cause analysis. Lemma-RCA has distinctive and appreciable characteristics like large-scale, multi-mod... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "3;5;4;4",
"confidence_avg": 4,
"soundness": "2;2;2;3",
"soundness_avg": 2.25,
"contribution": "3;1;2;2",
"contribution_avg": 2,
"presentation": "2;3;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.568047"
} | {
"id": "scLwJLixZB",
"metareview": "The paper contributes \n(1) large dataset for root cause analysis\n(2) 14 baselines evaluated .\n\nThe reviewers feel that if ICLR had a separate Dataset track this would be a sure accept. \nThe methodological contributions are modest and hence it is not clear on how this paper ... | {
"decision": "Reject"
} |
0RHMnPj8no | 2410.05880v1 | Improved Sample Complexity for Private Nonsmooth Nonconvex Optimization | {
"content": "## Abstract\n\nAbstract We study differentially private (DP) optimization algorithms for stochastic and empirical\nobjectives which are neither smooth nor convex, and propose methods that return a Goldstein-stationary point with sample complexity bounds that improve on existing works.\nWe start by provi... | [
{
"id": "7BqRXnscjh",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The paper presents algorithms to improve sample complexity in differentially private (DP) nonsmooth, nonconvex (NSNC) optimization. The authors propose two zero-o... | {
"rating": "3;5;6;8",
"rating_avg": 5.5,
"confidence": "4;4;3;3",
"confidence_avg": 3.5,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;2;3",
"contribution_avg": 2.25,
"presentation": "2;3;3;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.568738"
} | {
"id": "d7LDcZxWVL",
"metareview": "This paper presents new single-pass and multi-pass algorithms for differentially private (DP) optimization of nonsmooth nonconvex. Authors provide new state-of-art art sample complexity for both these types of algorithms. Algorithm similar to Zhang et al., 2024, but authors prov... | {
"decision": "Reject"
} |
0RUQmLFF1D | 2410.04634v1 | Is What You Ask For What You Get? Investigating Concept Associations in Text-to-Image Models | {
"content": "## Abstract\n\nAbstract Text-to-image (T2I) models are increasingly used in impactful real-life applications. As such, there is a growing need to audit these models to ensure that they generate desirable, task-appropriate images.\nHowever, systematically inspecting the associations between prompts and g... | [
{
"id": "wU0qbrGnJn",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "The authors propose Concept2Concept, a framework that characterizes the conditional distributions of vision-language models using interpretable concepts and metri... | {
"rating": "3;5;6;6",
"rating_avg": 5,
"confidence": "3;3;4;4",
"confidence_avg": 3.5,
"soundness": "2;2;3;4",
"soundness_avg": 2.75,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "2;3;4;4",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.569532"
} | {
"id": "JPpSN9qoFV",
"metareview": "This paper introduces a framework to audit T2I models and prompt datasets, which analyzes association between text prompts and generated images with interpretable tools. Four knowledgeable reviewers went over this submission. The reviewers recognized that the paper tackles an im... | {
"decision": "Reject"
} |
0UCkWfcfb9 | 2406.07657v1 | OPTune: Efficient Online Preference Tuning | {
"content": "## Abstract\n\nAbstract Reinforcement learning with human feedback (RLHF) is critical for aligning Large Language Models (LLMs) with human preference.\nCompared to the widely studied offline version of RLHF, e.g. direct preference optimization (DPO), recent works have shown that the online variants achi... | [
{
"id": "ZTUPRCGmf0",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper targets LLM alignment with human preferences in an online manner. OPTune involves two main strategies to reduce computational costs while maintaining a... | {
"rating": "3;3;5;5;5",
"rating_avg": 4.2,
"confidence": "3;5;4;3;4",
"confidence_avg": 3.8,
"soundness": "2;1;3;3;3",
"soundness_avg": 2.4,
"contribution": "2;1;2;2;2",
"contribution_avg": 1.8,
"presentation": "1;2;3;2;3",
"presentation_avg": 2.2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.570519"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0UCoWxPhQ4 | 2406.01130v1 | SAVA: Scalable Learning-Agnostic Data Valuation | {
"content": "## Abstract\n\nAbstract Selecting suitable data for training machine learning models is crucial since large, web-scraped, real datasets contain noisy artifacts that affect the quality and relevance of individual data points. These artifacts will impact the performance and generalization of the model. We... | [
{
"id": "S2ePN70JFQ",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper investigates of the problem of extending Optimal Transport (OT) distance-based data valuation methods for larger scale problems. The paper points out t... | {
"rating": "5;6;6",
"rating_avg": 5.666666666666667,
"confidence": "3;4;4",
"confidence_avg": 3.6666666666666665,
"soundness": "3;3;3",
"soundness_avg": 3,
"contribution": "2;3;3",
"contribution_avg": 2.6666666666666665,
"presentation": "3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.571549"
} | {
"id": "8NxyTQRdY9",
"metareview": "This paper introduces SAVA, a new and more scalable variant of the data valuation method LAVA. LAVA was restricted by the amount of computation required when computing the optimal transport distance. SAVA addresses this problem by computing these metrics in a batch manner. Exper... | {
"decision": "Accept (Poster)"
} |
0YXckVo7Kw | 2410.09733v1 | MMCOMPOSITION: Revisiting the Compositionality of Pre-trained Vision-Language Models | {
"content": "## Abstract\n\nAbstract The advent of large Vision-Language Models (VLMs) has significantly advanced multimodal understanding,\nenabling more sophisticated and accurate integration of visual and textual information across various tasks, including image and video captioning, visual question answering, an... | [
{
"id": "A8i0h6PeY7",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper introduces MMComposition, a QA benchmark that evaluates the compositional capabilities of modern vision-language models. MMComposition encompasses a ra... | {
"rating": "5;5;5;6",
"rating_avg": 5.25,
"confidence": "3;4;4;4",
"confidence_avg": 3.75,
"soundness": "3;2;2;3",
"soundness_avg": 2.5,
"contribution": "3;2;2;3",
"contribution_avg": 2.5,
"presentation": "3;2;2;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.572304"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0aTIvSJ83I | 2406.07107v3 | Agnostic Sharpness-Aware Minimization | {
"content": "## Abstract\n\nAbstract Sharpness-aware minimization (SAM) has been instrumental in improving deep neural network training by minimizing both the training loss and the sharpness of the loss landscape, leading the model into flatter minima that are associated with better generalization properties. In ano... | [
{
"id": "xdoqYtJqJC",
"initial_rating": 3,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This work aims to combine Sharpness-Aware Minimization with Model-Agnostic Meta-Learning, by having worst-case robustified versions of the loss in both the inner ... | {
"rating": "3;3;3;3",
"rating_avg": 3,
"confidence": "5;5;4;3",
"confidence_avg": 4.25,
"soundness": "2;2;2;2",
"soundness_avg": 2,
"contribution": "1;2;1;2",
"contribution_avg": 1.5,
"presentation": "2;2;2;2",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.573113"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0bcRCD7YUx | 2406.05370v2 | VALL-E 2: Neural Codec Language Models are Human Parity Zero-Shot Text to Speech Synthesizers | {
"content": "## Abstract\n\nAbstract This paper introduces VALL-E 2, the latest advancement in neural codec language models that marks a milestone in zero-shot text-to-speech synthesis (TTS), achieving human parity for the first time .\nBased on its predecessor, VALL-E, the new iteration introduces two significant e... | [
{
"id": "qfWhMcEIWF",
"initial_rating": 3,
"confidence": 5,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "VALL-E 2 is an LM-based TTS model based on VALL-E. It proposes two new methods:\n\n1. **Repetition Aware Sampling**: In this method, during the sampling process, ... | {
"rating": "3;3;6;8",
"rating_avg": 5,
"confidence": "4;5;4;5",
"confidence_avg": 4.5,
"soundness": "3;2;3;4",
"soundness_avg": 3,
"contribution": "2;2;3;4",
"contribution_avg": 2.75,
"presentation": "3;2;3;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.573941"
} | {
"id": "qRTy09pQLR",
"metareview": "The paper proposes to incorporate a new sampling procedure and a different grouping of codes to improve VALL-E for zero-shot TTS.\n\nI recommend a rejection because all reviewers unanimously agree that paper lacks novelty despite the strong performance. \n\nThere are various oth... | {
"decision": "Reject"
} |
0cadcLKbt7 | 2410.00531v1 | TPI-LLM: Serving 70B-scale LLMs Efficiently on Low-resource Edge Devices | {
"content": "## Abstract\n\nAbstract Large model inference is shifting from cloud to edge due to concerns about the privacy of user interaction data. However, edge devices often struggle with limited computing power, memory, and bandwidth, requiring collaboration across multiple devices to run and speed up LLM infer... | [
{
"id": "YeIB0tSnEt",
"initial_rating": 1,
"confidence": 5,
"soundness": 1,
"contribution": 1,
"presentation": 1,
"summary": "The paper introduces a technique to run 70B LLM on CPU based (edge) devices. The system uses a tensor parallel framework to distribute attention heads across mult... | {
"rating": "1;5;5;5",
"rating_avg": 4,
"confidence": "5;4;4;4",
"confidence_avg": 4.25,
"soundness": "1;3;2;2",
"soundness_avg": 2,
"contribution": "1;2;3;2",
"contribution_avg": 2,
"presentation": "1;3;2;2",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.574632"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0fD3iIBhlV | 2405.15471v1 | Emergence of a High-Dimensional Abstraction Phase in Language Transformers | {
"content": "## Abstract\n\nAbstract A language model (LM) is a mapping from a linguistic context to an output token. However, much remains to be known about this mapping, including how its geometric properties relate to its function. We take a high-level geometric approach to its analysis, observing, across five pr... | [
{
"id": "ttYcoW2vwp",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The paper uses the technique of intrinsic dimension estimation as a tool for analyzing properties of different transformer LLM layers. 5 different LLMs are analyz... | {
"rating": "5;5;6;6;8",
"rating_avg": 6,
"confidence": "3;3;4;3;5",
"confidence_avg": 3.6,
"soundness": "2;3;2;3;3",
"soundness_avg": 2.6,
"contribution": "3;3;2;2;3",
"contribution_avg": 2.6,
"presentation": "3;3;3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.575692"
} | {
"id": "yZxEhQm1mU",
"metareview": "The paper employs intrinsic dimension (ID) estimation as a technique to analyze the properties of different layers in transformer-based LLMs. While inspired by previous work, this study expands the scope by including five LLMs and introducing more extensive probing and downstrea... | {
"decision": "Accept (Poster)"
} |
0fhzSFsGUT | 2406.02052v1 | PETRA: Parallel End-to-end Training with Reversible Architectures | {
"content": "## Abstract\n\nAbstract Reversible architectures have been shown to be capable of performing on par with their non-reversible architectures, being applied in deep learning for memory savings and generative modeling. In this work, we show how reversible architectures can solve challenges in parallelizing... | [
{
"id": "T5hGsFAPG5",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 4,
"summary": "The authors propose a new algorithm for training reversible models. Compared to backpropagation, it can be run on each layer in parallel and with a reduced memory... | {
"rating": "5;6;8;8",
"rating_avg": 6.75,
"confidence": "4;4;5;4",
"confidence_avg": 4.25,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;3;3;4",
"contribution_avg": 3,
"presentation": "3;4;4;3",
"presentation_avg": 3.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.576433"
} | {
"id": "Q8UL1eppFY",
"metareview": "The authors propose a new algorithm for training reversible models. Compared to backpropagation, it can be run on each layer in parallel and with a reduced memory cost. They show empirically the advantages of their algorithm on RevNet models for image classification.",
"additi... | {
"decision": "Accept (Spotlight)"
} |
0gOQeSHNX1 | 2410.06405v1 | Tackling the Abstraction and Reasoning Corpus with Vision Transformers: the Importance of 2D Representation, Positions, and Objects | {
"content": "## Abstract\n\nAbstract The Abstraction and Reasoning Corpus (ARC) is a popular benchmark focused on visual reasoning in the evaluation of Artificial Intelligence systems. In its original framing, an ARC task requires solving a program synthesis problem over small 2D images using a few input-output trai... | [
{
"id": "7VV9O4aWod",
"initial_rating": 5,
"confidence": 5,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper studies vision transformer in abstract visual reasoning ARC tasks which do not include any text or background knowledge, and focus purely on visual abs... | {
"rating": "3;5;5;8",
"rating_avg": 5.25,
"confidence": "4;4;5;4",
"confidence_avg": 4.25,
"soundness": "3;2;3;3",
"soundness_avg": 2.75,
"contribution": "2;3;2;3",
"contribution_avg": 2.5,
"presentation": "2;3;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.577177"
} | {
"id": "caKFIgzimO",
"metareview": "This is an interesting paper that designs a specific approach to tackle the ARC-AGI benchmark using object-centric priors and other task-specific components on top of a vision transformer architecture. \n\nWhile the method is interesting and the results look encouraging, reviewe... | {
"decision": "Reject"
} |
0gVatTOgEv | 2410.07172v1 | Glider: Global and Local Instruction-Driven Expert Router | {
"content": "## Abstract\n\nAbstract The availability of performant pre-trained models has led to a proliferation of fine-tuned expert models that are specialized to a particular domain or task. This has enabled the creation of powerful and adaptive routing-based “Model MoErging\" (Yadav et al., 2024 ) methods with ... | [
{
"id": "faduwKmugv",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 1,
"summary": "This paper focuses on addressing the trade-off between performance improvement and generalization ability in expert modules. It assumes that this issue arises fro... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "3;4;3;3",
"confidence_avg": 3.25,
"soundness": "2;2;3;2",
"soundness_avg": 2.25,
"contribution": "2;1;2;2",
"contribution_avg": 1.75,
"presentation": "3;2;3;1",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.578047"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0gqCIaBRQ9 | 2403.04236v1 | Regularized DeepIV with Model Selection | {
"content": "## Abstract\n\nAbstract In this paper, we study nonparametric estimation of instrumental variable (IV) regressions. While recent advancements in machine learning have introduced flexible methods for IV estimation, they often encounter one or more of the following limitations: (1) restricting the IV regr... | [
{
"id": "epyZtG50iA",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "This paper studies a two stage procedure for regression in the scenario where the errors are not conditionally independent. They first learn a conditional density... | {
"rating": "5;5;5;6",
"rating_avg": 5.25,
"confidence": "4;4;3;3",
"confidence_avg": 3.5,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;3;3;3",
"contribution_avg": 2.75,
"presentation": "2;3;2;4",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.579277"
} | {
"id": "bMXnxptYot",
"metareview": "This study represents a solid and good contribution. However, the novelty and usefulness of the approach have not been fully appreciated, at least by the reviewers. The paper's improvements in norm selection and error evaluation are not presented in a form that is easily accepte... | {
"decision": "Reject"
} |
0h6v4SpLCY | 2402.11981v2 | Universal generalization guarantees for Wasserstein distributionally robust models | {
"content": "## Abstract\n\nAbstract Distributionally robust optimization has emerged as an attractive way\nto train robust machine learning models, capturing data uncertainty and distribution shifts. Recent statistical analyses have proved that robust models built from Wasserstein ambiguity sets have nice generaliz... | [
{
"id": "UB7x1jbnJK",
"initial_rating": 8,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper presents novel bounds on for the DRO loss using the Wasserstein distance. In particular, they address the question of finding the minimal $\\rho$ used ... | {
"rating": "6;6;8",
"rating_avg": 6.666666666666667,
"confidence": "3;3;4",
"confidence_avg": 3.3333333333333335,
"soundness": "3;3;3",
"soundness_avg": 3,
"contribution": "3;3;3",
"contribution_avg": 3,
"presentation": "3;3;4",
"presentation_avg": 3.3333333333333335
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.580490"
} | {
"id": "SF5HnFQRqZ",
"metareview": "This paper analyzes distributionally robust optimization, which effectively addresses data uncertainty and distribution shifts in training robust machine learning models. Existing generalization guarantees are often approximate or limited to specific cases with hard-to-verify ... | {
"decision": "Accept (Spotlight)"
} |
0hyShAPeBj | 2410.04201v1 | IT$^3$: Idempotent Test-Time Training | {
"content": "## Abstract\n\nAbstract This paper introduces Idempotent Test-Time Training (IT 3 ),\na novel approach to addressing the challenge of distribution shift.\nWhile supervised-learning methods assume matching train and test distributions, this is rarely the case for machine learning systems deployed in the ... | [
{
"id": "TjkaMzbhqw",
"initial_rating": 3,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 4,
"summary": "- This paper proposes a test-time-training based approach to address the distribution shift or OOD generalization problem by learning models that are idempotent.\... | {
"rating": "3;3;3;5",
"rating_avg": 3.5,
"confidence": "5;3;3;4",
"confidence_avg": 3.75,
"soundness": "2;2;2;2",
"soundness_avg": 2,
"contribution": "2;2;2;3",
"contribution_avg": 2.25,
"presentation": "3;2;4;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.581380"
} | {
"id": "9QOQR3UzKU",
"metareview": "This paper addresses the test time training problem to handle distribution shifts. The paper proposes a method to learn a model that is idempotent. This is an interesting idea which the reviewers appreciated. The authors' rebuttal was considered and some of the reviewers also en... | {
"decision": "Reject"
} |
0iscEAo2xB | 2411.07414v1 | Comparing Targeting Strategies for Maximizing Social Welfare with Limited Resources | {
"content": "## Abstract\n\nAbstract Machine learning is increasingly used to select which individuals receive limited-resource interventions in domains such as human services, education, development, and more. However, it is often not apparent what the right quantity is for models to predict. In particular, policym... | [
{
"id": "Cbv8nLwP1O",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The authors compare the utility of targeting interventions based on estimated treatment effects with the utility of targeting based on predicted risk. The former ... | {
"rating": "5;5;6;10",
"rating_avg": 6.5,
"confidence": "3;4;4;4",
"confidence_avg": 3.75,
"soundness": "2;3;3;4",
"soundness_avg": 3,
"contribution": "3;3;2;4",
"contribution_avg": 3,
"presentation": "3;2;3;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.582285"
} | {
"id": "ZWPP9TKp9r",
"metareview": "The paper compares two targeting strategies for allocating limited resources in social welfare applications: risk-based targeting and treatment-effect-based targeting. Using data from five real-world randomized controlled trials (RCTs), the authors find that targeting based on t... | {
"decision": "Accept (Poster)"
} |
0jmFRA64Vw | 2403.09904v1 | FedComLoc: Communication-Efficient Distributed Training of Sparse and Quantized Models | {
"content": "## Abstract\n\nAbstract Federated Learning (FL) has garnered increasing attention due to its unique characteristic of allowing heterogeneous clients to process their private data locally and interact with a central server, while being respectful of privacy.\nA critical bottleneck in FL is the communicat... | [
{
"id": "c6LSimPWAl",
"initial_rating": 3,
"confidence": 4,
"soundness": 1,
"contribution": 1,
"presentation": 2,
"summary": "This paper presents an empirical study on a new algorithm, FedComLoc, which extends Scaffnew by integrating compression techniques: TopK and quantization. Three s... | {
"rating": "3;3;3",
"rating_avg": 3,
"confidence": "4;4;4",
"confidence_avg": 4,
"soundness": "2;2;1",
"soundness_avg": 1.6666666666666667,
"contribution": "1;2;1",
"contribution_avg": 1.3333333333333333,
"presentation": "2;2;2",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.583170"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0koPj0cJV6 | 2410.02099v1 | A Watermark for Black-Box Language Models | {
"content": "## Abstract\n\nAbstract Watermarking has recently emerged as an effective strategy for detecting the outputs of large language models (LLMs). Most existing schemes require white-box access to the model’s next-token probability distribution, which is typically not accessible to downstream users of an LLM... | [
{
"id": "j0gMeHkYdt",
"initial_rating": 1,
"confidence": 5,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper proposes an LLM watermarking scheme that is applicable in black-box scenarios, i.e., when the party watermarking the text does not have access to the sa... | {
"rating": "3;5;5;6;6",
"rating_avg": 5,
"confidence": "4;3;3;3;4",
"confidence_avg": 3.4,
"soundness": "2;2;2;4;3",
"soundness_avg": 2.6,
"contribution": "1;3;2;2;3",
"contribution_avg": 2.2,
"presentation": "1;2;2;3;3",
"presentation_avg": 2.2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.583855"
} | {
"id": "tRBnHlBzxo",
"metareview": "The paper presents a novel watermark scheme for black-box language models. After a extensive discussion, there are several critical issues remain unresolved. Reviewer 3sQV raised the concerns on the definition of \"distortion-free\". Reviewer Lnyz noted that the proposed waterma... | {
"decision": "Reject"
} |
0mJZplhexS | 2406.17117v2 | Speeding Up Image Classifiers with Little Companions | {
"content": "## Abstract\n\nAbstract Scaling up neural networks has been a key recipe to the success of large language and vision models. However, in practice, up-scaled models can be disproportionately costly in terms of computations, providing only marginal improvements in performance; for example, EfficientViT-L3... | [
{
"id": "Tf4fZo2uV0",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The paper proposes a method named Little-Big to accelerate image classification with neural networks. Little-Big uses a light-weight model to quickly classify all... | {
"rating": "3;5;5;5",
"rating_avg": 4.5,
"confidence": "3;4;3;3",
"confidence_avg": 3.25,
"soundness": "3;2;2;3",
"soundness_avg": 2.5,
"contribution": "2;2;3;2",
"contribution_avg": 2.25,
"presentation": "3;4;3;3",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.584569"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0mo2yqOS6Z | 2407.00356v1 | Enhancing Accuracy and Parameter Efficiency of Neural Representations for Network Parameterization | {
"content": "## Abstract\n\nAbstract In this work, we investigate the fundamental trade-off regarding accuracy and parameter efficiency in the parameterization of neural network weights using predictor networks. We present a surprising finding that, when recovering the original model accuracy is the sole objective, ... | [
{
"id": "X3pAO96v3X",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The author study the fundamental trade-off regarding accuracy and parameter efficiency in neural network weight parameterization using predictor networks. They pr... | {
"rating": "5;5;5;6;6",
"rating_avg": 5.4,
"confidence": "5;4;4;4;3",
"confidence_avg": 4,
"soundness": "1;3;2;3;3",
"soundness_avg": 2.4,
"contribution": "2;3;1;3;3",
"contribution_avg": 2.4,
"presentation": "3;2;2;3;3",
"presentation_avg": 2.6
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.585177"
} | {
"id": "tlvMsSM0jZ",
"metareview": "This paper explores the trade-off between accuracy and parameter efficiency in neural networks using smaller predictor networks to predict neural network weights. With successive rounds of reconstruction, by decoupling the reconstruction and distillation processes, the model's ... | {
"decision": "Reject"
} |
0mtz0pet1z | 2409.13097v1 | Incremental Causal Effect for Time to Treatment Initialization | {
"content": "## Abstract\n\nAbstract We consider time to treatment initialization. This can commonly occur in preventive medicine, such as disease screening and vaccination; it can also occur with non-fatal health conditions such as HIV infection without the onset of AIDS; or in tech industry where items wait to be ... | [
{
"id": "H0sAXkDZpy",
"initial_rating": 3,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 1,
"summary": "In many settings, whether or not a subject receives a treatment at any given time point may be a function of their covariates. In these settings, we can reason a... | {
"rating": "3;6;6;6",
"rating_avg": 5.25,
"confidence": "3;4;3;2",
"confidence_avg": 3,
"soundness": "3;2;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;2;3",
"contribution_avg": 2.25,
"presentation": "1;3;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.586127"
} | {
"id": "chn5G6uwZR",
"metareview": "The paper introduces a continuous-time version of the \"time to treatment\" problem, which I found to be particularly relevant in applications such as AI for healthcare and is, to the best of my knowledge, understudied. It can be valuable for the ICLR community to be exposed to ... | {
"decision": "Accept (Poster)"
} |
0n4bS0R5MM | 2407.12781v2 | VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control | {
"content": "## Abstract\n\nAbstract Modern text-to-video synthesis models demonstrate coherent, photorealistic generation of complex videos from a text description.\nHowever, most existing models lack fine-grained control over camera movement, which is critical for downstream applications related to content creatio... | [
{
"id": "BLNCUZyzqR",
"initial_rating": 8,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper presents a camera control method for transformer-based video generation models that enhances control while ensuring visual quality. The proposed approa... | {
"rating": "5;6;6;8;8",
"rating_avg": 6.6,
"confidence": "5;4;5;3;4",
"confidence_avg": 4.2,
"soundness": "2;3;3;3;3",
"soundness_avg": 2.8,
"contribution": "2;3;2;3;2",
"contribution_avg": 2.4,
"presentation": "2;3;3;3;3",
"presentation_avg": 2.8
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.586872"
} | {
"id": "C4CbujugSK",
"metareview": "This paper proposes a camera-control video diffusion model. The central claimed contribution is its transformer-based architecture. The method builds on SnapVideo and introduces a new attn-based conditional module for injecting the camera information. STOA performances are obtai... | {
"decision": "Accept (Poster)"
} |
0nJt9aVGtl | 2410.09002v1 | WaveDiffusion: Exploring Full Waveform Inversion via Joint Diffusion in the Latent Space | {
"content": "## Abstract\n\nAbstract Full Waveform Inversion (FWI) is a vital technique for reconstructing high-resolution subsurface velocity maps from seismic waveform data, governed by partial differential equations (PDEs) that model wave propagation. Traditional machine learning approaches typically map seismic ... | [
{
"id": "LbfJkmBoht",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The manuscript introduces a new approach to invert acoustic wave equation data based on a joint generative process. Although there were earlier papers on the use ... | {
"rating": "3;3;6;6",
"rating_avg": 4.5,
"confidence": "5;5;3;4",
"confidence_avg": 4.25,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "2;1;3;3",
"contribution_avg": 2.25,
"presentation": "2;3;4;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.587833"
} | {
"id": "U76p6QSser",
"metareview": "This paper addresses full waveform inversion by training autoencoders on models and data, with shared latent spaces, followed by training a diffusion model in the latent space to generate samples. While the reviewers acknowledged the idea as reasonable and potentially effective,... | {
"decision": "Reject"
} |
0nxocR2qx4 | 2404.04102v2 | ROPO: Robust Preference Optimization for Large Language Models | {
"content": "## Abstract\n\nAbstract Preference alignment is pivotal for empowering large language models (LLMs) to generate helpful and harmless responses.\nHowever, the performance of preference alignment is highly sensitive to the prevalent noise in the preference data.\nRecent efforts for this problem either mar... | [
{
"id": "bwuYVcLt6L",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "The paper studies preference alignment under the condition when there are poorly-annotated preference pairs. The authors propose a robust preference optimization ... | {
"rating": "5;6;6",
"rating_avg": 5.666666666666667,
"confidence": "4;4;4",
"confidence_avg": 4,
"soundness": "2;3;3",
"soundness_avg": 2.6666666666666665,
"contribution": "2;3;3",
"contribution_avg": 2.6666666666666665,
"presentation": "3;3;2",
"presentation_avg": 2.6666666666666665
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.588528"
} | {
"id": "Wrs9GORVAK",
"metareview": "The paper presents the RObust Preference Optimization (ROPO) framework, a novel approach to improving preference alignment in LLMs under noisy conditions. By integrating a noise-tolerant loss function and robustness-guided rejection sampling, ROPO aims to mitigate the impact of ... | {
"decision": "Reject"
} |
0oWGVvC6oq | 2405.16581v3 | On Bits and Bandits: Quantifying the Regret-Information Trade-off | {
"content": "## Abstract\n\nAbstract In many sequential decision problems, an agent performs a repeated task. He then suffers regret and obtains information that he may use in the following rounds. However, sometimes the agent may also obtain information and avoid suffering regret by querying external sources. We st... | [
{
"id": "b90eqB53zF",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper studies regret minimization when extra information about the prior is revealed. \nIn particular, the authors consider contextual bandit problems, where... | {
"rating": "5;6;6;8",
"rating_avg": 6.25,
"confidence": "3;3;3;3",
"confidence_avg": 3,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "3;2;2;3",
"contribution_avg": 2.5,
"presentation": "1;3;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.589381"
} | {
"id": "QvqvN793er",
"metareview": "This paper develops new information-theoretic methods for analyzing sequential decision-making algorithms. The methods are then used to recover existing lower bounds for a range of sequential decision-making problems. The reviewers gave overall very positive feedback and unanimo... | {
"decision": "Accept (Poster)"
} |
0pLCDJVVRD | 2408.12578v2 | A Percolation Model of Emergence: Analyzing Transformers Trained on a Formal Language | {
"content": "## Abstract\n\nAbstract Increase in data, size, or compute can lead to sudden learning of specific capabilities by a neural network—a phenomenon often called “emergence”.\nBeyond scientific understanding, establishing the causal factors underlying such emergent capabilities is crucial to enable risk reg... | [
{
"id": "XfEoujJ4Kt",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 4,
"summary": "The paper studies emergent capabilities in transformers via two case studies. In the first study, they look at learning of formal languages (in particular, a lang... | {
"rating": "5;5;6;8",
"rating_avg": 6,
"confidence": "4;3;3;3",
"confidence_avg": 3.25,
"soundness": "3;2;3;3",
"soundness_avg": 2.75,
"contribution": "3;2;3;3",
"contribution_avg": 2.75,
"presentation": "3;4;2;4",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.590579"
} | {
"id": "UehPPcYApd",
"metareview": "This paper examines emergence in neural networks through the lens of percolation theory, using transformers trained on formal languages. The work provides a formal definition of emergence, demonstrates phase transitions in learning, and establishes theoretical connections to per... | {
"decision": "Accept (Poster)"
} |
0pbxX2jatP | 2410.13204v1 | Measuring Free-Form Decision-Making Inconsistency of Language Models in Military Crisis Simulations | {
"content": "## Abstract\n\nAbstract There is an increasing interest in using language models (LMs) for automated decision-making, with multiple countries actively testing LMs to aid in military crisis decision-making.\nTo scrutinize relying on LM decision-making in high-stakes settings, we examine the inconsistency... | [
{
"id": "um4JZR1VYr",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 4,
"summary": "This paper investigates the inconsistency of large language models (LLMs) in terms of ablations like sentence order and semantics when applied in war games. The a... | {
"rating": "3;5;5",
"rating_avg": 4.333333333333333,
"confidence": "5;4;4",
"confidence_avg": 4.333333333333333,
"soundness": "3;2;3",
"soundness_avg": 2.6666666666666665,
"contribution": "2;2;2",
"contribution_avg": 2,
"presentation": "4;3;4",
"presentation_avg": 3.6666666666666665
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.591714"
} | {
"id": "uIxSnD0iRx",
"metareview": "**Summary:**\n\nThe authors analyze how previously observed inconsistency in outputs of LLMs influences their behavior during military decision-making and mental health simulations. They use BERTScore for identifying model inconsistency. Their findings indicate LLMs may be too i... | {
"decision": "Reject"
} |
0py3h7pops | 2410.10160v1 | Will the Inclusion of Generated Data Amplify Bias Across Generations in Future Image Classification Models? | {
"content": "## Abstract\n\nAbstract {NoHyper} † † ⋆ ⋆ \\star ⋆ indicates equal contribution with random order. As the demand for high-quality training data escalates, researchers have increasingly turned to generative models to create synthetic data, addressing data scarcity and enabling continuous model improvemen... | [
{
"id": "mIVTD43Gqt",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper investigates the bias implications of incorporating generated data from generative models into training downstream tasks. The authors propose an iterat... | {
"rating": "3;5;6;6",
"rating_avg": 5,
"confidence": "4;3;3;4",
"confidence_avg": 3.5,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;2;2",
"contribution_avg": 2,
"presentation": "2;2;2;3",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.592714"
} | {
"id": "J7ZL5CrrzU",
"metareview": "This paper explores the implications of generated data being used to train generative models, which is a realistic setting given how data is often shared and scraped on the internet. Specific focus is given to how biases can emerge or be exacerbated across subgroups in the data.... | {
"decision": "Reject"
} |
0qrTH5AZVt | 2410.12439v1 | ConLUX: Concept-Based Local Unified Explanations | {
"content": "## Abstract\n\nAbstract With the rapid advancements of various machine learning models, there is a significant demand for model-agnostic explanation techniques, which can explain these models across different architectures.\nMainstream model-agnostic explanation techniques generate local explanations ba... | [
{
"id": "8F0yQmBuVX",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The paper proposes to use foundation models to discover concepts to augment on methods like LIME to provide concept-based local explanations. The method is evalua... | {
"rating": "3;5;6",
"rating_avg": 4.666666666666667,
"confidence": "4;3;4",
"confidence_avg": 3.6666666666666665,
"soundness": "3;2;3",
"soundness_avg": 2.6666666666666665,
"contribution": "2;2;2",
"contribution_avg": 2,
"presentation": "2;2;3",
"presentation_avg": 2.3333333333333335
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.593517"
} | {
"id": "QXkUwT7jxT",
"metareview": "The paper introduces an intriguing method using large pre-trained models for concept discovery and perturbation, but several issues limit its reliability and generalizability. Concept-level perturbations can distort inputs, failing to capture local decision boundaries effectivel... | {
"decision": "Reject"
} |
0rS9o1uKqu | 2410.16884v1 | Training-Like Data Reconstruction | {
"content": "## Abstract\n\nAbstract Machine Learning models are often trained on proprietary and private data that cannot be shared, though the trained models themselves are distributed openly assuming that sharing model weights is privacy preserving, as training data is not expected to be inferred from the model w... | [
{
"id": "OpnoQGTv5y",
"initial_rating": 3,
"confidence": 4,
"soundness": 1,
"contribution": 2,
"presentation": 2,
"summary": "This paper introduces a new method for reconstructing data that resembles the training dataset of an ML model. The method is based on two steps: inversion, where ... | {
"rating": "1;3;3;3",
"rating_avg": 2.5,
"confidence": "5;3;3;4",
"confidence_avg": 3.75,
"soundness": "3;2;2;1",
"soundness_avg": 2,
"contribution": "1;2;2;2",
"contribution_avg": 1.75,
"presentation": "2;2;2;2",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.594195"
} | {
"id": "N4GM97Z0qL",
"metareview": "The reviewers were unfortunately not excited about the paper. In particular there were concerns about the experimental setup, complexity of the scheme, and the scale of reconstruction.",
"additional_comments": "The authors could not convince the reviewers to bump up the score.... | {
"decision": "Reject"
} |
0sU4myabw1 | 2411.00004v1 | RapidDock: Unlocking Proteome-scale Molecular Docking | {
"content": "## Abstract\n\nAbstract Accelerating molecular docking – the process of predicting how molecules bind to protein targets – could boost small-molecule drug discovery and revolutionize medicine. Unfortunately, current molecular docking tools are too slow to screen potential drugs against all relevant prot... | [
{
"id": "salc2kHrfU",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 4,
"summary": "The authors tackle the problem of proteome-scale docking, the goal of which is predicting the binding pose of a ligand against many thousands of proteins. To do t... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "3;3;5;4",
"confidence_avg": 3.75,
"soundness": "3;1;3;3",
"soundness_avg": 2.5,
"contribution": "2;3;2;2",
"contribution_avg": 2.25,
"presentation": "3;1;2;4",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.594728"
} | {
"id": "vjzWT4Svbc",
"metareview": "The paper proposes a rapid docking method based on a Transformer. The model takes as input a ligand and a protein, both of which are tokenized, and their 3D structures are passed through all self-attention layers. The protein is embedded using ESM-2.\n\nThe method achieves stron... | {
"decision": "Reject"
} |
0sr8bS4S2H | 2410.18603v1 | AgentStore: Scalable Integration of Heterogeneous Agents As Specialized Generalist Computer Assistant | {
"content": "## Abstract\n\nAbstract Digital agents capable of automating complex computer tasks have attracted considerable attention due to their immense potential to enhance human-computer interaction. However, existing agent methods exhibit deficiencies in their generalization and specialization capabilities, es... | [
{
"id": "XPeGrrY1an",
"initial_rating": 5,
"confidence": 4,
"soundness": 1,
"contribution": 2,
"presentation": 2,
"summary": "This paper presents AgentStore, which allows integrating and dynamically use a range of domain-specific agents (the difference is mainly the base model, how the m... | {
"rating": "3;3;5;6",
"rating_avg": 4.25,
"confidence": "3;4;4;4",
"confidence_avg": 3.75,
"soundness": "2;2;1;3",
"soundness_avg": 2,
"contribution": "2;2;2;3",
"contribution_avg": 2.25,
"presentation": "3;2;2;4",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.595424"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0tAXMiSufG | 2407.14622v1 | BOND: Aligning LLMs with Best-of-N Distillation | {
"content": "## Abstract\n\nAbstract Reinforcement learning from human feedback (RLHF) is a key driver of quality and safety in state-of-the-art large language models.\nYet, a surprisingly simple and strong inference-time strategy is Best-of- N sampling that selects the best generation among N 𝑁 N italic_N candidat... | [
{
"id": "yh6wouxHVw",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper is focusing on the RLHF alignment problem, in particular on emulating the Best-of-N distribution which is known to perform very well, but is very costly... | {
"rating": "5;6;6;6;6",
"rating_avg": 5.8,
"confidence": "4;4;4;3;4",
"confidence_avg": 3.8,
"soundness": "3;4;3;4;3",
"soundness_avg": 3.4,
"contribution": "3;3;3;3;3",
"contribution_avg": 3,
"presentation": "2;2;2;3;3",
"presentation_avg": 2.4
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.596189"
} | {
"id": "rBimtAsa0d",
"metareview": "This paper proposes to align an LM based on the best-of-N distribution. This paper is backed up by formal mathematical characterizations (which may be useful for others working with these types of distributions), as well as solid empirical results. There were some concerns with ... | {
"decision": "Accept (Poster)"
} |
0tAn34IkXI | 2406.15664v3 | Flat Posterior Does Matter For Bayesian Model Averaging | {
"content": "## Abstract\n\nAbstract Bayesian neural network (BNN) approximates the posterior distribution of model parameters and utilizes the posterior for prediction via Bayesian Model Averaging (BMA). The quality of the posterior approximation is critical for achieving accurate and robust predictions. It is know... | [
{
"id": "tFwerVl47c",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "The paper proposes a sharpness-aware Bayesian neural network (BNNs) to ensure the found modes are flat. A new Bayesian transfer learning scheme is also developed ... | {
"rating": "3;3;5;6",
"rating_avg": 4.25,
"confidence": "3;4;3;4",
"confidence_avg": 3.5,
"soundness": "3;2;2;3",
"soundness_avg": 2.5,
"contribution": "1;2;2;3",
"contribution_avg": 2,
"presentation": "3;3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.597081"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0tIiMNNmdm | 2311.12618v1 | Limitations of measure-first protocols in quantum machine learning | {
"content": "## Abstract\n\nAbstract In recent works, much progress has been made with regards to so-called randomized measurement strategies, which include the famous methods of classical shadows and shadow tomography.\nIn such strategies, unknown quantum states are first measured (or “learned”), to obtain classica... | [
{
"id": "IfqiYF5eou",
"initial_rating": 3,
"confidence": 3,
"soundness": 1,
"contribution": 1,
"presentation": 2,
"summary": "The authors design a quantum machine learning task that exhibits a sample and time complexity separation between two classes of quantum algorithms acting on input... | {
"rating": "3;6;6",
"rating_avg": 5,
"confidence": "3;5;4",
"confidence_avg": 4,
"soundness": "1;4;3",
"soundness_avg": 2.6666666666666665,
"contribution": "1;4;3",
"contribution_avg": 2.6666666666666665,
"presentation": "2;3;3",
"presentation_avg": 2.6666666666666665
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.597854"
} | {
"id": "OPmYCG5qCm",
"metareview": "This paper studied limitations for quantum machine learning algorithms that use fixed measurement schemes on the input quantum states. This is motivated by both recent advances in randomized measurement protocols as well as machine learning for quantum states. In particular, the... | {
"decision": "Reject"
} |
0uFTqvQhML | 2405.14475v2 | MagicDrive3D: Controllable 3D Generation for Any-View Rendering in Street Scenes | {
"content": "## Abstract\n\nAbstract While controllable generative models for images and videos have achieved remarkable success, high-quality models for 3D scenes, particularly in unbounded scenarios like autonomous driving, remain underdeveloped due to high data acquisition costs. In this paper, we introduce Magic... | [
{
"id": "ti2bl7A429",
"initial_rating": 6,
"confidence": 5,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper proposes a novel approach for 3D street scene generation, with a strong emphasis on multi-condition controllability, including BEV (Bird’s Eye View) ma... | {
"rating": "3;5;6;6",
"rating_avg": 5,
"confidence": "5;4;5;5",
"confidence_avg": 4.75,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "2;3;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.598519"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0vMLqSdsKW | 2409.13210v1 | A Unified Causal Framework for Auditing Recommender Systems for Ethical Concerns | {
"content": "## Abstract\n\nAbstract As recommender systems become widely deployed in different domains,\nthey increasingly influence their users’ beliefs and preferences.\nAuditing recommender systems is crucial as it not only ensures the continuous improvement of recommendation algorithms but also safeguards again... | [
{
"id": "ksLcEU89WT",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The paper presents a unified causal framework for auditing recommender systems with focus on user agency. The authors make three main contributions:\n1. A general... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "3;3;5;3",
"confidence_avg": 3.5,
"soundness": "2;3;2;3",
"soundness_avg": 2.5,
"contribution": "2;3;2;2",
"contribution_avg": 2.25,
"presentation": "3;2;2;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.599957"
} | {
"id": "gqhybUXrra",
"metareview": "This paper presents a causal framework for auditing recommender systems with a focus on ethical considerations such as user agency, stability, and reachability. It introduces two novel classes of metrics: future- and past-reachability and stability, which measure a user’s influe... | {
"decision": "Reject"
} |
0wQCSXJbwt | 2410.07812v1 | Temporal-Difference Variational Continual Learning | {
"content": "## Abstract\n\nAbstract A crucial capability of Machine Learning models in real-world applications is the ability to continuously learn new tasks. This adaptability allows them to respond to potentially inevitable shifts in the data-generating distribution over time. However, in Continual Learning (CL) ... | [
{
"id": "KTGXBLWZ7X",
"initial_rating": 3,
"confidence": 5,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "In this paper, the authors propsoed a new version of variational continual learning (VCL) which combines n-step regularization loss with temporal difference. The ... | {
"rating": "3;3;5;6",
"rating_avg": 4.25,
"confidence": "3;5;4;5",
"confidence_avg": 4.25,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "1;2;3;2",
"contribution_avg": 2,
"presentation": "3;2;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.601162"
} | {
"id": "PQKw4c912g",
"metareview": "This paper draws links between new Variational Continual Learning methods and Temporal-Difference mthods. \n\nThis meta-review is relatively short as all reviewers agreed that the paper requires more experiments to verify the claims made in the paper (MNIST scale experiments are... | {
"decision": "Reject"
} |
0wmfzWPAFu | 2409.14989v1 | Methods for Convex $(L_0,L_1)$-Smooth Optimization: Clipping, Acceleration, and Adaptivity | {
"content": "## Abstract\n\nAbstract Due to the non-smoothness of optimization problems in Machine Learning, generalized smoothness assumptions have gained much attention in recent years. One of the most popular assumptions of this type is ( L 0 , L 1 ) subscript 𝐿 0 subscript 𝐿 1 (L_{0},L_{1}) ( italic_L start_PO... | [
{
"id": "e1i8sNQmeP",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 1,
"presentation": 3,
"summary": "This paper analyzes the iteration complexities of several algorithms targeted at convex $(L_0,L_1)$-smooth optimization. In comparison to previous works, the auth... | {
"rating": "5;5;5;6;8",
"rating_avg": 5.8,
"confidence": "3;4;3;5;4",
"confidence_avg": 3.8,
"soundness": "2;3;3;3;4",
"soundness_avg": 3,
"contribution": "2;2;1;3;4",
"contribution_avg": 2.4,
"presentation": "3;3;3;4;4",
"presentation_avg": 3.4
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.602390"
} | {
"id": "giC7vCndpw",
"metareview": "This paper studies the optimization algorithm for strongly convex and (L0,L1)-smooth functions. They derived the improved convergence rates for multiple algorithm including Gradient Descent with Gradient Clipping, Gradient Descent with Polyak stepsizes, and Adaptive Gradient Des... | {
"decision": "Accept (Poster)"
} |
0xUEBQV54B | 2407.21787v2 | Large Language Monkeys: Scaling Inference Compute with Repeated Sampling | {
"content": "## Abstract\n\nAbstract Scaling the amount of compute used to train language models has dramatically improved their capabilities. However, when it comes to inference, we often limit the amount of compute to only one attempt per problem. Here, we explore inference compute as another axis for scaling by i... | [
{
"id": "1WU2aHd5br",
"initial_rating": 6,
"confidence": 5,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The paper studies scaling laws for a new axis of compute for LLMs: inference time compute. They empirically find that coverage (pass@N) improves with the number o... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "4;4;4;4",
"confidence_avg": 4,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "1;2;2;1",
"contribution_avg": 1.5,
"presentation": "3;4;2;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.603576"
} | {
"id": "20yiXJG8lv",
"metareview": "This paper studies scaling laws for LLMs as a function of the inference time budget. It makes empirical observations that coverage improves with increasing inference time budget and can be modeled with an exponentiated power law. For domains with automatic verifiers (e.g., code ... | {
"decision": "Reject"
} |
0y3hGn1wOk | 2411.03554v1 | Benchmarking Vision Language Model Unlearning via Fictitious Facial Identity Dataset | {
"content": "## Abstract\n\nAbstract Machine unlearning has emerged as an effective strategy for forgetting specific information in the training data. However, with the increasing integration of visual data, privacy concerns in Vision Language Models (VLMs) remain underexplored.\nTo address this, we introduce Facial... | [
{
"id": "RCEHSuNHoN",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper introduces an unlearning benchmark for Vision Language Models (VLMs) under the Right to be Forgotten setting. After defining the VLM unlearning tasks, ... | {
"rating": "5;5;5;6;6",
"rating_avg": 5.4,
"confidence": "4;5;4;5;3",
"confidence_avg": 4.2,
"soundness": "2;3;2;3;3",
"soundness_avg": 2.6,
"contribution": "2;3;2;3;3",
"contribution_avg": 2.6,
"presentation": "3;3;3;4;3",
"presentation_avg": 3.2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.604340"
} | {
"id": "lFbg13oPBc",
"metareview": "The paper introduces FIUBENCH, a benchmark for evaluating unlearning algorithms in Vision-Language Models (VLMs) under the \"Right to be Forgotten\" setting, using synthetic facial data and a two-stage evaluation pipeline. There are also some weaknesses of this paper, including ... | {
"decision": "Accept (Poster)"
} |
0yTf37PXcH | 2410.13733v1 | Improving Multi-modal Large Language Model through Boosting Vision Capabilities | {
"content": "## Abstract\n\nAbstract We focus on improving the visual understanding capability for boosting the vision-language models. We propose Arcana , a multiModal language model, which introduces two crucial techniques. First, we present Multimodal LoRA (MM-LoRA), a module designed to enhance the decoder. Unli... | [
{
"id": "jQJXMInTmP",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper proposes Arcana, a multi-modal large language model (MLLM) designed to improve visual perception capabilities. Arcana introduces two key techniques: MM-... | {
"rating": "3;5;5;5;8",
"rating_avg": 5.2,
"confidence": "4;5;4;4;5",
"confidence_avg": 4.4,
"soundness": "2;3;2;3;3",
"soundness_avg": 2.6,
"contribution": "2;3;1;2;3",
"contribution_avg": 2.2,
"presentation": "2;4;3;3;4",
"presentation_avg": 3.2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.605043"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
0ydseYDKRi | 2411.03820v1 | Beyond The Rainbow: High Performance Deep Reinforcement Learning On A Desktop PC | {
"content": "## Abstract\n\nAbstract Rainbow Deep Q-Network (DQN) demonstrated combining multiple independent enhancements could significantly boost a reinforcement learning (RL) agent’s performance. In this paper, we present “Beyond The Rainbow” (BTR), a novel algorithm that integrates six improvements from across ... | [
{
"id": "XGHb5FOKvh",
"initial_rating": 8,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper introduces Beyond The Rainbow (BTR), a novel reinforcement learning (RL) algorithm that enhances Rainbow DQN by integrating six key improvements. The B... | {
"rating": "3;5;6;8",
"rating_avg": 5.5,
"confidence": "4;4;4;3",
"confidence_avg": 3.75,
"soundness": "1;2;3;3",
"soundness_avg": 2.25,
"contribution": "3;3;2;3",
"contribution_avg": 2.75,
"presentation": "2;2;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.605732"
} | {
"id": "MG5Lc5YTt4",
"metareview": "This paper proposes Beyond The Rainbow (BTR) an RL algorithm combining multiple recent advancements into a single method with the single aim of outperforming on the Atari benchmark. There is a fairly high bar for this type of work to be accepted, given that the Atari environment... | {
"decision": "Reject"
} |
0yvZm2AjUr | 2406.19501v1 | Monitoring Latent World States in Language Models with Propositional Probes | {
"content": "## Abstract\n\nAbstract Language models are susceptible to bias, sycophancy, backdoors, and other tendencies that lead to unfaithful responses to the input context. Interpreting internal states of language models could help monitor and correct unfaithful behavior. We hypothesize that language models rep... | [
{
"id": "AOfKv0aoEX",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper studies how LLMs might encode propositions stated in the context, like \"Greg is a nurse. Laura is a physicist.\", by looking at the activations associa... | {
"rating": "6;6;6;8",
"rating_avg": 6.5,
"confidence": "3;4;3;4",
"confidence_avg": 3.5,
"soundness": "3;2;3;4",
"soundness_avg": 3,
"contribution": "3;3;3;4",
"contribution_avg": 3.25,
"presentation": "3;3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.606445"
} | {
"id": "39DxiuvLk0",
"metareview": "This paper proposes a method to interpret language models by extracting logical propositions that they claim represent the internal state of the model, by discovering bindings in a low-dimensional linear subspace of the model. The subspace discovery algorithm and the proposition... | {
"decision": "Accept (Spotlight)"
} |
12B3jBTL0V | 2410.14031v1 | Modeling the Human Visual System: Comparative Insights from Response-Optimized and Task-Optimized Vision Models, Language Models, and different Readout Mechanisms | {
"content": "## Abstract\n\nAbstract Over the past decade, predictive modeling of neural responses in the primate visual system has advanced significantly, largely driven by various deep neural network approaches. These include models optimized directly for visual recognition, cross-modal alignment through contrasti... | [
{
"id": "UaUFY3CsAQ",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 4,
"presentation": 2,
"summary": "In this work, the authors present a comprehensive suite of analyses comparing vision / language DNN models to human fMRI data. Using a novel “readout” mechanism d... | {
"rating": "3;3;5;6",
"rating_avg": 4.25,
"confidence": "3;4;3;4",
"confidence_avg": 3.5,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;3;3;4",
"contribution_avg": 3,
"presentation": "2;1;3;2",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.607873"
} | {
"id": "BI1cjmL7AZ",
"metareview": "The reviewers collectively noted that the paper suffers from subpar presentation quality, as highlighted by reviewer gWb8, with cluttered and confusing figures and unclear descriptions. Furthermore, while the proposed Semantic Spatial Transformer readout shows improvements in pr... | {
"decision": "Reject"
} |
13PclvlVBa | 2407.20254v2 | EEGMamba: Bidirectional State Space Model with Mixture of Experts for EEG Multi-task Classification | {
"content": "## Abstract\n\nAbstract In recent years, with the development of deep learning, electroencephalogram (EEG) classification networks have achieved certain progress. Transformer-based models can perform well in capturing long-term dependencies in EEG signals. However, their quadratic computational complexi... | [
{
"id": "O8seF0IL8q",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper presents EEGMamba, a model tailored for multi-task EEG classification. It aims to overcome the limitations of existing models in terms of computational ... | {
"rating": "3;3;5;6;6",
"rating_avg": 4.6,
"confidence": "4;4;4;4;4",
"confidence_avg": 4,
"soundness": "2;2;3;3;3",
"soundness_avg": 2.6,
"contribution": "2;2;2;3;3",
"contribution_avg": 2.4,
"presentation": "3;3;3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.608599"
} | {
"id": "e5b0Twjq6t",
"metareview": "The work considers the problem of multitask decoding from EEG (to understand a model that can predict for different tasks without retraining). To so so it introduces a new architecture by using a bidirectional mamba state-space model and a MoE approach that can orient the model ... | {
"decision": "Reject"
} |
14fFV0chUS | 2410.05643v2 | TRACE: Temporal Grounding Video LLM via Causal Event Modeling | {
"content": "## Abstract\n\nAbstract Video Temporal Grounding (VTG) is a crucial capability for video understanding models and plays a vital role in downstream tasks such as video browsing and editing.\nTo effectively handle various tasks simultaneously and enable zero-shot prediction, there is a growing trend in em... | [
{
"id": "Sh1itr2J38",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper proposes a new method for Video Temporal Grounding (VTG) tasks, named TRACE. TRACE uses a causal event modeling framework to represent videos as a sequ... | {
"rating": "5;5;6;6",
"rating_avg": 5.5,
"confidence": "4;3;4;3",
"confidence_avg": 3.5,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "3;2;3;2",
"contribution_avg": 2.5,
"presentation": "3;2;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.609415"
} | {
"id": "4BH4HsKNEj",
"metareview": "The paper receives 3 positive and 1 negative ratings after rebuttal, with 3 upgraded scores. Initially, the reviewers had several concerns about some technical clarity, motivations of using timestamps and scores, more contexts with the relevant work, more analysis on model param... | {
"decision": "Accept (Poster)"
} |
15ASUbzg0N | 2410.12822v1 | AVID: Adapting Video Diffusion Models to World Models | {
"content": "## Abstract\n\nAbstract Large-scale generative models have achieved remarkable success in a number of domains. However, for sequential decision-making problems, such as robotics, action-labelled data is often scarce and therefore scaling-up foundation models for decision-making remains a challenge.\nA p... | [
{
"id": "TAJxZxuoxW",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "The paper proposes a mechanism for adapting current image-conditioned video diffusion models to action-conditioned video diffusion models. They do this by trainin... | {
"rating": "5;5;6;6",
"rating_avg": 5.5,
"confidence": "3;4;3;3",
"confidence_avg": 3.25,
"soundness": "3;2;3;2",
"soundness_avg": 2.5,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "2;3;4;2",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.610203"
} | {
"id": "XKrpvPC1gX",
"metareview": "This is a well-written paper about training an action-conditioning adapter for a pre-trained, frozen video diffusion model. \n\nThe reviewer reception of this paper was mixed: while the writing and presentation are clear and of high quality, there were concerns about the general... | {
"decision": "Reject"
} |
15UetYngA7 | 2408.07990v1 | FuseChat: Knowledge Fusion of Chat Models | {
"content": "## Abstract\n\nAbstract While training large language models (LLMs) from scratch can indeed lead to models with distinct capabilities and strengths, it incurs substantial costs and may lead to redundancy in competencies. Knowledge fusion aims to integrate existing LLMs of diverse architectures and capab... | [
{
"id": "xu2L3ZL1oY",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 4,
"summary": "This paper proposes a new framework, FuseChat, to fuse diverse LLMs into a single LLM capable of performing various tasks. They first apply pairwise knowledge fu... | {
"rating": "5;6;6",
"rating_avg": 5.666666666666667,
"confidence": "4;5;4",
"confidence_avg": 4.333333333333333,
"soundness": "2;3;3",
"soundness_avg": 2.6666666666666665,
"contribution": "2;3;3",
"contribution_avg": 2.6666666666666665,
"presentation": "3;2;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.611051"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
15dVqf7VXR | 2405.17079v1 | Learning with User-Level Local Differential Privacy | {
"content": "## Abstract\n\nAbstract User-level privacy is important in distributed systems. Previous research primarily focuses on the central model, while the local models have received much less attention. Under the central model, user-level DP is strictly stronger than the item-level one. However, under the loca... | [
{
"id": "LnjnfB4rdS",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper first analyzes the mean estimation problem and then extends the findings to stochastic optimization, classification, and regression. Specifically, the ... | {
"rating": "3;5;6",
"rating_avg": 4.666666666666667,
"confidence": "4;3;4",
"confidence_avg": 3.6666666666666665,
"soundness": "2;2;3",
"soundness_avg": 2.3333333333333335,
"contribution": "2;2;3",
"contribution_avg": 2.3333333333333335,
"presentation": "3;2;3",
"presentation_avg": 2.66666666666666... | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.611974"
} | {
"id": "c4r3ra4kzb",
"metareview": "## Summary of Contributions\n\nThis paper studies learning tasks in user-level local DP (LDP) setting. Here each of the $n$ users has $m$ items drawn from an unknown distribution, and we want to satisfy LDP where the privacy unit is each user (i.e. all $m$ items can change). The... | {
"decision": "Reject"
} |
16O8GCm8Wn | 2410.18775v1 | Robust Watermarking Using Generative Priors Against Image Editing: From Benchmarking to Advances | {
"content": "## Abstract\n\nAbstract Current image watermarking methods are vulnerable to advanced image editing techniques enabled by large-scale text-to-image models. These models can distort embedded watermarks during editing, posing significant challenges to copyright protection. In this work, we introduce W-Ben... | [
{
"id": "1PbjBoz1aF",
"initial_rating": 6,
"confidence": 2,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "This paper presents VINE, a watermarking method designed to withstand various image editing techniques enabled by advanced generative models. It also introduces W... | {
"rating": "5;6;6;6;6",
"rating_avg": 5.8,
"confidence": "4;3;4;4;2",
"confidence_avg": 3.4,
"soundness": "2;3;2;3;2",
"soundness_avg": 2.4,
"contribution": "2;3;3;3;2",
"contribution_avg": 2.6,
"presentation": "3;3;3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.612935"
} | {
"id": "lGMWnuxesU",
"metareview": "This paper is working on robust watermarking. Authors first proposed W-Bench to evaluate the robustness of watermarking methods against a wide range of image editing techniques. Then proposed VINE to improve watermarking robustness against all these different edits. Experimental... | {
"decision": "Accept (Poster)"
} |
1762Fbr4HK | 2410.02079v1 | Deep Generative Modeling for Identification of Noisy, Non-Stationary Dynamical Systems | {
"content": "## Abstract\n\nAbstract A significant challenge in many fields of science and engineering is making sense of time-dependent measurement data by recovering governing equations in the form of differential equations. We focus on finding parsimonious ordinary differential equation (ODE) models for nonlinear... | [
{
"id": "QoaFhcgDRo",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "The paper proposes ‘dynamic sparse identification of nonlinear dynamics’ (dynamic SINDy), a deep learning framework for identifying governing equations in noisy, ... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "4;4;4;4",
"confidence_avg": 4,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "3;2;2;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.613766"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
19QWQSsbOA | 2407.05259v1 | Multi-scale Conditional Generative Modeling for Microscopic Image Restoration | {
"content": "## Abstract\n\nAbstract The advance of diffusion-based generative models in recent years has revolutionized state-of-the-art (SOTA) techniques in a wide variety of image analysis and synthesis tasks, whereas their adaptation on image restoration, particularly within computational microscopy remains theo... | [
{
"id": "jkTsZCtPo1",
"initial_rating": 6,
"confidence": 5,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The paper introduces a multi-scale conditional generative model (MSCGM) aimed at enhancing microscopic image restoration by combining wavelet transforms and a Bro... | {
"rating": "3;5;6;6",
"rating_avg": 5,
"confidence": "5;4;4;5",
"confidence_avg": 4.5,
"soundness": "2;1;3;3",
"soundness_avg": 2.25,
"contribution": "2;2;3;2",
"contribution_avg": 2.25,
"presentation": "1;2;3;3",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.614741"
} | {
"id": "AZA25KuW6C",
"metareview": "This work introduces a multi-scale generative model to enhance conditional image restoration by initiating the Brownian Bridge diffusion process specifically at the lowest-frequency subband and applying generative adversarial networks at subsequent multi-scale high-frequency sub... | {
"decision": "Reject"
} |
1AYrzmDK4V | 2407.14206v1 | Watermark Smoothing Attacks against Language Models | {
"content": "## Abstract\n\nAbstract Watermarking is a technique used to embed a hidden signal in the probability distribution of text generated by large language models (LLMs), enabling attribution of the text to the originating model. We introduce smoothing attacks and show that existing watermarking methods are n... | [
{
"id": "tAsxkAOkaF",
"initial_rating": 1,
"confidence": 4,
"soundness": 3,
"contribution": 1,
"presentation": 3,
"summary": "The paper proposes an automatic method for editing watermarked text from a language model to evade watermark detection using another (weaker) language model. The ... | {
"rating": "1;3;3;6",
"rating_avg": 3.25,
"confidence": "4;4;3;3",
"confidence_avg": 3.5,
"soundness": "3;2;2;2",
"soundness_avg": 2.25,
"contribution": "1;2;2;3",
"contribution_avg": 2,
"presentation": "3;2;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.615559"
} | {
"id": "nViDNQi9nB",
"metareview": "**Paper Summary:**\n\nThe paper proposes an attack on distortionary text watermarks, using text from an unwatermarked weak LM to rewrite text from a strong watermarked LM in a way that erases the watermark.\n\n**Strengths:**\n\nThe attack is interesting and it is simple: a token... | {
"decision": "Reject"
} |
1BdPHbuimc | 2403.17359v1 | Chain-of-Action: Faithful and Multimodal Question Answering through Large Language Models | {
"content": "## Abstract\n\nAbstract Large Language Models (LLMs) demonstrate their versatility in a variety of tasks. However, there are still two inevitable bottlenecks during practical applications: (1) unfaithful hallucination that is not consistent with real-time or specific domains’ facts; (2) weak reasoning o... | [
{
"id": "6HbAS1PEI9",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 1,
"summary": "The paper introduces the Chain-of-Action (CoA) framework, a novel approach to multimodal and retrieval-augmented question answering that enhances the faithfulness... | {
"rating": "5;5;8",
"rating_avg": 6,
"confidence": "3;4;3",
"confidence_avg": 3.3333333333333335,
"soundness": "3;2;3",
"soundness_avg": 2.6666666666666665,
"contribution": "3;2;3",
"contribution_avg": 2.6666666666666665,
"presentation": "4;1;3",
"presentation_avg": 2.6666666666666665
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.616057"
} | {
"id": "iLkQc4ygKa",
"metareview": "This paper introduces the Chain-of-Action (CoA) framework, a new approach to improving multimodal and retrieval-augmented QA for LLMs. The framework tackles two core challenges in QA—unfaithful reasoning and information hallucination—by decomposing complex questions into sequent... | {
"decision": "Accept (Poster)"
} |
1CLzLXSFNn | 2410.16032v1 | TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis | {
"content": "## Abstract\n\nAbstract Time series analysis plays a critical role in numerous applications, supporting tasks such as forecasting, classification, anomaly detection, and imputation. In this work, we present the time series pattern machine (TSPM), a model designed to excel in a broad range of time series... | [
{
"id": "PygUsAWjj8",
"initial_rating": 10,
"confidence": 4,
"soundness": 4,
"contribution": 4,
"presentation": 4,
"summary": "The paper presents TIMEMIXER++, a general-purpose model for various time series tasks, including forecasting, classification, anomaly detection, and imputation. ... | {
"rating": "5;6;10",
"rating_avg": 7,
"confidence": "4;5;4",
"confidence_avg": 4.333333333333333,
"soundness": "1;3;4",
"soundness_avg": 2.6666666666666665,
"contribution": "1;4;4",
"contribution_avg": 3,
"presentation": "3;3;4",
"presentation_avg": 3.3333333333333335
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Oral",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.616718"
} | {
"id": "X5qhV3Yy8C",
"metareview": "This paper presents TimeMixer++, a general-purpose model for various time series tasks, including forecasting, classification, anomaly detection, and imputation. The proposed model achieves state-of-the-art performance across 8 time series analytical tasks, consistently surpass... | {
"decision": "Accept (Oral)"
} |
1DEEVAl5QX | 2401.12478v2 | Mini-batch Submodular Maximization | {
"content": "## Abstract\n\nAbstract We present the first mini-batch algorithm for maximizing a non-negative monotone decomposable submodular function, F = ∑ i = 1 N f i 𝐹 superscript subscript 𝑖 1 𝑁 superscript 𝑓 𝑖 F=\\sum_{i=1}^{N}f^{i} italic_F = ∑ start_POSTSUBSCRIPT italic_i = 1 end_POSTSUBSCRIPT start_POS... | [
{
"id": "pOyNi0J9yi",
"initial_rating": 3,
"confidence": 4,
"soundness": 3,
"contribution": 1,
"presentation": 3,
"summary": "This paper considers maximization of decomposable monotone submodular functions over a ground set of size $n$, meaning that the objective function $f$ is a sum of... | {
"rating": "3;5;6",
"rating_avg": 4.666666666666667,
"confidence": "4;4;4",
"confidence_avg": 4,
"soundness": "3;3;3",
"soundness_avg": 3,
"contribution": "1;2;2",
"contribution_avg": 1.6666666666666667,
"presentation": "3;3;4",
"presentation_avg": 3.3333333333333335
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.617517"
} | {
"id": "PHa4AnjKX8",
"metareview": "The paper studies the problem of maximizing a decomposable submodular function subject to constraints. The main contribution of the paper is a sampling-based algorithm that uniformly samples and reweighs a subset of the functions in order to create a much smaller sparsified inst... | {
"decision": "Reject"
} |
1DIdt2YOPw | 2404.10960v1 | Uncertainty-Based Abstention in LLMs Improves Safety and Reduces Hallucinations | {
"content": "## Abstract\n\nAbstract A major barrier towards the practical deployment of large language models (LLMs) is their lack of reliability. Three situations where this is particularly apparent are correctness, hallucinations when given unanswerable questions, and safety. In all three cases, models should ide... | [
{
"id": "AZ6MQtNhUb",
"initial_rating": 3,
"confidence": 5,
"soundness": 4,
"contribution": 2,
"presentation": 4,
"summary": "The paper studies how token-level and semantic uncertainty metrics on generated LLM text relate to accuracy on knowledge-intensive tasks, hallucination on unanswe... | {
"rating": "3;3;3;5;5",
"rating_avg": 3.8,
"confidence": "5;5;5;4;4",
"confidence_avg": 4.6,
"soundness": "2;3;4;3;2",
"soundness_avg": 2.8,
"contribution": "2;1;2;3;2",
"contribution_avg": 2,
"presentation": "3;3;4;3;2",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.618348"
} | {
"id": "6iGzXb3TF5",
"metareview": "PC is entering meta-review on behalf of SAC/AC:\n\nThe reviewers did not believe that this paper was a strong contribution given the limited novelty of the work, and lack of anchoring in the field.",
"additional_comments": "TBD"
} | {
"decision": "Reject"
} |
1EnpStvBU8 | 2403.03003v1 | Feast Your Eyes: Mixture-of-Resolution Adaptation for Multimodal Large Language Models | {
"content": "## Abstract\n\nAbstract Despite remarkable progress, existing multimodal large language models (MLLMs) are still inferior in granular visual recognition. Contrary to previous works, we study this problem from the perspective of image resolution, and reveal that a combination of low- and high-resolution ... | [
{
"id": "FxDVard2bM",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper aims to enhance MLLM by enlarging resolution of input images. By combining features from ViT and a CNN encoder through an adapter, performances of MLLM... | {
"rating": "5;5;6;6;6",
"rating_avg": 5.6,
"confidence": "4;4;4;4;4",
"confidence_avg": 4,
"soundness": "3;3;3;3;3",
"soundness_avg": 3,
"contribution": "2;2;3;3;3",
"contribution_avg": 2.6,
"presentation": "3;3;3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.619082"
} | {
"id": "NDBJAjMbqh",
"metareview": "This paper proposes a mixture-of-resolution adaption method for multimodal large language model (MLLM). It consists of two visual pathways for images of different resolutions, ViT/ConveNext for low/high-resolution. The information from high-res input is adapted to low-res featur... | {
"decision": "Accept (Poster)"
} |
1Ffzgglq2I | 2406.10445v3 | Binary Reward Labeling: Bridging Offline Preference and Reward-Based Reinforcement Learning | {
"content": "## Abstract\n\nAbstract Offline reinforcement learning has become one of the most practical RL settings. However, most existing works on offline RL focus on the standard setting with scalar reward feedback. It remains unknown how to universally transfer the existing rich understanding of offline RL from... | [
{
"id": "vFQY8IFNmc",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The paper presents a novel framework aimed at bridging the gap between offline preference-based reinforcement learning (PBRL) and standard offline reward-based re... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "3;2;3;4",
"confidence_avg": 3,
"soundness": "3;2;2;2",
"soundness_avg": 2.25,
"contribution": "3;2;2;2",
"contribution_avg": 2.25,
"presentation": "2;1;2;2",
"presentation_avg": 1.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.619703"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
1GTARJhxtq | 2405.20541v1 | Perplexed by Perplexity: Perplexity-Based Data Pruning With Small Reference Models | {
"content": "## Abstract\n\nAbstract In this work, we investigate whether small language models can determine high-quality subsets of large-scale text datasets that improve the performance of larger language models.\nWhile existing work has shown that pruning based on the perplexity of a larger model can yield high-... | [
{
"id": "ZywBbWh82H",
"initial_rating": 8,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper proposes that smaller language models effectively prune large datasets in a way that benefits the training of much larger model. Applying perplexity-bas... | {
"rating": "3;5;6;8",
"rating_avg": 5.5,
"confidence": "3;5;4;3",
"confidence_avg": 3.75,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;3;3;3",
"contribution_avg": 2.75,
"presentation": "3;3;4;3",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.620447"
} | {
"id": "AYGnFt1rRW",
"metareview": "This paper presents a study on using smaller language models to prune large datasets effectively, aiming to improve the training efficiency of larger models. The proposed perplexity-based pruning technique is evaluated on two distinct datasets, The Pile and Dolma, which vary in ... | {
"decision": "Accept (Poster)"
} |
1HCN4pjTb4 | 2410.04887v1 | Wide Neural Networks Trained with Weight Decay Provably Exhibit Neural Collapse | {
"content": "## Abstract\n\nAbstract Deep neural networks (DNNs) at convergence consistently represent the training data in the last layer via a highly symmetric geometric structure referred to as neural collapse. This empirical evidence has spurred a line of theoretical research aimed at proving the emergence of ne... | [
{
"id": "zPXleeD4ia",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "Dear authors, thank you for submitting your work to ICLR 2025. The paper considers the phenomenon of 'neural collapse' and attempt to extend current rather specia... | {
"rating": "5;6;6;8;8",
"rating_avg": 6.6,
"confidence": "4;3;3;4;2",
"confidence_avg": 3.2,
"soundness": "3;3;3;4;4",
"soundness_avg": 3.4,
"contribution": "3;3;2;4;4",
"contribution_avg": 3.2,
"presentation": "2;3;1;4;3",
"presentation_avg": 2.6
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Oral",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.621277"
} | {
"id": "HXO6sQqUzn",
"metareview": "This paper explores neural collapse, a geometric structure observed in the last layer of deep neural networks (DNNs) at convergence, where training data is consistently represented. Moving beyond the unconstrained features model, the authors study DNNs with at least two linear l... | {
"decision": "Accept (Oral)"
} |
1Iuw1jcIrf | 2410.08196v1 | MathCoder2: Better Math Reasoning from Continued Pretraining on Model-translated Mathematical Code | {
"content": "## Abstract\n\nAbstract Code has been shown to be effective in enhancing the mathematical reasoning abilities of large language models due to its precision and accuracy. Previous works involving continued mathematical pretraining often include code that utilizes math-related packages, which are primaril... | [
{
"id": "MCK0blSBL8",
"initial_rating": 8,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper presents a novel approach for enhancing mathematical reasoning in large language models (LLMs). Unlike previous models that used math-related code with... | {
"rating": "6;8;8",
"rating_avg": 7.333333333333333,
"confidence": "4;4;3",
"confidence_avg": 3.6666666666666665,
"soundness": "3;4;3",
"soundness_avg": 3.3333333333333335,
"contribution": "3;4;3",
"contribution_avg": 3.3333333333333335,
"presentation": "3;4;3",
"presentation_avg": 3.33333333333333... | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.622031"
} | {
"id": "rs921tBM1K",
"metareview": "The paper proposes a novel method for generating paired pretraining data that combines mathematical code with corresponding reasoning steps to improve LLMs' mathematical reasoning capabilities through continued pretraining. Using this approach, the authors introduce MathCode-Pil... | {
"decision": "Accept (Spotlight)"
} |
1IuwdOI4Zb | 2410.10306v1 | Animate-X: Universal Character Image Animation with Enhanced Motion Representation | {
"content": "## Abstract\n\nAbstract Character image animation, which generates high-quality videos from a reference image and target pose sequence, has seen significant progress in recent years. However, most existing methods only apply to human figures, which usually do not generalize well on anthropomorphic chara... | [
{
"id": "SClEUBMPLQ",
"initial_rating": 5,
"confidence": 5,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This work presents an animation framework capable of animating anthropomorphic characters, along with an accompanying benchmark for animated anthropomorphic chara... | {
"rating": "5;5;6;6",
"rating_avg": 5.5,
"confidence": "5;5;5;4",
"confidence_avg": 4.75,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "2;2;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.622681"
} | {
"id": "PHaTjppS8O",
"metareview": "The paper received very positive ratings from the reviewers. They highlight improved results, new ideas, the introduction of a new benchmarks, as well as the ability of the method to animate objects not having a distinct skeleton structure. The reviewers also pointed out that at... | {
"decision": "Accept (Poster)"
} |
1IwoEFyErz | 2410.21088v1 | Shallow Diffuse: Robust and Invisible Watermarking through Low-Dimensional Subspaces in Diffusion Models | {
"content": "## Abstract\n\nAbstract The widespread use of AI-generated content from diffusion models has raised significant concerns regarding misinformation and copyright infringement. Watermarking is a crucial technique for identifying these AI-generated images and preventing their misuse. In this paper, we intro... | [
{
"id": "v0j1VEnymZ",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "This paper proposed Shallow Diffuse, a watermarking technique for diffusion models. The method is well-motivated and with proper theoretical justification. The pr... | {
"rating": "5;5;5;5",
"rating_avg": 5,
"confidence": "5;4;5;4",
"confidence_avg": 4.5,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "2;2;2;3",
"contribution_avg": 2.25,
"presentation": "2;3;2;1",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.623469"
} | {
"id": "JIwq2WKsW3",
"metareview": "This paper studies diffusion-based digital watermarking to deal with the AI-generated content tracing problem. The proposed approach, ``Shallow Diffuse,'' tries to decouple both the watermarking and diffusion processes by leveraging the presence of a low-dimensional subspace in ... | {
"decision": "Reject"
} |
1JhSJIYX3p | 2410.17787v1 | Large Language Models Engineer Too Many Simple Features for Tabular Data | {
"content": "## Abstract\n\nAbstract Tabular machine learning problems often require time-consuming and labor-intensive feature engineering.\nRecent efforts have focused on using large language models (LLMs) to capitalize on their potential domain knowledge.\nAt the same time, researchers have observed ethically con... | [
{
"id": "VKAQFWrcSt",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The authors investigate how well LLMs can engineer features for tabular datasets. Specifically, they look at the frequencies of operators, and find that there is ... | {
"rating": "3;3;3;5",
"rating_avg": 3.5,
"confidence": "4;5;4;4",
"confidence_avg": 4.25,
"soundness": "2;3;2;4",
"soundness_avg": 2.75,
"contribution": "1;1;2;2",
"contribution_avg": 1.5,
"presentation": "2;3;2;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.624159"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
1KvYxcAihR | 2410.10479v1 | TMGBench: A Systematic Game Benchmark for Evaluating Strategic Reasoning Abilities of LLMs | {
"content": "## Abstract\n\nAbstract The rapid advancement of large language models (LLMs) has accelerated their application in reasoning, with strategic reasoning drawing increasing attention.\nTo evaluate the strategic reasoning capabilities of LLMs, game theory, with its concise structure, has become the preferre... | [
{
"id": "VjKE4yyOFF",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "The paper introduces TMGBENCH, a benchmark for systematically evaluating the strategic reasoning abilities of LLMs. By evaluating some LLMs on TMGBENCH, the paper... | {
"rating": "5;5;5;8",
"rating_avg": 5.75,
"confidence": "4;4;3;2",
"confidence_avg": 3.25,
"soundness": "3;3;2;4",
"soundness_avg": 3,
"contribution": "2;3;2;4",
"contribution_avg": 2.75,
"presentation": "2;1;3;4",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:00.624756"
} | {
"id": "EDXmhYTVrY",
"metareview": "The paper introduces an LLM benchmark based on game theory games (prisoners dilemma and its ilk, not games that people actually play). Reviewers are lukewarm about the paper, partly because of a perceived dearth of novelty and justification, which I consider not to be reason eno... | {
"decision": "Reject"
} |
1L52bHEL5d | 2404.15161v1 | Test-Time Adaptation for Combating Missing Modalities in Egocentric Videos | {
"content": "## Abstract\n\nAbstract Understanding videos that contain multiple modalities is crucial, especially in egocentric videos, where combining various sensory inputs significantly improves tasks like action recognition and moment localization.\nHowever, real-world applications often face challenges with inc... | [
{
"id": "rnHnCn6uH3",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "To tackle the issue of modality missing in real-time tasks, this framework offers an online self-supervised learning method called MiDl. MiDl uses mutual informat... | {
"rating": "5;5;6;6",
"rating_avg": 5.5,
"confidence": "4;3;4;4",
"confidence_avg": 3.75,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "2;3;3;3",
"contribution_avg": 2.75,
"presentation": "3;4;3;2",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.625377"
} | {
"id": "PeFHya5yKy",
"metareview": "The rebuttal provided clarifications about the proposed method and its analysis that were useful for assessing the paper's contribution and responded adequately to most reviewer concerns. All reviewers recommend acceptance after discussion (with four marginally above the accepta... | {
"decision": "Accept (Poster)"
} |
1MHgMGoqsH | 2409.19561v1 | Unifying Back-Propagation and Forward-Forward Algorithms through Model Predictive Control | {
"content": "## Abstract\n\nAbstract We introduce a Model Predictive Control (MPC) framework for training deep neural networks,\nsystematically unifying the Back-Propagation (BP)\nand Forward-Forward (FF) algorithms.\nAt the same time, it gives rise to a range of\nintermediate training algorithms with varying look-f... | [
{
"id": "o9ZilKrGlA",
"initial_rating": 3,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "Drawing inspiration from the model predictive control framework, this work proposes a framework for integrating back-propagation (BP) and the forward-forward (FF)... | {
"rating": "3;3;3",
"rating_avg": 3,
"confidence": "3;4;3",
"confidence_avg": 3.3333333333333335,
"soundness": "2;2;2",
"soundness_avg": 2,
"contribution": "2;1;2",
"contribution_avg": 1.6666666666666667,
"presentation": "2;2;2",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:00.626081"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
1Njl73JKjB | 2405.08366v3 | Towards Principled Evaluations of Sparse Autoencoders for Interpretability and Control | {
"content": "## Abstract\n\nAbstract Disentangling model activations into meaningful features is a central problem in\ninterpretability. However, the absence of ground-truth for these features in\nrealistic scenarios makes validating recent approaches, such as sparse\ndictionary learning, elusive. To address this ch... | [
{
"id": "3Fh1ANhB3y",
"initial_rating": 8,
"confidence": 2,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The authors propose a (principled) method allowing to create supervised dictionaries for space features, which allow for evaluating the degree of disentanglement ... | {
"rating": "3;5;6;8",
"rating_avg": 5.5,
"confidence": "4;3;2;2",
"confidence_avg": 2.75,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "3;2;3;3",
"contribution_avg": 2.75,
"presentation": "1;4;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:00.627293"
} | {
"id": "v5xmE7Qi9c",
"metareview": "This paper explores sparse autoencoders on language tasks that have a known ground truth to evaluate whether SAEs can provide similar interpretability and control as supervised feature dictionaries.\n\nSince most language tasks do not have a single ground truth, the authors orig... | {
"decision": "Accept (Poster)"
} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.