paper_id string | arxiv_id string | title string | markdown dict | reviews list | scores dict | metadata dict | meta_review dict | decision dict |
|---|---|---|---|---|---|---|---|---|
CWAvMSNUqT | 2409.02727v2 | Pooling And Attention: What Are Effective Designs For LLM-Based Embedding Models? | {
"content": "## Abstract\n\nAbstract The significant advancements of Large Language Models (LLMs) in generative tasks have led to a growing body of work exploring LLM-based embedding models. While these models, employing different pooling and attention strategies, have achieved state-of-the-art performance on public... | [
{
"id": "gHjyqt7qfa",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The paper conducts many experiments by training several LLM-based embedding models using the same training data and base model, but varying their pooling and atte... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "4;4;3;3",
"confidence_avg": 3.5,
"soundness": "2;1;2;3",
"soundness_avg": 2,
"contribution": "2;2;2;2",
"contribution_avg": 2,
"presentation": "3;3;4;3",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.235966"
} | {
"id": "mWnBK7bIQ3",
"metareview": "This submission presents a study of design choices in LLM-based embedding models, focusing on pooling and attention strategies. The paper conducts controlled experiments using identical base models and training data to isolate the impact of different architectural choices. The m... | {
"decision": "Reject"
} |
CaRkGrdewB | 2410.23883v1 | 'No' Matters: Out-of-Distribution Detection in Multimodality Long Dialogue | {
"content": "## Abstract\n\nAbstract Out-of-distribution (OOD) detection in multimodal contexts is essential for identifying deviations in combined inputs from different modalities, particularly in applications like open-domain dialogue systems or real-life dialogue interactions. This paper aims to improve the user ... | [
{
"id": "FQQ3FUbqom",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper addresses the challenge of out-of-distribution (OOD) detection in multimodal contexts, particularly focusing on the combined input of dialogues and ima... | {
"rating": "3;3;5;5;5",
"rating_avg": 4.2,
"confidence": "2;3;2;4;3",
"confidence_avg": 2.8,
"soundness": "2;2;2;2;2",
"soundness_avg": 2,
"contribution": "1;2;3;2;2",
"contribution_avg": 2,
"presentation": "3;3;3;3;2",
"presentation_avg": 2.8
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.236603"
} | {
"id": "Racv07skMR",
"metareview": "This paper proposes a framework for Out-of-Distribution (OOD) detection in multimodal, multi-turn dialogue scenarios, integrating image and dialogue inputs with a novel scoring method. While the approach is clearly explained and supported by experiments, the reviewers found issu... | {
"decision": "Reject"
} |
Cb4YXpqBIc | 2410.10663v1 | Cross-Modal Few-Shot Learning: a Generative Transfer Learning Framework | {
"content": "## Abstract\n\nAbstract Most existing studies on few-shot learning focus on unimodal settings, where models are trained to generalize on unseen data using only a small number of labeled examples from the same modality. However, real-world data are inherently multi-modal, and unimodal approaches limit th... | [
{
"id": "qvYLAdaUSD",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "In this paper, the authors introduced the Cross-modal Few-Shot Learning (CFSL) benchmark that aims to recognize instances from multiple modalities in a data effic... | {
"rating": "3;5;6;6",
"rating_avg": 5,
"confidence": "4;3;4;4",
"confidence_avg": 3.75,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "1;2;3;3",
"contribution_avg": 2.25,
"presentation": "2;2;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:01.237273"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
CbpWPbYHuv | 2411.03884v1 | Polynomial Composition Activations: Unleashing the Dynamics of Large Language Models | {
"content": "## Abstract\n\nAbstract Transformers have found extensive applications across various domains due to the powerful fitting capabilities. This success can be partially attributed to their inherent nonlinearity. Thus, in addition to the ReLU function employed in the original transformer architecture, resea... | [
{
"id": "ncGvGTGiOW",
"initial_rating": 6,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "This work presents a method to extend activation functions using polynomials.\nTwo examples of these extended activation functions are introduced: PolyReLU and Po... | {
"rating": "3;6;6;8",
"rating_avg": 5.75,
"confidence": "3;4;3;4",
"confidence_avg": 3.5,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "1;3;3;3",
"contribution_avg": 2.5,
"presentation": "3;4;4;3",
"presentation_avg": 3.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.238296"
} | {
"id": "E3dnzqoDJW",
"metareview": "**Summary**\n\nThis work introduces a novel activation function called PolyCom, which is a composition of polynomials and other types of functions featuring specific instances like PolyReLU and PolyNorm. These are seamlessly integrated into transformer architectures, offering se... | {
"decision": "Accept (Poster)"
} |
Ccwp4tFEtE | 2408.15240v2 | Generative Verifiers: Reward Modeling as Next-Token Prediction | {
"content": "## Abstract\n\nAbstract Verifiers or reward models are often used to enhance the reasoning performance of large language models (LLMs). A common approach is the Best-of-N method, where N candidate solutions generated by the LLM are ranked by a verifier, and the best one is selected. While LLM-based veri... | [
{
"id": "UaHBOmQZtt",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper proposes Generative Verifiers (GenRM), a novel framework for verification in large language models (LLMs), which reframes reward modeling as a generati... | {
"rating": "3;5;6",
"rating_avg": 4.666666666666667,
"confidence": "4;4;3",
"confidence_avg": 3.6666666666666665,
"soundness": "2;2;3",
"soundness_avg": 2.3333333333333335,
"contribution": "1;2;3",
"contribution_avg": 2,
"presentation": "2;2;3",
"presentation_avg": 2.3333333333333335
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.239032"
} | {
"id": "bVTsMSPgZJ",
"metareview": "This paper propose training verifiers using the ubiquitous next-token prediction objective, jointly on verification and solution generation. The method appears to be novel and experimental results outperforms DPO verifiers, and LLM-as-a-Judge. \n\nStrength: A novel approach to b... | {
"decision": "Accept (Poster)"
} |
CexatBp6rx | 2407.01331v1 | Restyling Unsupervised Concept Based Interpretable Networks with Generative Models | {
"content": "## Abstract\n\nAbstract Developing inherently interpretable models for prediction has gained prominence in recent years. A subclass of these models, wherein the interpretable network relies on learning high-level concepts, are valued because of closeness of concept representations to human communication... | [
{
"id": "aBa6AhRhcJ",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper proposes a new approach for inherently interpretable models by mapping the concept space to the latent space of a pre-trained generative model. In part... | {
"rating": "5;6;6;6;6",
"rating_avg": 5.8,
"confidence": "4;4;3;3;3",
"confidence_avg": 3.4,
"soundness": "3;3;3;3;3",
"soundness_avg": 3,
"contribution": "2;3;3;3;3",
"contribution_avg": 2.8,
"presentation": "3;3;3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.239827"
} | {
"id": "jFAJrdoqzi",
"metareview": "This paper proposes a concept translator for inherently interpretable models. This translator uses pre-trained generative models, and maps concepts into the learned latent space of this pre-trained generative model.\n\nImportant concerns, such as relation with CBM models, the r... | {
"decision": "Accept (Poster)"
} |
CfXRcN4iUw | 2410.08524v1 | IGNN-Solver: A Graph Neural Solver for Implicit Graph Neural Networks | {
"content": "## Abstract\n\nAbstract Implicit graph neural networks (IGNNs), which exhibit strong expressive power with a single layer, have recently demonstrated remarkable performance in capturing long-range dependencies (LRD) in underlying graphs while effectively mitigating the over-smoothing problem. However, I... | [
{
"id": "wjqIsBepM7",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper presents IGNN-Solver, a novel approach to accelerate fixed-point solving in implicit graph neural networks (IGNNs), addressing the scalability challeng... | {
"rating": "3;3;5;5;6",
"rating_avg": 4.4,
"confidence": "4;4;3;3;4",
"confidence_avg": 3.6,
"soundness": "2;2;2;2;3",
"soundness_avg": 2.2,
"contribution": "2;2;2;2;2",
"contribution_avg": 2,
"presentation": "2;2;3;2;3",
"presentation_avg": 2.4
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:01.240712"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
CfZPzH7ftt | 2410.03783v2 | Improving Neural Optimal Transport via Displacement Interpolation | {
"content": "## Abstract\n\nAbstract Optimal Transport (OT) theory investigates the cost-minimizing transport map that moves a source distribution to a target distribution. Recently, several approaches have emerged for learning the optimal transport map for a given cost function using neural networks. We refer to th... | [
{
"id": "288nwkQQnc",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The authors propose a novel method (DIOTM) to solve the optimal transport mapping problem for the quadratic transport cost (Wasserstein-2 OT) with neural networks... | {
"rating": "5;6;6;8",
"rating_avg": 6.25,
"confidence": "3;4;4;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;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.241429"
} | {
"id": "13DcTMdZd5",
"metareview": "The reviewers agree that the paper presents a novel and interesting method for solving the optimal transport mapping problem using neural networks. They appreciate the idea of exploiting displacement interpolation and the theoretically justified HJB-based regularization. The e... | {
"decision": "Accept (Poster)"
} |
CfdPELywGN | 2406.15275v2 | How language models extrapolate outside the training data: A Case study in Textualized Gridworld | {
"content": "## Abstract\n\nAbstract Language models have demonstrated impressive capabilities across various natural language processing tasks, yet they struggle with planning tasks requiring multi-step simulations. Inspired by human cognitive processes, this paper investigates the optimal planning power of languag... | [
{
"id": "E0Mf1Ys1fe",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "The authors study LMs ability to solve path finding in synthetic GridWorld problems. They propose fine-tuning an LM to first produce path search traces (called c... | {
"rating": "3;3;5;5;8",
"rating_avg": 4.8,
"confidence": "3;4;2;4;4",
"confidence_avg": 3.4,
"soundness": "2;2;3;3;3",
"soundness_avg": 2.6,
"contribution": "2;2;2;2;3",
"contribution_avg": 2.2,
"presentation": "2;3;3;2;3",
"presentation_avg": 2.6
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.242159"
} | {
"id": "Q4NENM05n4",
"metareview": "This paper studies the generalization capabilities of data-driven models trained on a toy grid-world task. The find that a CoT approach (dubbed \"cognitive maps\") enables better generalization to unseen environments.\n\nOn the positive side, this paper studies an interesting (a... | {
"decision": "Reject"
} |
CiiLchbRe3 | 2405.14219v2 | Understanding the Training and Generalization of Pretrained Transformer for Sequential Decision Making | {
"content": "## Abstract\n\nAbstract In this paper, we consider the supervised pre-trained transformer for a class of sequential decision-making problems. The class of considered problems is a subset of the general formulation of reinforcement learning in that there is no transition probability matrix; though seemin... | [
{
"id": "LtvhB2lOSw",
"initial_rating": 6,
"confidence": 3,
"soundness": 4,
"contribution": 3,
"presentation": 3,
"summary": "This paper presents a supervised training framework for transformers applied to sequential decision-making problems—a subset of reinforcement learning (RL) tasks ... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "3;2;2;3",
"confidence_avg": 2.5,
"soundness": "2;3;3;4",
"soundness_avg": 3,
"contribution": "1;2;2;3",
"contribution_avg": 2,
"presentation": "1;3;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.243121"
} | {
"id": "cQR8hUpoxl",
"metareview": "This paper presents a supervised training framework for transformers applied to sequential decision-making problems—a subset of reinforcement learning (RL) tasks that lack an explicit transition matrix. In this framework, transformers are pre-trained on the target task class by ... | {
"decision": "Reject"
} |
CkCFoN3j4s | 2410.01805v1 | Locret: Enhancing Eviction in Long-Context LLM Inference with Trained Retaining Heads | {
"content": "## Abstract\n\nAbstract Large language models (LLMs) have shown remarkable advances in supporting long-context comprehension and processing tasks.\nHowever, scaling the generation inference of LLMs to such long contexts incurs significant additional computation load, and demands a substantial GPU memory... | [
{
"id": "AeMan22iwV",
"initial_rating": 8,
"confidence": 4,
"soundness": 2,
"contribution": 3,
"presentation": 3,
"summary": "The paper proposes LOCRET, an framework designed to enhance memory efficiency in long-context large language model (LLM) inference by using retaining heads to sco... | {
"rating": "3;5;5;6;8",
"rating_avg": 5.4,
"confidence": "5;5;4;4;4",
"confidence_avg": 4.4,
"soundness": "2;3;3;3;2",
"soundness_avg": 2.6,
"contribution": "3;3;2;2;3",
"contribution_avg": 2.6,
"presentation": "3;2;3;2;3",
"presentation_avg": 2.6
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.244779"
} | {
"id": "pFIhkMcuWY",
"metareview": "The paper introduces LOCRET, a framework for memory-efficient long-context inference in LLMs. LOCRET employs lightweight retaining heads trained to predict the causal importance of KV cache units, enabling selective cache eviction. Using this mechanism with `chunked prefill`, LO... | {
"decision": "Reject"
} |
CkKEuLmRnr | 2410.05298v1 | How Do Large Language Models Understand Graph Patterns? A Benchmark for Graph Pattern Comprehension | {
"content": "## Abstract\n\nAbstract Benchmarking the capabilities and limitations of large language models (LLMs) in graph-related tasks is becoming an increasingly popular and crucial area of research. Recent studies have shown that LLMs exhibit a preliminary ability to understand graph structures and node feature... | [
{
"id": "o45CaZkW7S",
"initial_rating": 6,
"confidence": 4,
"soundness": 4,
"contribution": 4,
"presentation": 2,
"summary": "This paper presents a benchmark that evaluates SOTA LLM graph pattern understanding and whether any graph reasoning is gleaned from pretraining in graph-based tas... | {
"rating": "3;5;6;6",
"rating_avg": 5,
"confidence": "4;4;5;4",
"confidence_avg": 4.25,
"soundness": "3;2;3;4",
"soundness_avg": 3,
"contribution": "1;3;3;4",
"contribution_avg": 2.75,
"presentation": "4;2;3;2",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.245515"
} | {
"id": "karKB9NeZj",
"metareview": "The paper examines the ability of LLMs to understand graph patterns from synthetic and real data, and their capacity to discover these patterns from data. The authors vary the description of these patterns based on terminology and topology to observe the impact on performance. \... | {
"decision": "Accept (Poster)"
} |
CkgKSqZbuC | 2407.05557v1 | $R^2$-Guard: Robust Reasoning Enabled LLM Guardrail via Knowledge-Enhanced Logical Reasoning | {
"content": "## Abstract\n\nAbstract As large language models (LLMs) become increasingly prevalent across various applications, it is critical to establish safety guardrails to moderate input/output of LLMs and ensure compliance with safety policies.\nExisting guardrail models, such as OpenAI Mod and LlamaGuard, tre... | [
{
"id": "YjUHJTyCrY",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "Existing LLM guardrails treat different categories of safety failures independently. In contrast, R2-Guard proposes a reasoning-enabled LLM guardrail that can per... | {
"rating": "6;6;8",
"rating_avg": 6.666666666666667,
"confidence": "3;4;3",
"confidence_avg": 3.3333333333333335,
"soundness": "3;3;4",
"soundness_avg": 3.3333333333333335,
"contribution": "3;3;4",
"contribution_avg": 3.3333333333333335,
"presentation": "4;3;2",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.246246"
} | {
"id": "DpxaJ3Ufgq",
"metareview": "This paper introduces a novel approach to language model safety that combines probabilistic graphical models with traditional category-specific guardrails. The system implements additional reasoning through a PGM grounded in first-order logical rules, enabling it to capture rela... | {
"decision": "Accept (Spotlight)"
} |
ClkfwM3STw | 2406.12928v1 | Evaluating the Generalization Ability of Quantized LLMs: Benchmark, Analysis, and Toolbox | {
"content": "## Abstract\n\nAbstract Large language models (LLMs) have exhibited exciting progress in multiple scenarios, while the huge computational demands hinder their deployments in lots of real-world applications. As an effective means to reduce memory footprint and inference cost, quantization also faces chal... | [
{
"id": "eLRE9czn3U",
"initial_rating": 5,
"confidence": 2,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "The authors proposed a benchmark for evaluating the post-training quantized large language models (LLMs) generalization ability. They considered two scenarios and... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "5;4;2;3",
"confidence_avg": 3.5,
"soundness": "3;3;2;2",
"soundness_avg": 2.5,
"contribution": "2;2;2;2",
"contribution_avg": 2,
"presentation": "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": "2026-01-14T22:16:01.247030"
} | {
"id": "SlebSTnYqj",
"metareview": "This paper evaluates the generalization ability of quantized LLMs through a benchmark suite, offering an evaluation system, detailed analyses, and a modular toolbox. The study examines the impact of calibration data distribution on quantized LLMs using over 40 datasets and popul... | {
"decision": "Reject"
} |
Cn5Z0MUPZT | 2410.17621v1 | Process Supervision-Guided Policy Optimization for Code Generation | {
"content": "## Abstract\n\nAbstract Reinforcement learning (RL) with unit test feedback has enhanced large language models’ (LLMs) code generation, but relies on sparse rewards provided only after complete code evaluation, limiting learning efficiency and incremental improvements. When generated code fails all unit... | [
{
"id": "z6kK7fV0p1",
"initial_rating": 8,
"confidence": 3,
"soundness": 4,
"contribution": 3,
"presentation": 3,
"summary": "The paper introduces a practical approach for training models with process-oriented feedback for code generation. The approach uses a novel automatic (LLM + unit ... | {
"rating": "1;3;5;8",
"rating_avg": 4.25,
"confidence": "4;5;4;3",
"confidence_avg": 4,
"soundness": "2;2;3;4",
"soundness_avg": 2.75,
"contribution": "1;2;2;3",
"contribution_avg": 2,
"presentation": "1;3;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.247863"
} | {
"id": "bCvEUZyHaY",
"metareview": "> The paper at hand concerns itself with RL fine-tuning of LLMs from unit test feedback, i.e., obtained in code generation tasks by testing the LLM output against a set of unit tests. The main feature here is the use of a process reward model (PRM) to supply dense rewards at mul... | {
"decision": "Reject"
} |
Cnwz9jONi5 | 2410.05584v2 | Rethinking Reward Model Evaluation: Are We Barking up the Wrong Tree? | {
"content": "## Abstract\n\nAbstract Reward Models (RMs) are crucial for aligning language models with human preferences.\nCurrently, the evaluation of RMs depends on measuring accuracy against a validation set of manually annotated preference data.\nAlthough this method is straightforward and widely adopted, the re... | [
{
"id": "330gmFVsG5",
"initial_rating": 8,
"confidence": 3,
"soundness": 4,
"contribution": 3,
"presentation": 3,
"summary": "Research question of the paper is on how we should evaluate the quality of the reward models for RLHF. The paper conducts experiments to evaluate the evaluation m... | {
"rating": "5;5;6;8",
"rating_avg": 6,
"confidence": "3;4;4;3",
"confidence_avg": 3.5,
"soundness": "2;2;3;4",
"soundness_avg": 2.75,
"contribution": "2;3;4;3",
"contribution_avg": 3,
"presentation": "2;2;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.248558"
} | {
"id": "NZ3Lh4AMRt",
"metareview": "This paper questions the current evaluation procedure of reward models in LLM post-training. It shows that the accuracy of reward models does not necessarily translate to the improvement in downstream RLHF tasks.\n\nStrengths: \nThis paper studies an important and well-motivate... | {
"decision": "Accept (Spotlight)"
} |
CoQw1dXtGb | 2411.07249v1 | SPDIM: Source-Free Unsupervised Conditional and Label Shift Adaptation in EEG | {
"content": "## Abstract\n\nAbstract The non-stationary nature of electroencephalography (EEG) introduces distribution shifts across domains (e.g., days and subjects), posing a significant challenge to EEG-based neurotechnology generalization.\nWithout labeled calibration data for target domains, the problem is a so... | [
{
"id": "Csl8kZWaDK",
"initial_rating": 8,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 4,
"summary": "This study focuses on the realistic issue of label shifts in EEG across subjects and/or sessions (relative class proportions in target domains when source domains... | {
"rating": "5;5;5;5;8",
"rating_avg": 5.6,
"confidence": "3;5;2;3;3",
"confidence_avg": 3.2,
"soundness": "2;3;3;3;3",
"soundness_avg": 2.8,
"contribution": "3;3;3;2;3",
"contribution_avg": 2.8,
"presentation": "2;3;1;2;4",
"presentation_avg": 2.4
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.249403"
} | {
"id": "XMGfdQIjPh",
"metareview": "This paper was considered \"Rigorous\" with \"clear presentation\", and \"a great example of theory-guided methods design for EEG\". by reviewer PCcy and also endorsed by reviewers CvWN and ocKU, in particular after a valuable discussion with ocKU that surely helped clarify the ... | {
"decision": "Accept (Poster)"
} |
Cpr6Wv2tfr | 2410.04083v2 | OPTAMI: Global Superlinear Convergence of High-order Methods | {
"content": "## Abstract\n\nAbstract Second-order methods for convex optimization outperform first-order methods in terms of theoretical iteration convergence, achieving rates up to O ( k − 5 ) 𝑂 superscript 𝑘 5 O(k^{-5}) italic_O ( italic_k start_POSTSUPERSCRIPT - 5 end_POSTSUPERSCRIPT ) for highly-smooth funct... | [
{
"id": "57OiMi4wFk",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper focuses on the analysis of high order methods for nearly convex functions (i.e. star-convex or convex) with additional growth properties. The authors le... | {
"rating": "3;5;6;8",
"rating_avg": 5.5,
"confidence": "4;4;3;4",
"confidence_avg": 3.75,
"soundness": "3;3;3;4",
"soundness_avg": 3.25,
"contribution": "3;2;2;3",
"contribution_avg": 2.5,
"presentation": "1;2;3;4",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.250479"
} | {
"id": "pY00iX5bT9",
"metareview": "The paper focuses on the analysis of high order methods for mu-strongly star-convex functions. They propose an adaptive variant of Nesterov Accelerated Tensor Method called NATA. The authors also provide OPTAMI, a python library for high order methods. The reviewers found the pa... | {
"decision": "Accept (Poster)"
} |
Cr1XlGBGVm | 2405.12001v3 | Scrutinize What We Ignore: Reining In Task Representation Shift Of Context-Based Offline Meta Reinforcement Learning | {
"content": "## Abstract\n\nAbstract Offline meta reinforcement learning (OMRL) has emerged as a promising approach for interaction avoidance and strong generalization performance by leveraging pre-collected data and meta-learning techniques.\nPrevious context-based approaches predominantly rely on the intuition tha... | [
{
"id": "IaNQTVWl92",
"initial_rating": 8,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This work attempts to explain the performance improvement of the Offline Meta RL (ORML) optimization framework. It identifies that the variation of task represent... | {
"rating": "3;5;6;8",
"rating_avg": 5.5,
"confidence": "4;4;3;3",
"confidence_avg": 3.5,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "3;2;3;3",
"contribution_avg": 2.75,
"presentation": "3;2;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.251459"
} | {
"id": "WVb9D81xfa",
"metareview": "This work investigates task representation shift in offline meta reinforcement learning, providing theoretical insights and practical strategies to control this shift, such as tuning batch sizes and accumulation steps. Experiments are well designed and executed, across two diff... | {
"decision": "Accept (Poster)"
} |
CrOHzVtWmH | 2409.02416v1 | Relative-Translation Invariant Wasserstein Distance | {
"content": "## Abstract\n\nAbstract We introduce a new family of distances, relative-translation invariant Wasserstein distances ( R W p 𝑅 subscript 𝑊 𝑝 RW_{p} italic_R italic_W start_POSTSUBSCRIPT italic_p end_POSTSUBSCRIPT ), for measuring the similarity of two probability distributions under distribution sh... | [
{
"id": "6sVLFEvKdF",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "Many machine learning pipelines rely on training objective losses involving comparing probabilities measures. One of the mostly used in these pipelines are optima... | {
"rating": "3;3;3;5;5",
"rating_avg": 3.8,
"confidence": "4;3;4;4;3",
"confidence_avg": 3.6,
"soundness": "2;3;2;2;3",
"soundness_avg": 2.4,
"contribution": "2;1;2;3;2",
"contribution_avg": 2,
"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": "2026-01-14T22:16:01.252211"
} | {
"id": "6n1jrfJK05",
"metareview": "The authors propose a novel relative-translation invariant Wasserstein ($RW_p$) distance to deal with the translation shift for Wasserstein distance. The authors prove that it satisfies the metric property. The authors propose two algorithmic approach for the proposed problem: t... | {
"decision": "Reject"
} |
CuwjD3cazX | 2409.06411v1 | Length Desensitization in Direct Preference Optimization | {
"content": "## Abstract\n\nAbstract Direct Preference Optimization (DPO) is widely utilized in the Reinforcement Learning from Human Feedback (RLHF) phase to align Large Language Models (LLMs) with human preferences, thereby enhancing both their harmlessness and efficacy. However, it has been observed that DPO tend... | [
{
"id": "DrCsDflgzH",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The authors demonstrate the existence of length sensitivity in the DPO algorithm and analyze this issue theoretically. They propose the LD-DPO algorithm to addres... | {
"rating": "5;5;5",
"rating_avg": 5,
"confidence": "4;4;4",
"confidence_avg": 4,
"soundness": "2;3;3",
"soundness_avg": 2.6666666666666665,
"contribution": "3;2;3",
"contribution_avg": 2.6666666666666665,
"presentation": "4;4;2",
"presentation_avg": 3.3333333333333335
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.252949"
} | {
"id": "COl67g9Oxo",
"metareview": "This paper studies the verbosity bias of the DPO algorithm. It investigates the reasons for verbosity and provides solutions with empirical evaluation.\n\nStrengths: \nThis paper is well-written and easy to understand. The proposed algorithm, LD-DPO, works well on selected task... | {
"decision": "Reject"
} |
CvGqMD5OtX | 2410.01943v1 | CHASE-SQL: Multi-Path Reasoning and Preference Optimized Candidate Selection in Text-to-SQL | {
"content": "## Abstract\n\nAbstract In tackling the challenges of large language model (LLM) performance for Text-to-SQL tasks, we introduce CHASE-SQL, a new framework that employs innovative strategies, using test-time compute in multi-agent modeling to improve candidate generation and selection. CHASE-SQL leverag... | [
{
"id": "ApLSl1nFbv",
"initial_rating": 6,
"confidence": 5,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The article discusses the formation of a pipeline for text2sql, which includes the following main parts: combining multiple methods (including DC-CoT, Query Plan-... | {
"rating": "3;6;6;8",
"rating_avg": 5.75,
"confidence": "5;4;5;5",
"confidence_avg": 4.75,
"soundness": "3;3;3;4",
"soundness_avg": 3.25,
"contribution": "2;3;3;3",
"contribution_avg": 2.75,
"presentation": "3;4;3;4",
"presentation_avg": 3.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.253591"
} | {
"id": "h9UKRmcIxq",
"metareview": "This paper proposes the CHASE-SQL framework, which improves text-to-sql tasks using LLMs. Overall the reviews are quite positive as CHASE-SQL is novel, efficient, and has good results on benchmarks like Bird and Spider. The paper is also well written and provides plenty of detai... | {
"decision": "Accept (Poster)"
} |
CvttyK4XzV | 2410.00153v1 | Beyond Single Concept Vector: Modeling Concept Subspace in LLMs with Gaussian Distribution | {
"content": "## Abstract\n\nAbstract Probing learned concepts in large language models (LLMs) is crucial for understanding how semantic knowledge is encoded internally. Training linear classifiers on probing tasks is a principle approach to denote the vector of a certain concept in the representation space. However,... | [
{
"id": "RpUn11Uo9d",
"initial_rating": 8,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper proposes a robust alternative to the standard practice of learning linear probes in LLMs to find concept representations (for steering).\nSpecifically, ... | {
"rating": "3;5;8;8",
"rating_avg": 6,
"confidence": "3;4;3;4",
"confidence_avg": 3.5,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "3;3;3;3",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.254255"
} | {
"id": "4aLVIjFXSq",
"metareview": "This paper moves from modeling concept representations as directions in a representation space to modeling them as normal distributions (with diagonal covariance) in the representation space. Reviewers appreciated the novelty and clarity, and are optimistic that characterizing t... | {
"decision": "Accept (Poster)"
} |
CxS8mlkOH7 | 2410.12526v1 | Shaping a Stabilized Video by Mitigating Unintended Changes for Concept-Augmented Video Editing | {
"content": "## Abstract\n\nAbstract Text-driven video editing utilizing generative diffusion models has garnered significant attention due to their potential applications. However, existing approaches are constrained by the limited word embeddings provided in pre-training, which hinders nuanced editing targeting op... | [
{
"id": "pUKChjrStK",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper introduces an improved text-driven video editing approach using generative diffusion models. It overcomes limitations of traditional methods by incorpo... | {
"rating": "3;3;3;5;6",
"rating_avg": 4,
"confidence": "4;4;3;5;3",
"confidence_avg": 3.8,
"soundness": "3;2;1;2;3",
"soundness_avg": 2.2,
"contribution": "3;2;1;2;3",
"contribution_avg": 2.2,
"presentation": "3;2;2;2;3",
"presentation_avg": 2.4
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.254907"
} | {
"id": "zLxJWLwnkd",
"metareview": "This paper receives reviews of 5,3,6,3,3. The AC follows the decisions of the reviewers to reject the paper. The main concerns of this paper are: 1) The need to improve on the writing to make the technical contributions clearer. 2) The lack of novelty since the use of LoRA with ... | {
"decision": "Reject"
} |
CxwtuhU40F | 2211.09321v2 | Interpretable Dimensionality Reduction by Feature-preserving Manifold Approximation and Projection | {
"content": "## Abstract\n\nAbstract Nonlinear dimensionality reduction lacks interpretability due to the absence of source features in low-dimensional embedding space.\nWe propose an interpretable method featMAP to preserve source features by tangent space embedding.\nThe core of our proposal is to utilize local si... | [
{
"id": "GwSMk5BG7a",
"initial_rating": 6,
"confidence": 4,
"soundness": 4,
"contribution": 3,
"presentation": 3,
"summary": "This paper advances manifold learning by preserving both the topological structure and source features within the tangent space, facilitating interpretability thr... | {
"rating": "3;5;6;6;6",
"rating_avg": 5.2,
"confidence": "3;4;4;3;4",
"confidence_avg": 3.6,
"soundness": "2;2;3;3;4",
"soundness_avg": 2.8,
"contribution": "2;2;3;3;3",
"contribution_avg": 2.6,
"presentation": "3;3;2;3;3",
"presentation_avg": 2.8
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.255561"
} | {
"id": "WuRhrMv9kM",
"metareview": "While nonlinear dimensionality reduction often lacks interpretability, this paper proposes an interpretable method FeatureMAP to preserve source features by tangent space embedding. Experiments on several datasets validate the effectiveness. After the rebuttal, it receives two b... | {
"decision": "Reject"
} |
Cz8KnDYj1L | 2410.22517v1 | Attention Speaks Volumes: Localizing and Mitigating Bias in Language Models | {
"content": "## Abstract\n\nAbstract We explore the internal mechanisms of how bias emerges in large language models (LLMs) when provided with ambiguous comparative prompts: inputs that compare or enforce choosing between two or more entities without providing clear context for preference. Most approaches for bias m... | [
{
"id": "nFdvg2Q5mL",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper introduces a simple and straightforward method for localizing and mitigating bias in large language models. The approach focuses on the attention score... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "3;4;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;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:01.256236"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
D0LuQNZfEl | 2403.07937v2 | Speech Robust Bench: A Robustness Benchmark For Speech Recognition | {
"content": "## Abstract\n\nAbstract As Automatic Speech Recognition (ASR) models become ever more pervasive, it is important to ensure that they make reliable predictions under corruptions present in the physical and digital world.\nWe propose Speech Robust Bench ( SRB ), a comprehensive benchmark for evaluating th... | [
{
"id": "wlLgldCrMB",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "The authors propose a comprehensive benchmark for automatic speech recognition (ASR) models. The benchmark aims to cover a large variety of real-world data-qualit... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "4;2;3;4",
"confidence_avg": 3.25,
"soundness": "3;2;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "3;2;1;3",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.256950"
} | {
"id": "IOEqYV7ddC",
"metareview": "This work introduces a new benchmark dataset and pipeline designed to evaluate ASR robustness under various scenarios. The proposed framework consists of four key components: (1) clean datasets and noise sources, (2) a bank of perturbations, (3) ASR transcription extraction, and... | {
"decision": "Accept (Poster)"
} |
D0XpSucS3l | 2411.04434v1 | Scaling Laws for Pre-training Agents and World Models | {
"content": "## Abstract\n\nAbstract The performance of embodied agents has been shown to improve by increasing model parameters, dataset size, and compute.\nThis has been demonstrated in domains from robotics to video games, when generative learning objectives on offline datasets (pre-training) are used to model an... | [
{
"id": "nSliSM2JDO",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "## Summary\nThis paper investigates the impact of model size, dataset size, and compute on embodied AI tasks like behavior cloning (BC) and world modeling (WM). U... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "4;4;3;3",
"confidence_avg": 3.5,
"soundness": "1;2;3;3",
"soundness_avg": 2.25,
"contribution": "1;2;2;2",
"contribution_avg": 1.75,
"presentation": "3;3;2;2",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.257616"
} | {
"id": "Bori62Qwsz",
"metareview": "This paper investigates scaling laws in embodied AI, focusing on behavior cloning (BC) and world modeling (WM) tasks. Using a large dataset from gameplay in Bleeding Edge, the study examines the interplay of model size, dataset size, and compute on pre-training losses. The autho... | {
"decision": "Reject"
} |
D1Y2XFgsPI | 2407.19804v1 | Imputation for prediction: beware of diminishing returns. | {
"content": "## Abstract\n\nAbstract Missing values are prevalent across various fields, posing challenges for training and deploying predictive models. In this context, imputation is a common practice, driven by the hope that accurate imputations will enhance predictions. However, recent theoretical and empirical s... | [
{
"id": "uKPhvbD8ab",
"initial_rating": 3,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "This paper empirically studies the effects of various missing value imputation methods, in particular whether better imputation accuracy yields higher prediction ... | {
"rating": "3;3;8;8;8",
"rating_avg": 6,
"confidence": "5;3;4;4;4",
"confidence_avg": 4,
"soundness": "3;2;4;3;3",
"soundness_avg": 3,
"contribution": "1;2;3;3;3",
"contribution_avg": 2.4,
"presentation": "3;3;3;4;1",
"presentation_avg": 2.8
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.258282"
} | {
"id": "oGuYrAIpHq",
"metareview": "The paper proposes an empirical study that analyses the effectiveness of advanced imputation methods for missing values.\nAll reviewers are happy about the findings of the paper and agree to accept the paper.",
"additional_comments": "Reviewers were happy about the rebuttals a... | {
"decision": "Accept (Spotlight)"
} |
D23JcXiUwf | 2411.01829v1 | Formal Theorem Proving by Rewarding LLMs to Decompose Proofs Hierarchically | {
"content": "## Abstract\n\nAbstract Mathematical theorem proving is an important testbed for large language models’ deep and abstract reasoning capability. This paper focuses on improving LLMs’ ability to write proofs in formal languages that permit automated proof verification/evaluation. Most previous results pro... | [
{
"id": "wAs6xolc7A",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 4,
"presentation": 2,
"summary": "According to my understanding, this paper has the following two main contributions:\n\n1. **A Setting for Theorem Proving without the Help of Human-Written Lemmas... | {
"rating": "3;3;6;6",
"rating_avg": 4.5,
"confidence": "4;4;4;4",
"confidence_avg": 4,
"soundness": "2;3;4;3",
"soundness_avg": 3,
"contribution": "1;2;3;4",
"contribution_avg": 2.5,
"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": "2026-01-14T22:16:01.258981"
} | {
"id": "fjQqbNh7Ql",
"metareview": "This paper focuses on neural theorem proving in contexts where the relevant premises are unavailable. The authors propose Proof Decomposer (ProD), a RL approach that utilizes hindsight experience replay to reward a LLM in hierarchically decomposing a given theorem into lemmas, a... | {
"decision": "Reject"
} |
D2as3jDmRA | 2409.02097v3 | LinFusion: 1 GPU, 1 Minute, 16K Image | {
"content": "## Abstract\n\nAbstract Modern diffusion models, particularly those utilizing a Transformer-based UNet for denoising, rely heavily on self-attention operations to manage complex spatial relationships, thus achieving impressive generation performance.\nHowever, this existing paradigm faces significant ch... | [
{
"id": "8WOr80DgFb",
"initial_rating": 6,
"confidence": 5,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper introduces LinFusion, a versatile pipeline designed to enhance GPU memory efficiency and boost sampling speed across various diffusion models for image... | {
"rating": "5;5;6;8",
"rating_avg": 6,
"confidence": "4;4;4;4",
"confidence_avg": 4,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "2;2;3;4",
"contribution_avg": 2.75,
"presentation": "3;3;4;3",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.259664"
} | {
"id": "6zSH1j17tY",
"metareview": "This paper introduces LinFusion, a versatile pipeline designed to enhance GPU memory efficiency and accelerate sampling speeds in various diffusion models for image generation. LinFusion investigates recent linear-attention mechanisms to identify key factors contributing to thei... | {
"decision": "Reject"
} |
D48jvLN45W | 2411.08017v1 | Wavelet Latent Diffusion (WaLa): Billion-Parameter 3D Generative Model with Compact Wavelet Encodings | {
"content": "## Abstract\n\nAbstract Large-scale 3D generative models require substantial computational resources yet often fall short in capturing fine details and complex geometries at high resolutions. We attribute this limitation to the inefficiency of current representations, which lack the compactness required... | [
{
"id": "kw0lpbFzWD",
"initial_rating": 6,
"confidence": 4,
"soundness": 4,
"contribution": 4,
"presentation": 3,
"summary": "This paper presents an approach to encode 3D shapes (TSDF) into a latent space using Wavelet representation via VQ-VAE. Additionally, it trains several latent dif... | {
"rating": "3;3;3;6",
"rating_avg": 3.75,
"confidence": "5;5;5;4",
"confidence_avg": 4.75,
"soundness": "4;3;3;4",
"soundness_avg": 3.5,
"contribution": "2;1;1;4",
"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": "2026-01-14T22:16:01.260572"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
D5X6nPGFUY | 2410.18857v1 | Probabilistic Language-Image Pre-Training | {
"content": "## Abstract\n\nAbstract Vision-language models (VLMs) embed aligned image-text pairs into a joint space but often rely on deterministic embeddings, assuming a one-to-one correspondence between images and texts. This oversimplifies real-world relationships, which are inherently many-to-many, with multipl... | [
{
"id": "6JfhzmfkXr",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "The paper proposes a novel approach for vision-language models (VLMs) that leverages probabilistic embeddings to capture better the inherent many-to-many relation... | {
"rating": "5;5;5;6;6",
"rating_avg": 5.4,
"confidence": "3;2;3;5;3",
"confidence_avg": 3.2,
"soundness": "3;2;3;3;3",
"soundness_avg": 2.8,
"contribution": "2;2;3;3;2",
"contribution_avg": 2.4,
"presentation": "3;2;2;2;2",
"presentation_avg": 2.2
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.261604"
} | {
"id": "Mrx0XsdFf3",
"metareview": "The paper introduces Probabilistic Language-Image Pre-training (ProLIP), an approach to vision-language models that integrates probabilistic modelling to capture the inherent uncertainty in image-text relationships. ProLIP departs from conventional deterministic embeddings by ma... | {
"decision": "Accept (Poster)"
} |
D6zn6ozJs7 | 2406.08772v2 | MMFakeBench: A Mixed-Source Multimodal Misinformation Detection Benchmark for LVLMs | {
"content": "## Abstract\n\nAbstract Current multimodal misinformation detection (MMD) methods often assume a single source and type of forgery for each sample, which is insufficient for real-world scenarios where multiple forgery sources coexist.\nThe lack of a benchmark for mixed-source misinformation has hindered... | [
{
"id": "8Oh7u3ftrK",
"initial_rating": 8,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 4,
"summary": "This paper looks at creating a multimodal misinformation dataset called MMFakeBench that can be used to evaluate detection methods and Large-Vision Models (LVLMs)... | {
"rating": "5;5;6;8;8",
"rating_avg": 6.4,
"confidence": "4;4;4;5;4",
"confidence_avg": 4.2,
"soundness": "3;2;3;3;3",
"soundness_avg": 2.8,
"contribution": "3;1;3;3;3",
"contribution_avg": 2.6,
"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": "2026-01-14T22:16:01.262546"
} | {
"id": "0ZQwZKfu9h",
"metareview": "This work introduces a mixed-source multimodal misinformation detection dataset (MMD). The authors provide comprehensive evaluations on the proposed evaluation settings, and introduce a new baseline for this dataset. Most of the reviewers have agreed that the authors have succes... | {
"decision": "Accept (Poster)"
} |
D756s2YQ6b | 2410.05697v1 | Diffusing to the Top: Boost Graph Neural Networks with Minimal Hyperparameter Tuning | {
"content": "## Abstract\n\nAbstract Graph Neural Networks (GNNs) are proficient in graph representation learning and achieve promising performance on versatile tasks such as node classification and link prediction.\nUsually, a comprehensive hyperparameter tuning is essential for fully unlocking GNN’s top performanc... | [
{
"id": "zhvS9bh94l",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 3,
"presentation": 3,
"summary": "The paper introduces GNN-Diff, a latent diffusion model designed to enhance GNN training while minimizing the need for hyperparameter tuning. The author claim tha... | {
"rating": "3;5;6;8",
"rating_avg": 5.5,
"confidence": "4;3;3;2",
"confidence_avg": 3,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "3;2;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": "2026-01-14T22:16:01.263486"
} | {
"id": "lieFtAlsbl",
"metareview": "The paper introduces a method for improving the performance of GNNs with minimal hyperparameter tuning. It propose a graph-conditioned generative model to generate (better) hyperparameter configurations based on lightly-tuned suboptimal parameters. This approach to parameter tu... | {
"decision": "Accept (Poster)"
} |
D7PQ54l5Q1 | 2409.08551v1 | Think Twice Before You Act: Improving Inverse Problem Solving With MCMC | {
"content": "## Abstract\n\nAbstract Recent studies demonstrate that diffusion models can serve as a strong prior for solving inverse problems. A prominent example is Diffusion Posterior Sampling (DPS), which approximates the posterior distribution of data given the measure using Tweedie’s formula. Despite the merit... | [
{
"id": "KT5BeSWMMS",
"initial_rating": 3,
"confidence": 5,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "The authors propose a new algorithm for solving inverse problems with pre-trained diffusion models. The proposed algorithm is a variation of Diffusion Posterior S... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "5;4;4;4",
"confidence_avg": 4.25,
"soundness": "3;2;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;2;2",
"contribution_avg": 2,
"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": "2026-01-14T22:16:01.264201"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
D9GoWJJxS5 | 2406.10576v2 | Bypass Back-propagation: Optimization-based Structural Pruning for Large Language Models via Policy Gradient | {
"content": "## Abstract\n\nAbstract Compared to the moderate size of neural network models, structural weight pruning on the Large-Language Models (LLMs) imposes a novel challenge on the efficiency of the pruning algorithms, due to the heavy computation/memory demands of the LLMs. Recent efficient LLM pruning metho... | [
{
"id": "BQdz1yUPOW",
"initial_rating": 5,
"confidence": 4,
"soundness": 4,
"contribution": 3,
"presentation": 2,
"summary": "In this work, the authors leverage policy gradient estimation to optimize pruning masks without relying on backpropagation. They validate the efficacy of their ap... | {
"rating": "3;5;5;6;6",
"rating_avg": 5,
"confidence": "4;4;4;3;4",
"confidence_avg": 3.8,
"soundness": "2;2;4;3;3",
"soundness_avg": 2.8,
"contribution": "2;3;3;3;4",
"contribution_avg": 3,
"presentation": "2;3;2;2;3",
"presentation_avg": 2.4
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.264870"
} | {
"id": "vhdwn51dO3",
"metareview": "The paper titled \"Bypass Back-propagation: Optimization-based Structural Pruning for Large Language Models via Policy Gradient\" proposes a novel method for pruning large language models (LLMs) that bypasses traditional back-propagation techniques. The authors introduce an opti... | {
"decision": "Reject"
} |
D9JSxF2Xhx | 2312.07252v2 | Identifying Drivers of Predictive Aleatoric Uncertainty | {
"content": "## Abstract\n\nAbstract Explainability and uncertainty quantification are two pillars of trustable artificial intelligence. However, the reasoning behind uncertainty estimates is generally left unexplained. Identifying the drivers of uncertainty complements explanations of point predictions in recognizi... | [
{
"id": "bH7k0HQbAe",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "The paper proposes a novel method for explaining predictive aleatoric uncertainty by explaining the variance output in a heteroscedastic regression model. Their m... | {
"rating": "3;3;5;6",
"rating_avg": 4.25,
"confidence": "3;4;3;3",
"confidence_avg": 3.25,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "1;2;2;2",
"contribution_avg": 1.75,
"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": "2026-01-14T22:16:01.265594"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
DAEXilQHYU | 2402.04062v2 | Link Prediction with Relational Hypergraphs | {
"content": "## Abstract\n\nAbstract Link prediction with knowledge graphs has been thoroughly studied in graph machine learning, leading to a rich landscape of graph neural network architectures with successful applications. Nonetheless, it remains challenging to transfer the success of these architectures to relat... | [
{
"id": "KOePo8DJTS",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "This paper proposes a new method for link prediction in relational hypergraphs, where the task focuses on k-ary relations. It first investigates the expressive po... | {
"rating": "5;6;6;6",
"rating_avg": 5.75,
"confidence": "4;3;3;4",
"confidence_avg": 3.5,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "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": "2026-01-14T22:16:01.266569"
} | {
"id": "pxeJ6lCo7g",
"metareview": "This submission violated the anonymity policy. In the provided link (https://anonymous.4open.science/r/HCNet/README.md), when clicking the blue-colored text, Link Prediction with Relational Hypergraphs, the arXiv version of the paper is shown, which reveals all authors' informat... | {
"decision": "Reject"
} |
DC8bsa9bzY | 2410.13211v1 | Estimating the Probabilities of Rare Outputs in Language Models | {
"content": "## Abstract\n\nAbstract We consider the problem of low probability estimation : given a machine learning model and a formally-specified input distribution, how can we estimate the probability of a binary property of the model’s output, even when that probability is too small to estimate by random sampli... | [
{
"id": "DmrUWcvGVf",
"initial_rating": 10,
"confidence": 4,
"soundness": 3,
"contribution": 4,
"presentation": 4,
"summary": "This paper studies low-probability estimation - the problem of estimating the probability that model outputs will satisfy a certain criteria on inputs sampled fr... | {
"rating": "3;6;6;10",
"rating_avg": 6.25,
"confidence": "4;3;2;4",
"confidence_avg": 3.25,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "2;3;3;4",
"contribution_avg": 3,
"presentation": "3;2;4;4",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.267710"
} | {
"id": "4dBeVs3YZf",
"metareview": "All four reviewers recognize the importance and novelty of the paper’s core problem: estimating low-probability (rare) outputs of language models on a formally-specified input distribution. This problem is non-trivial and has significant implications for understanding worst-case... | {
"decision": "Accept (Spotlight)"
} |
DCandSZ2F1 | 2410.08017v2 | Fast Feedforward 3D Gaussian Splatting Compression | {
"content": "## Abstract\n\nAbstract With 3D Gaussian Splatting (3DGS) advancing real-time and high-fidelity rendering for novel view synthesis, storage requirements pose challenges for their widespread adoption. Although various compression techniques have been proposed, previous art suffers from a common limitatio... | [
{
"id": "7ZmyYuwZjC",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper proposes a new compression model for 3D Gaussian Splatting (3DGS) representations. It aims to reduce compression time in an optimization-free pipeline, ... | {
"rating": "5;6;6;6",
"rating_avg": 5.75,
"confidence": "3;4;4;4",
"confidence_avg": 3.75,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "2;3;3;3",
"contribution_avg": 2.75,
"presentation": "2;2;4;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.268395"
} | {
"id": "LqLvPiVems",
"metareview": "This paper presents an efficient 3D Gaussian Splatting (3DGS) compression method. The proposed approach is novel, offering an optimization-free pipeline for 3DGS compression in a feed-forward manner. A multi-path entropy module controls the tradeoff between size and quality, whi... | {
"decision": "Accept (Poster)"
} |
DD11okKg13 | 2407.15589v3 | Exploring the Effectiveness of Object-Centric Representations in Visual Question Answering: Comparative Insights with Foundation Models | {
"content": "## Abstract\n\nAbstract Object-centric (OC) representations, which represent the state of a visual scene by modeling it as a composition of objects, have the potential to be used in various downstream tasks to achieve systematic compositional generalization and facilitate reasoning. However, these claim... | [
{
"id": "EkTGAYZNbW",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper presents a comprehensive empirical study comparing object-centric (OC) representations with foundation models on Visual Question Answering tasks. Throu... | {
"rating": "3;5;6;6",
"rating_avg": 5,
"confidence": "4;5;3;3",
"confidence_avg": 3.75,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "2;4;3;4",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.269216"
} | {
"id": "GqXpPa1hae",
"metareview": "The submission conducts an empirical study on the impact of object-centric representations for visual question answering, and aims to put the observations into perspective with \"foundation models\". The original submission primarily relied on synthetic datasets for their analys... | {
"decision": "Accept (Poster)"
} |
DDNFTaVQdU | 2307.07735v2 | Faster Algorithms for Structured Linear and Kernel Support Vector Machines | {
"content": "## Abstract\n\nQuadratic programming is a ubiquitous prototype in convex programming. Many combinatorial optimizations on graphs and machine learning problems can be formulated as quadratic programming; for example, Support Vector Machines (SVMs). Linear and kernel SVMs have been among the most popular ... | [
{
"id": "nl1nwQOQtm",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The authors design the first nearly-linear time algorithm for solving SVM problems via quadratic programs whenever the quadratic objective admits a low-rank facto... | {
"rating": "3;6;6",
"rating_avg": 5,
"confidence": "3;3;4",
"confidence_avg": 3.3333333333333335,
"soundness": "3;4;4",
"soundness_avg": 3.6666666666666665,
"contribution": "2;3;4",
"contribution_avg": 3,
"presentation": "2;3;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.270591"
} | {
"id": "a60nGajEoW",
"metareview": "This paper investigates solving SVMs via quadratic programming and proposes a novel, efficient, nearly-linear time algorithm for cases where the quadratic objective admits a low-rank factorization and the number of linear constraints is small—settings that are typical for SVMs. ... | {
"decision": "Accept (Poster)"
} |
DEOV74Idsg | 2407.04903v2 | MMSci: A Dataset for Graduate-Level Multi-Discipline Multimodal Scientific Understanding | {
"content": "## Abstract\n\nAbstract The rapid development of Multimodal Large Language Models (MLLMs) is making AI-driven scientific assistants increasingly feasible, with interpreting scientific figures being a crucial task. However, existing datasets and benchmarks focus mainly on basic charts and limited science... | [
{
"id": "hi3uQ4jIj0",
"initial_rating": 8,
"confidence": 4,
"soundness": 3,
"contribution": 4,
"presentation": 3,
"summary": "The paper introduces MMSci, a multimodal, multidisciplinary dataset collected from open-access scientific articles published in Nature Communications. Spanning 72... | {
"rating": "3;5;5;8",
"rating_avg": 5.25,
"confidence": "5;4;4;4",
"confidence_avg": 4.25,
"soundness": "1;2;3;3",
"soundness_avg": 2.25,
"contribution": "3;2;3;4",
"contribution_avg": 3,
"presentation": "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": "2026-01-14T22:16:01.272055"
} | {
"id": "vcnhMjaPJQ",
"metareview": "The paper presents MMSci, a rich dataset sourced from Nature Communications, containing 131K top-tier articles and figures across 72 scientific disciplines. It sets up benchmarks comprising 1K tasks to assess the performance of Large Multimodal Models (LMMs) in comprehending int... | {
"decision": "Reject"
} |
DGjzxNRbKU | 2402.03077v2 | Markov Persuasion Processes: Learning to Persuade From Scratch | {
"content": "## Abstract\n\nAbstract In Bayesian persuasion , an informed sender strategically discloses information to a receiver so as to persuade them to undertake desirable actions.\nRecently, Markov persuasion processes (MPPs) have been introduced to capture sequential scenarios where a sender faces a stream of... | [
{
"id": "kqvkJQdcee",
"initial_rating": 3,
"confidence": 5,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper considers learning in a Markov persuasion process (MPP) where the learner cannot take actions directly, and must persuade a receiver to take the intend... | {
"rating": "3;3;5;5;6",
"rating_avg": 4.4,
"confidence": "3;5;3;2;3",
"confidence_avg": 3.2,
"soundness": "3;3;3;2;4",
"soundness_avg": 3,
"contribution": "1;2;2;2;3",
"contribution_avg": 2,
"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": "2026-01-14T22:16:01.273009"
} | {
"id": "yFOPMcMxEu",
"metareview": "The reviewers acknowledged that the paper investigates an interesting problem setting in sequential Bayesian persuasion where a sender has limited information about the receiver and provides a thorough theoretical analysis for this setting. However, the reviewers pointed out sev... | {
"decision": "Reject"
} |
DJSZGGZYVi | 2410.06940v1 | Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think | {
"content": "## Abstract\n\nAbstract Recent studies have shown that the denoising process in (generative) diffusion models can induce meaningful (discriminative) representations inside the model, though the quality of these representations still lags behind those learned through recent self-supervised learning metho... | [
{
"id": "kIhFzftsQH",
"initial_rating": 8,
"confidence": 4,
"soundness": 4,
"contribution": 4,
"presentation": 3,
"summary": "The paper presents a novel argument that a primary bottleneck in training large-scale diffusion models for generation lies in learning effective representations. ... | {
"rating": "6;6;6;8;8;8",
"rating_avg": 7,
"confidence": "4;3;4;4;4;4",
"confidence_avg": 3.8333333333333335,
"soundness": "3;3;3;4;4;4",
"soundness_avg": 3.5,
"contribution": "3;3;3;3;3;4",
"contribution_avg": 3.1666666666666665,
"presentation": "4;3;3;3;4;3",
"presentation_avg": 3.333333333333333... | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Oral",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.274179"
} | {
"id": "gUi5idCATZ",
"metareview": "This paper introduces REPresentation Alignment (REPA), a simple yet effective regularization technique that significantly enhances the efficiency and performance of diffusion transformers by aligning their representations with pretrained self-supervised visual encoders. Reviewer... | {
"decision": "Accept (Oral)"
} |
DKA7Hx7PSt | 2409.20449v2 | Linear Projections of Teacher Embeddings for Few-Class Distillation | {
"content": "## Abstract\n\nAbstract Knowledge Distillation (KD) has emerged as a promising approach for transferring knowledge from a larger, more complex teacher model to a smaller student model. Traditionally, KD involves training the student to mimic the teacher’s output probabilities, while more advanced techni... | [
{
"id": "LL1wj0xv9u",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This manuscript proposes Learning Embedding Linear Projections, a method for distilling knowledge from a teacher model's representations. The proposed method iden... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "5;3;4;3",
"confidence_avg": 3.75,
"soundness": "1;3;3;3",
"soundness_avg": 2.5,
"contribution": "2;3;3;3",
"contribution_avg": 2.75,
"presentation": "2;3;3;2",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.275028"
} | {
"id": "jN3z62oVz9",
"metareview": "This paper aims to handle binary or few-class classification when applying knowledge distillation. To achieve this goal, the authors increase the number of classes by projecting the final embedding vectors of the student and the teacher into several PCA subsets. In this way, the... | {
"decision": "Reject"
} |
DKZjYuB6gc | 2408.09310v3 | Narrowing the Focus: Learned Optimizers for Pretrained Models | {
"content": "## Abstract\n\nAbstract In modern deep learning, the models are learned by applying gradient updates using an optimizer, which transforms the updates based on various statistics. Optimizers are often hand-designed and tuning their hyperparameters is a big part of the training process. Learned optimizers... | [
{
"id": "foqfIa6tMH",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "This work introduces L3RS, a learned learning rate scheduler that ensembles updates from base optimizers like Adam and SGD. \n\nL3RS is parameterized as a shared ... | {
"rating": "3;5;5;5",
"rating_avg": 4.5,
"confidence": "3;4;4;4",
"confidence_avg": 3.75,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;3;2;2",
"contribution_avg": 2.25,
"presentation": "3;2;1;2",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.275794"
} | {
"id": "FcNNL8ucmV",
"metareview": "The submission introduces L3RS (Learned Layer-wise Learning Rate Scheduler), an optimizer that learns to combine the updates provided by hand-designed optimizers such as SGD and Adam, as well as heuristics such as multi-scale EMA.\nAfter the initial round of reviews, this receiv... | {
"decision": "Reject"
} |
DKgAFfCs5F | 2410.12592v1 | Cocoon: Robust Multi-Modal Perception with Uncertainty-Aware Sensor Fusion | {
"content": "## Abstract\n\nAbstract An important paradigm in 3D object detection is the use of multiple modalities to enhance accuracy in both normal and challenging conditions, particularly for long-tail scenarios. To address this, recent studies have explored two directions of adaptive approaches: MoE-based adapt... | [
{
"id": "y2bRyGvhqI",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "This paper introduces an object- and feature-level uncertainty-aware fusion framework for robust multi-modal sensor fusion in 3D detection scene of autonomous dri... | {
"rating": "5;5;6",
"rating_avg": 5.333333333333333,
"confidence": "3;4;4",
"confidence_avg": 3.6666666666666665,
"soundness": "3;3;3",
"soundness_avg": 3,
"contribution": "3;2;3",
"contribution_avg": 2.6666666666666665,
"presentation": "4;3;2",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.276504"
} | {
"id": "ktMKUxgyEK",
"metareview": "This paper proposes a new object- and feature-level uncertainty-aware multimodal fusion framework for 3D object detection tasks. The proposed framework adopts a feature aligner and a feature impression strategy to achieve uncertainty quantification for heterogeneous representati... | {
"decision": "Accept (Poster)"
} |
DKkQtRMowq | 2410.10877v1 | Improving Data Efficiency via Curating LLM-Driven Rating Systems | {
"content": "## Abstract\n\nAbstract Instruction tuning is critical for adapting large language models (LLMs) to downstream tasks, and recent studies have demonstrated that small amounts of human-curated data can outperform larger datasets, challenging traditional data scaling laws. While LLM-based data quality rati... | [
{
"id": "WxZaMAYCAs",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "The authors introduce a data curation algorithm, DS^2, to correct inaccuracies in LLM-based data quality evaluation strategies. The authors first show how, lever... | {
"rating": "5;5;6;6",
"rating_avg": 5.5,
"confidence": "4;4;4;4",
"confidence_avg": 4,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "3;3;2;2",
"contribution_avg": 2.5,
"presentation": "3;2;3;2",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.277367"
} | {
"id": "SmJ0bm2xKn",
"metareview": "The paper introduces a curation method for finetuning data that selects data samples and demonstrates that a model finetuned on a small dataset curated with the method can outperform a model trained on the larger full-scale dataset. \n\nAll reviewers find the proposed approach i... | {
"decision": "Accept (Poster)"
} |
DLDuVbxORA | 2409.13652v1 | OATS: Outlier-Aware Pruning Through Sparse and Low Rank Decomposition | {
"content": "## Abstract\n\nAbstract The recent paradigm shift to large-scale foundation models has brought about a new era for deep learning that, while has found great success in practice, has also been plagued by prohibitively expensive costs in terms of high memory consumption and compute. To mitigate these issu... | [
{
"id": "1lDbd62yTE",
"initial_rating": 8,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 4,
"summary": "The paper introduces OATS (Outlier-Aware Pruning Through Sparse and Low-Rank Decomposition), a method designed to compress large transformer models without the ne... | {
"rating": "3;6;8;8",
"rating_avg": 6.25,
"confidence": "5;4;4;3",
"confidence_avg": 4,
"soundness": "2;3;4;3",
"soundness_avg": 3,
"contribution": "1;3;3;3",
"contribution_avg": 2.5,
"presentation": "2;2;4;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.278141"
} | {
"id": "OZk7kckhdo",
"metareview": "**Summary**\n\nThe paper presents OATS (Outlier-Aware Pruning Through Sparse and Low-Rank Decomposition), a technique developed to compress large transformer models and reduce memory and compute costs without requiring costly retraining. This method is based on decomposing the w... | {
"decision": "Accept (Poster)"
} |
DLhjxxXYwH | 2407.19044v2 | Advancing Neural Network Performance through Emergence-Promoting Initialization Scheme | {
"content": "## Abstract\n\nAbstract We introduce a novel yet straightforward neural network initialization scheme that modifies conventional methods like Xavier and Kaiming initialization. Inspired by the concept of emergence and leveraging the emergence measures proposed by Li (2023), our method adjusts the layer-... | [
{
"id": "okwVO1LvCl",
"initial_rating": 3,
"confidence": 3,
"soundness": 2,
"contribution": 1,
"presentation": 1,
"summary": "From the perspective of emergence-promoting, the paper presents a method to initialize the neural networks for training. Starting from abstract definition of emer... | {
"rating": "3;3;3",
"rating_avg": 3,
"confidence": "4;4;3",
"confidence_avg": 3.6666666666666665,
"soundness": "2;2;2",
"soundness_avg": 2,
"contribution": "2;2;1",
"contribution_avg": 1.6666666666666667,
"presentation": "2;2;1",
"presentation_avg": 1.6666666666666667
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.279046"
} | {
"id": "aw8IDkmKbw",
"metareview": "This paper proposes a neural network initialization scheme by adjusting the layer-wise weight scaling factors to improve the conventional methods Xavier and Kaiming initialization. Experiments on various architectures are conducted. However, multiple reviewers raise concerns reg... | {
"decision": "Reject"
} |
DOA1WSPZSi | 2410.08085v1 | Can Knowledge Graphs Make Large Language Models More Trustworthy? An Empirical Study over Open-ended Question Answering | {
"content": "## Abstract\n\nAbstract Recent works integrating Knowledge Graphs (KGs) have led to promising improvements in enhancing reasoning accuracy of Large Language Models (LLMs).\nHowever, current benchmarks mainly focus on closed tasks, leaving a gap in the assessment of more complex, real-world scenarios. Th... | [
{
"id": "SuG1QJdRM9",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "This paper introduces OKGQA, a benchmark designed to assess the trustworthiness of LLMs augmented with KGs in open-ended QA scenarios. It also includes a perturbe... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "4;4;3;4",
"confidence_avg": 3.75,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"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": "2026-01-14T22:16:01.279727"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
DPlUWG4WMw | 2406.11520v2 | Operator Deep Smoothing for Implied Volatility | {
"content": "## Abstract\n\nAbstract We devise a novel method for nowcasting implied volatility based on neural operators.\nBetter known as implied volatility smoothing in the financial industry, nowcasting of implied volatility means constructing a smooth surface that is consistent with the prices presently observe... | [
{
"id": "GlERT7RhMq",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper introduces a novel approach to smoothing implied volatility using neural operators, focusing on Graph Neural Operator (GNO) architectures. Unlike tradi... | {
"rating": "5;5;6",
"rating_avg": 5.333333333333333,
"confidence": "2;2;4",
"confidence_avg": 2.6666666666666665,
"soundness": "2;3;3",
"soundness_avg": 2.6666666666666665,
"contribution": "3;2;2",
"contribution_avg": 2.3333333333333335,
"presentation": "2;2;3",
"presentation_avg": 2.33333333333333... | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.280502"
} | {
"id": "Ld0xdsHGll",
"metareview": "The authors proposed a neural operator approach for implied volatility surface smoothing, and the trained model can directly map observed data to implied volatility surfaces, eliminating the need for calibration every time. Some reviewers pointed out missing baseline methods, su... | {
"decision": "Accept (Poster)"
} |
DPynq6bSHn | 2409.17892v1 | EMMA-500: Enhancing Massively Multilingual Adaptation of Large Language Models | {
"content": "## Abstract\n\nAbstract In this work, we introduce EMMA-500 , a large-scale multilingual language model continue-trained on texts across 546 languages designed for enhanced multilingual performance, with a focus on improving language coverage for low-resource languages.\nTo facilitate continual pre-trai... | [
{
"id": "7iXr3bLShP",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "EMMA-500 outlines work around three technical contributions:\n* EMMA-500 -> A multilingual model trained on 546 languages\n* MaLA Corpus -> A multilingual corpus ... | {
"rating": "3;3;5",
"rating_avg": 3.6666666666666665,
"confidence": "4;4;4",
"confidence_avg": 4,
"soundness": "3;2;3",
"soundness_avg": 2.6666666666666665,
"contribution": "2;2;3",
"contribution_avg": 2.3333333333333335,
"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": "2026-01-14T22:16:01.281388"
} | {
"id": "Kt6VLhybx6",
"metareview": "The submission introduces EMMA-500, a multilingual language model trained across 546 languages via continual pre-training on the MaLA corpus, a multilingual dataset compiled from various existing sources. The paper highlight improvements in multilingual performance, particularly... | {
"decision": "Reject"
} |
DSsSPr0RZJ | 2409.07703v1 | DSBench: How Far Are Data Science Agents from Becoming Data Science Experts? | {
"content": "## Abstract\n\nAbstract Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) have demonstrated impressive language/vision reasoning abilities, igniting the recent trend of building agents for targeted applications such as shopping assistants or AI software engineers. Recently, many data... | [
{
"id": "ffJLgUw67W",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper introduces DSBench, a comprehensive data science benchmark designed to assess the performance of data science agents on real-world tasks. DSBench integ... | {
"rating": "6;6;6;6;8",
"rating_avg": 6.4,
"confidence": "4;4;2;3;4",
"confidence_avg": 3.4,
"soundness": "3;3;3;3;4",
"soundness_avg": 3.2,
"contribution": "3;2;2;2;4",
"contribution_avg": 2.6,
"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": "2026-01-14T22:16:01.282240"
} | {
"id": "PV4DCBK7PH",
"metareview": "The paper introduces DSBench, a comprehensive benchmark for evaluating data science agents through 540 tasks sourced from ModelOff and Kaggle competitions, offering a significant contribution to the field of AI-driven data science evaluation (Reviewers TVnT and EZxL). The benchm... | {
"decision": "Accept (Poster)"
} |
DSyHRkpI7v | 2405.00746v1 | Leveraging Sub-Optimal Data for Human-in-the-Loop Reinforcement Learning | {
"content": "## Abstract\n\nAbstract To create useful reinforcement learning (RL) agents, step zero is to design a suitable reward function that captures the nuances of the task. However, reward engineering can be a difficult and time-consuming process.\nInstead, human-in-the-loop (HitL) RL allows agents to learn re... | [
{
"id": "nxmTOSwxUE",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 1,
"presentation": 4,
"summary": "The submission proposed a simple framework, namely Sub-optimal Data Pre-training (SDP), that leverages reward-free, suboptimal data to warm-start the reward model... | {
"rating": "5;6;6;8",
"rating_avg": 6.25,
"confidence": "4;4;3;4",
"confidence_avg": 3.75,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "1;3;2;3",
"contribution_avg": 2.25,
"presentation": "4;3;4;4",
"presentation_avg": 3.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.282918"
} | {
"id": "5UdkyBdRhj",
"metareview": "This paper presents a method for learning reward model from human by leveraging suboptimal data to warm-start the learning process. Minimal environmental rewards are assigned to reward free data and then the reward function is learned through scalar or preference feedback from h... | {
"decision": "Accept (Poster)"
} |
DTQlvDCAql | 2408.17443v3 | HERMES: temporal-coHERent long-forM understanding with Episodes and Semantics | {
"content": "## Abstract\n\nAbstract Existing research often treats long-form videos as extended short videos, leading to several limitations: inadequate capture of long-range dependencies, inefficient processing of redundant information, and failure to extract high-level semantic concepts. To address these issues, ... | [
{
"id": "04CBqCx6VE",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "The paper presents HERMES, a framework designed for long-form video understanding that leverages episodic memory and semantic knowledge. The authors propose two m... | {
"rating": "3;3;5;8",
"rating_avg": 4.75,
"confidence": "4;5;4;5",
"confidence_avg": 4.5,
"soundness": "2;2;2;4",
"soundness_avg": 2.5,
"contribution": "2;2;2;4",
"contribution_avg": 2.5,
"presentation": "3;3;3;4",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Conference Withdrawn Submission",
"venueid": "ICLR.cc/2025/Conference/Withdrawn_Submission",
"processed_at": "2026-01-14T22:16:01.283519"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
DTatjJTDl1 | 2405.16381v1 | Trivialized Momentum Facilitates Diffusion Generative Modeling on Lie Groups | {
"content": "## Abstract\n\nAbstract The generative modeling of data on manifold is an important task, for which diffusion models in flat spaces typically need nontrivial adaptations. This article demonstrates how a technique called ‘trivialization’ can transfer the effectiveness of diffusion models in Euclidean spa... | [
{
"id": "IqvbnTdLNb",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The authors present a method for diffusion on Lie groups. The method is based on the recently published idea of trivialized momentum, that is, diffusion (i.e. noi... | {
"rating": "3;5;8;8",
"rating_avg": 6,
"confidence": "4;3;4;3",
"confidence_avg": 3.5,
"soundness": "2;2;4;4",
"soundness_avg": 3,
"contribution": "2;2;4;3",
"contribution_avg": 2.75,
"presentation": "2;3;3;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.284292"
} | {
"id": "GVp2KezcjO",
"metareview": "The paper studies diffusion processes on Lie groups by introducing the diffusion process on corresponding Lie algebra. This allows introducing the diffusion processes and training corresponding generative models on SO(n) and U(n) manifolds. All the reviewers have recognized the ... | {
"decision": "Accept (Poster)"
} |
DVlPp7Jd7P | 2410.01537v1 | Attention layers provably solve single-location regression | {
"content": "## Abstract\n\nAbstract. Attention-based models, such as Transformer, excel across various tasks but lack a comprehensive theoretical understanding, especially regarding token-wise sparsity and internal linear representations. To address this gap, we introduce the single-location regression task, where ... | [
{
"id": "PFDbB70K0L",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 4,
"summary": "This paper introduces a new theoretical approach to understand attention mechanisms by considering a task called single-location regression. The authors consider ... | {
"rating": "5;6;6;8",
"rating_avg": 6.25,
"confidence": "4;4;3;4",
"confidence_avg": 3.75,
"soundness": "3;3;3;4",
"soundness_avg": 3.25,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "3;3;4;4",
"presentation_avg": 3.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.285384"
} | {
"id": "ftDA6MVRpK",
"metareview": "All reviewers find the paper written well presenting analysis of attention learning a rather simple problem. Authors argue that such thorough analysis is missing for Attention; even though done for a simpler setting, involves non-trivial complexities in the analysis. Other conce... | {
"decision": "Accept (Poster)"
} |
DWLlTNhig1 | 2409.04617v2 | Sparse Rewards Can Self-Train Dialogue Agents | {
"content": "## Abstract\n\nAbstract Recent advancements in state-of-the-art (SOTA) Large Language Model (LLM) agents, especially in multi-turn dialogue tasks, have been primarily driven by supervised fine-tuning and high-quality human feedback. However, as base LLM models continue to improve, acquiring meaningful h... | [
{
"id": "8AB4ISvpWD",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 3,
"summary": "The paper introduces JOSH (Juxtaposed Outcomes for Simulation Harvesting), a self-alignment framework for large language model (LLM) agents to enhance multi-turn ... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "4;4;3;2",
"confidence_avg": 3.25,
"soundness": "1;2;2;3",
"soundness_avg": 2,
"contribution": "1;2;2;3",
"contribution_avg": 2,
"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": "2026-01-14T22:16:01.286315"
} | {
"id": "3ex7mrmeLr",
"metareview": "This work proposes an LLM self-training approach based on sparse reward simulation. The authors also plan to release their code and data. The reviewers appreciate the general idea of this work, the value of the ToolWOZ benchmark to the community, the performance improvements ach... | {
"decision": "Reject"
} |
DWWwGlPMFr | 2407.18941v1 | LEMoN: Label Error Detection using Multimodal Neighbors | {
"content": "## Abstract\n\nAbstract Large repositories of image-caption pairs are essential for the development of vision-language models. However, these datasets are often extracted from noisy data scraped from the web, and contain many mislabeled examples. In order to improve the reliability of downstream models,... | [
{
"id": "hVzdriIGzM",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper proposes a label error detection method for multimodal datasets. Specifically, the authors first use pre-trained vision-language models to extract imag... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "4;4;3;3",
"confidence_avg": 3.5,
"soundness": "2;3;2;3",
"soundness_avg": 2.5,
"contribution": "2;2;2;2",
"contribution_avg": 2,
"presentation": "3;3;2;3",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.287069"
} | {
"id": "KNU3NJEdQf",
"metareview": "The paper proposes a mechanism for coping with label/caption noise. The reviewers praise the intuitive method and extensive experiments. However they raise concerns about effectiveness (e.g. improvement over using CLIP, results in Table 4), applicability (to real-world noise) an... | {
"decision": "Reject"
} |
DZcmz9wU0i | 2410.09697v1 | Provable Convergence and Limitations of Geometric Tempering for Langevin Dynamics | {
"content": "## Abstract\n\nAbstract Geometric tempering is a popular approach to sampling from challenging multi-modal probability distributions by instead sampling from a sequence of distributions which interpolate, using the geometric mean, between an easier proposal distribution and the target distribution. In t... | [
{
"id": "qjMFBrHHjS",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This work studies the convergence guarantee of geometric tempering for the Langevin diffusion and its time-discretization the Langevin algorithm. The authors prov... | {
"rating": "6;6;8;8",
"rating_avg": 7,
"confidence": "4;3;4;3",
"confidence_avg": 3.5,
"soundness": "3;3;4;4",
"soundness_avg": 3.5,
"contribution": "3;3;3;3",
"contribution_avg": 3,
"presentation": "3;3;4;3",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.288203"
} | {
"id": "mC0uggpsL2",
"metareview": "The authors analyze the convergence of tempered Langevin dynamics. Most interestingly, the authors derive lower bounds where tempering leads to exponentially poor convergence. This result is quite novel as agreed upon by all reviewers. On top of this, the authors also provide th... | {
"decision": "Accept (Poster)"
} |
DaA0wAcTY7 | 2410.16512v1 | TIPS: Text-Image Pretraining with Spatial awareness | {
"content": "## Abstract\n\nAbstract While image-text representation learning has become very popular in recent years, existing models tend to lack spatial awareness and have limited direct applicability for dense understanding tasks.\nFor this reason, self-supervised image-only pretraining is still the go-to method... | [
{
"id": "vnLp2icUJT",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper presents a spatial-aware text-image pre-training method that combines contrastive image-text learning with self-supervised masked image modeling. Besid... | {
"rating": "5;6;6;8",
"rating_avg": 6.25,
"confidence": "4;4;4;3",
"confidence_avg": 3.75,
"soundness": "2;3;3;4",
"soundness_avg": 3,
"contribution": "2;3;3;4",
"contribution_avg": 3,
"presentation": "3;4;3;4",
"presentation_avg": 3.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.289421"
} | {
"id": "0kcWlwTSD1",
"metareview": "The paper presents an approach to integrating spatial awareness into text-image pretraining. The reviewers generally agree on the paper's strengths, including its extensive experiments, strong generalization across tasks, and intuitive dual-embedding technique. Some concerns wer... | {
"decision": "Accept (Poster)"
} |
Daq6Pw3TjN | 2410.05746v1 | Wolf2Pack: The AutoFusion Framework for Dynamic Parameter Fusion | {
"content": "## Abstract\n\nAbstract In the rapidly evolving field of deep learning, specialized models have driven significant advancements in tasks such as computer vision and natural language processing. However, this specialization leads to a fragmented ecosystem where models lack the adaptability for broader ap... | [
{
"id": "ZJi6sW4ttT",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 4,
"summary": "This paper concentrates on a very interesting problem: how to fuse two types of distinct model parameters pretrained for two different tasks into one model that c... | {
"rating": "5;5;6",
"rating_avg": 5.333333333333333,
"confidence": "4;4;3",
"confidence_avg": 3.6666666666666665,
"soundness": "3;2;3",
"soundness_avg": 2.6666666666666665,
"contribution": "2;2;3",
"contribution_avg": 2.3333333333333335,
"presentation": "1;1;4",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.290187"
} | {
"id": "XCB1Zpxuxz",
"metareview": "This paper presents an AutoFusion method for fusing parameters from distinct models for multi-task learning without relying on pre-trained checkpoints. It dynamically permutes model parameters at each layer, optimizing their combination through an unsupervised process.\n\nThis p... | {
"decision": "Reject"
} |
DblHBgD0GR | 2406.18944v4 | Rethinking and Defending Protective Perturbation in Personalized Diffusion Models | {
"content": "## Abstract\n\nAbstract Personalized diffusion models (PDMs) have become prominent for adapting pretrained text-to-image models to generate images of specific subjects using minimal training data. However, PDMs are susceptible to minor adversarial perturbations, leading to significant degradation when f... | [
{
"id": "nUZq6l8XS6",
"initial_rating": 6,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 1,
"summary": "The paper aims to improve the personalization performance of Diffusion Models on images with protective perturbation, a kind of noise avoiding images to be learne... | {
"rating": "3;6;6;8",
"rating_avg": 5.75,
"confidence": "4;3;3;3",
"confidence_avg": 3.25,
"soundness": "2;3;2;3",
"soundness_avg": 2.5,
"contribution": "2;3;2;3",
"contribution_avg": 2.5,
"presentation": "1;3;1;3",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.290948"
} | {
"id": "e2UyAgEZkc",
"metareview": "The paper proposes a view for the fine-tuning process of Personalized Diffusion Models (PDMs) as shortcut learning, motivated by causal analysis. The authors introduce a defense framework to help the model correctly associate images with their original semantic meanings.\n\nStr... | {
"decision": "Reject"
} |
DcZpQhVpp9 | 2411.07496v1 | ADMM for Structured Fractional Minimization | {
"content": "## Abstract\n\nAbstract We consider a class of structured fractional minimization problems, where the numerator includes a differentiable function, a simple nonconvex nonsmooth function, a concave nonsmooth function, and a convex nonsmooth function composed with a linear operator, while the denominator ... | [
{
"id": "KQMf5su4Mf",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "This paper proposes an ADMM for solving a class of structured fractional minimization problems. The main techniques are based on smoothing methods and two well-es... | {
"rating": "5;6;6",
"rating_avg": 5.666666666666667,
"confidence": "4;4;2",
"confidence_avg": 3.3333333333333335,
"soundness": "3;3;3",
"soundness_avg": 3,
"contribution": "3;2;3",
"contribution_avg": 2.6666666666666665,
"presentation": "2;1;2",
"presentation_avg": 1.6666666666666667
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.292038"
} | {
"id": "MTH3gruw4R",
"metareview": "The paper presents a novel ADMM-based optimization method, tailored for structured fractional minimization problems, which seem not to be well addressed in the literature. The main techniques are based on smoothing methods (via the Moreau envelope) and two well-established appro... | {
"decision": "Accept (Poster)"
} |
DdPeCRVyCd | 2409.12371v1 | Communication-Efficient Federated Low-Rank Update Algorithm and its Connection to Implicit Regularization | {
"content": "## Abstract\n\nAbstract Federated Learning (FL) faces significant challenges related to communication efficiency and heterogeneity. To address these issues, we explore the potential of using low-rank updates. Our theoretical analysis reveals that client’s loss exhibits a higher rank structure (gradients... | [
{
"id": "ZmWwRlv7NC",
"initial_rating": 3,
"confidence": 3,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The paper applies FedLoRU and its variants to impose the local update in a low-rank subspace to achieve implicit regularization.",
"strengths": "FedLoRU uses ... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "3;3;3;3",
"confidence_avg": 3,
"soundness": "2;2;2;2",
"soundness_avg": 2,
"contribution": "2;2;2;3",
"contribution_avg": 2.25,
"presentation": "2;2;3;2",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.293415"
} | {
"id": "LUQ3dCnvbs",
"metareview": "The paper proposes the FedLoRU method: a federated optimization method combining low-rank updates on the clients' side (equivalent to running LORA locally on each client), aggregating the updates on the servers, and repeating this process. The authors claim this method is motiva... | {
"decision": "Reject"
} |
DeVm3YUnpj | 2410.10934v2 | Agent-as-a-Judge: Evaluating Agents with Agents | {
"content": "## Abstract\n\nAbstract Contemporary evaluation techniques are inadequate for agentic systems. These approaches either focus exclusively on final outcomes—ignoring the step-by-step nature of agentic systems, or require excessive manual labour. To address this, we introduce the Agent-as-a-Judge framework... | [
{
"id": "QH3I9STjfx",
"initial_rating": 5,
"confidence": 3,
"soundness": 2,
"contribution": 3,
"presentation": 1,
"summary": "This paper introduces the Agent-as-a-Judge framework, an innovative approach for evaluating agentic systems by employing other agentic systems as evaluators. Buil... | {
"rating": "5;5;6",
"rating_avg": 5.333333333333333,
"confidence": "3;3;4",
"confidence_avg": 3.3333333333333335,
"soundness": "2;2;3",
"soundness_avg": 2.3333333333333335,
"contribution": "2;3;3",
"contribution_avg": 2.6666666666666665,
"presentation": "2;1;3",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.294384"
} | {
"id": "96lHd1jn6f",
"metareview": "The paper \"Agent-as-a-Judge: Evaluating Agents with Agents\" introduces a novel framework, Agent-as-a-Judge (AAAJ), for evaluating agentic systems (AI systems capable of autonomous decision-making and action) by leveraging other agentic systems as evaluators. This approach exte... | {
"decision": "Reject"
} |
Dem5LyVk8R | 2410.05655v1 | Efficient Policy Evaluation with Safety Constraint for Reinforcement Learning | {
"content": "## Abstract\n\nAbstract In reinforcement learning, classic on-policy evaluation methods often suffer from high variance and require massive online data to attain the desired accuracy. Previous studies attempt to reduce evaluation variance by searching for or designing proper behavior policies to collect... | [
{
"id": "g9hBUJ8GCI",
"initial_rating": 8,
"confidence": 2,
"soundness": 4,
"contribution": 4,
"presentation": 4,
"summary": "This paper provides an on-policy evaluation method that aims to reduce evaluation variance while also ensuring safety. This is done in the context of contextual b... | {
"rating": "5;5;5;8",
"rating_avg": 5.75,
"confidence": "4;4;3;2",
"confidence_avg": 3.25,
"soundness": "2;3;3;4",
"soundness_avg": 3,
"contribution": "2;2;2;4",
"contribution_avg": 2.5,
"presentation": "3;2;3;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.295282"
} | {
"id": "bu5WmIWoYn",
"metareview": "The paper provides an on-policy evaluation method that aims to reduce the variance while satisfying safety constraints.\nThe problem is an important one and the reviewers are generally positive about the paper. The experiments are also well executed, and I encourage the authors ... | {
"decision": "Accept (Poster)"
} |
DfTWrTwLzD | 2410.19318v1 | Two Are Better than One: Context Window Extension with Multi-Grained Self-Injection | {
"content": "## Abstract\n\nAbstract The limited context window of contemporary large language models (LLMs) remains a huge barrier to their broader application across various domains.\nWhile continual pre-training on long-context data is a straightforward and effective solution, it incurs substantial costs in terms... | [
{
"id": "n5ifuAKTBL",
"initial_rating": 5,
"confidence": 5,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper introduces SharedLLM, a novel approach for extending the context window of large language models (LLMs) by using a hierarchical architecture that pairs ... | {
"rating": "3;5;5;6",
"rating_avg": 4.75,
"confidence": "5;3;5;3",
"confidence_avg": 4,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "2;3;3;2",
"contribution_avg": 2.5,
"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": "2026-01-14T22:16:01.296013"
} | {
"id": "dqTjT6V2Rw",
"metareview": "This paper proposes a method for longer context length using two LMs, which play role of a compressor and a decoder each other. \n\nLonger context length without heavy computational costs and data for post-training are critical and important topic. \nHowever, main concerns are t... | {
"decision": "Reject"
} |
DhHIw9Nbl1 | 2410.02309v2 | Decoupling Layout from Glyph in Online Chinese Handwriting Generation | {
"content": "## Abstract\n\nAbstract Text plays a crucial role in the transmission of human civilization, and teaching machines to generate online handwritten text in various styles presents an interesting and significant challenge. However, most prior work has concentrated on generating individual Chinese fonts, le... | [
{
"id": "10rD81Fq00",
"initial_rating": 6,
"confidence": 4,
"soundness": 4,
"contribution": 3,
"presentation": 3,
"summary": "The paper introduces a novel approach for generating online handwritten Chinese text with specific styles. The authors naturally divide a text line into two compo... | {
"rating": "3;5;6;6",
"rating_avg": 5,
"confidence": "5;3;5;4",
"confidence_avg": 4.25,
"soundness": "2;3;4;4",
"soundness_avg": 3.25,
"contribution": "2;3;4;3",
"contribution_avg": 3,
"presentation": "2;2;3;3",
"presentation_avg": 2.5
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.296635"
} | {
"id": "LtRvc3FqPF",
"metareview": "This paper studies online handwritten Chinese text generation. After rebuttal, the overall rating of this paper is above the marginal acceptance threshold, but with mixed scores of 5,6,6,8. The Area Chair has read the paper, all reviews, and the authors' rebuttal. The main stren... | {
"decision": "Accept (Poster)"
} |
DhdqML3FdM | 2405.16674v2 | Limits of Deep Learning: Sequence Modeling through the Lens of Complexity Theory | {
"content": "## Abstract\n\nAbstract Deep learning models have achieved significant success across various applications but continue to struggle with tasks requiring complex reasoning over sequences, such as function composition and compositional tasks. Despite advancements, models like Structured State Space Models... | [
{
"id": "hFXwPsE9qt",
"initial_rating": 6,
"confidence": 4,
"soundness": 2,
"contribution": 3,
"presentation": 3,
"summary": "The paper theoretically and empirically studies the limitations of the computational power of SSMs in terms of effective space complexity and their ability to do ... | {
"rating": "3;5;8;8",
"rating_avg": 6,
"confidence": "3;3;3;3",
"confidence_avg": 3,
"soundness": "1;3;3;4",
"soundness_avg": 2.75,
"contribution": "3;3;3;4",
"contribution_avg": 3.25,
"presentation": "3;3;3;4",
"presentation_avg": 3.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.297317"
} | {
"id": "dzns25VJ3U",
"metareview": "This paper empirically and theoretically analyzes the computational limitations of Structured State Space Models (SSMs). The authors introduce three theorems: the inability of SSMs to compose functions, exponential scaling of compute when performing chain-of-thought, and an inab... | {
"decision": "Accept (Poster)"
} |
DhlbK7tAjz | 2407.20034v1 | MaskInversion: Localized Embeddings via Optimization of Explainability Maps | {
"content": "## Abstract\n\nAbstract Vision-language foundation models such as CLIP have achieved tremendous results in global vision-language alignment, but still show some limitations in creating representations for specific image regions. To address this problem, we propose MaskInversion, a method that leverages ... | [
{
"id": "rDQ81akkSW",
"initial_rating": 3,
"confidence": 5,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper introduces MaskInversion, a method that leverages pre-trained vision-language models (such as CLIP) to generate context-aware embeddings for specific i... | {
"rating": "3;6;6",
"rating_avg": 5,
"confidence": "5;3;4",
"confidence_avg": 4,
"soundness": "2;3;3",
"soundness_avg": 2.6666666666666665,
"contribution": "2;2;3",
"contribution_avg": 2.3333333333333335,
"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": "2026-01-14T22:16:01.297974"
} | {
"id": "MaeD7Pfsbb",
"metareview": "This paper addresses the poor localization problem of contrastive image-text pretraining (CLIP) models, which is a critical issue when using CLIP models in practice. The paper is well-written and motivated, and experiments show improved performance on the target localization tas... | {
"decision": "Reject"
} |
DjtJV3ke1j | 2211.14825v1 | Dynamic Kernel Sparsifiers | {
"content": "## Abstract\n\nA geometric graph associated with a set of points P = { x 1 , x 2 , ⋯ , x n } ⊂ ℝ d 𝑃 subscript 𝑥 1 subscript 𝑥 2 ⋯ subscript 𝑥 𝑛 superscript ℝ 𝑑 P=\\{x_{1},x_{2},\\cdots,x_{n}\\}\\subset\\mathbb{R}^{d} and a fixed kernel function 𝖪 : ℝ d × ℝ d → ℝ ≥ 0 : 𝖪 → superscript ℝ 𝑑 super... | [
{
"id": "9hBPKWCyu2",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "Given a set of $n$ points and a kernel function $k$, we can consider an all-pairs weighted graph where the weight of every pair is given by the corresponding kern... | {
"rating": "3;3;6;6",
"rating_avg": 4.5,
"confidence": "4;4;4;3",
"confidence_avg": 3.75,
"soundness": "1;2;3;3",
"soundness_avg": 2.25,
"contribution": "2;2;3;3",
"contribution_avg": 2.5,
"presentation": "2;2;1;3",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.299042"
} | {
"id": "zT5ckBrAPt",
"metareview": "The goal of this work is to efficiently update the kernel matrix when one of the data sample changes its representation. The main idea is to first use random projection to reduce the dimension while pair-wise distances are approximately maintained as per Johnson-Lindenstrauss, a... | {
"decision": "Reject"
} |
DkzZ1ooc7q | 2410.21269v1 | OmniSep: Unified Omni-Modality Sound Separation with Query-Mixup | {
"content": "## Abstract\n\nAbstract The scaling up has brought tremendous success in the fields of vision and language in recent years. When it comes to audio, however, researchers encounter a major challenge in scaling up the training data, as most natural audio contains diverse interfering signals. To address thi... | [
{
"id": "UjEMNhOqXg",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper presents OmniSep, a framework for omni-modal sound separation, which supports sound isolation using queries from multiple modalities, such as text, ima... | {
"rating": "5;5;5;6",
"rating_avg": 5.25,
"confidence": "4;3;3;3",
"confidence_avg": 3.25,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"contribution": "3;2;2;3",
"contribution_avg": 2.5,
"presentation": "3;2;1;3",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.300067"
} | {
"id": "eKY0cfUrkR",
"metareview": "This paper presents a novel system for query-based sound separation, called OmniSep, that accommodates multiple query modalities (text, image, and audio), either independently or in combination. Technical contributions consist in Query-Mixup for simultaneous multi-modal query pr... | {
"decision": "Accept (Poster)"
} |
DlqRpj68xe | 2410.01458v1 | From Reward Shaping to Q-Shaping: Achieving Unbiased Learning with LLM-Guided Knowledge | {
"content": "## Abstract\n\nAbstract Q-shaping is an extension of Q-value initialization and serves as an alternative to reward shaping for incorporating domain knowledge to accelerate agent training, thereby improving sample efficiency by directly shaping Q-values. This approach is both general and robust across di... | [
{
"id": "Swg5m4q0xm",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper introduces a novel framework called \"Q-shaping,\" which enhances Q-value initialization by integrating domain knowledge to accelerate training in rein... | {
"rating": "3;3;5",
"rating_avg": 3.6666666666666665,
"confidence": "3;4;3",
"confidence_avg": 3.3333333333333335,
"soundness": "2;2;2",
"soundness_avg": 2,
"contribution": "2;2;2",
"contribution_avg": 2,
"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": "2026-01-14T22:16:01.300707"
} | {
"id": "qtbbNPnaOU",
"metareview": "Authors present Q-shaping, an alternative to reward shaping using LLM guidance that directly modifies Q values. There are thorough empirical results showing an improvement in performance compared with existing reward-shaping methods. \n\nReviewers thought the Q-shaping contribut... | {
"decision": "Reject"
} |
DnfPX10Etk | 2410.11086v2 | JOOCI: A FRAMEWORK FOR LEARNING COMPREHENSIVE SPEECH REPRESENTATIONS | {
"content": "## Abstract\n\nAbstract Information in speech can be divided into two categories: “what is being said” (content) and “how it is expressed” (other). Current state-of-the-art (SOTA) techniques model speech at fixed segments, usually 10-25 ms, using a single embedding. Given the orthogonal nature of other ... | [
{
"id": "HCrx5mWaO8",
"initial_rating": 3,
"confidence": 5,
"soundness": 2,
"contribution": 1,
"presentation": 1,
"summary": "This paper proposes a method for speech representation learning, particularly, for disentangle content information from non-content information. The paper reporte... | {
"rating": "3;3;3;5",
"rating_avg": 3.5,
"confidence": "4;4;5;4",
"confidence_avg": 4.25,
"soundness": "2;2;2;3",
"soundness_avg": 2.25,
"contribution": "2;2;1;2",
"contribution_avg": 1.75,
"presentation": "3;2;1;3",
"presentation_avg": 2.25
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.301291"
} | {
"id": "H9uFPtJIGD",
"metareview": "The paper proposes an approach to disentangle linguistic information from non-linguistic ones by using an additional speaker model for supervision.\n\nI recommend a rejection because of the unsupported claims in the paper.\n\nAll reviewers raised concerns about unsupported clai... | {
"decision": "Reject"
} |
Do3whenqeY | 2410.03868v1 | Can Language Models Reason about Individualistic Human Values and Preferences? | {
"content": "## Abstract\n\nAbstract Recent calls for pluralistic alignment emphasize that AI systems should address the diverse needs of all people. Yet, efforts in this space often require sorting people into fixed buckets of pre-specified diversity-defining dimensions (e.g., demographics, personalities, communica... | [
{
"id": "XJeHtYUnbN",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 2,
"summary": "This paper introduces a dataset from the World Values Survey designed to evaluate language models' (LMs) reasoning on individualistic values. Unlike pluralistic a... | {
"rating": "3;3;6;6",
"rating_avg": 4.5,
"confidence": "2;4;4;3",
"confidence_avg": 3.25,
"soundness": "2;2;3;3",
"soundness_avg": 2.5,
"contribution": "2;2;3;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": "2026-01-14T22:16:01.302268"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
DoDNJdDntB | 2410.22573v1 | Flow Matching for Posterior Inference with Simulator Feedback | {
"content": "## Abstract\n\nAbstract Flow-based generative modeling is a powerful tool for solving inverse problems in physical sciences that can be used for sampling and likelihood evaluation with much lower inference times than traditional methods.\nWe propose to refine flows with additional control signals based ... | [
{
"id": "YckSgdrp41",
"initial_rating": 3,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 2,
"summary": "The paper studies the problem of modelling the posterior $p(\\theta\\mid x)$ in a generative model $\\theta\\to x$, where $p(x\\mid\\theta)$ is available as a sim... | {
"rating": "3;3;3;3;6",
"rating_avg": 3.6,
"confidence": "2;5;3;4;4",
"confidence_avg": 3.6,
"soundness": "2;1;1;3;3",
"soundness_avg": 2,
"contribution": "2;1;2;2;3",
"contribution_avg": 2,
"presentation": "2;2;2;2;3",
"presentation_avg": 2.2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.303202"
} | {
"id": "XsX39112Nw",
"metareview": "The reviewers are somewhat divided (3-3-3-6-6) about the paper, but they overall lean towards rejection. The paper introduces simulator feedback as an extension to flow matching for simulation-based inference. The approach is well-motivated, but the results are not convincing. T... | {
"decision": "Reject"
} |
Dq9VrVuLzV | 2410.00337v1 | SyntheOcc: Synthesize Geometric-Controlled Street View Images through 3D Semantic MPIs | {
"content": "## Abstract\n\nAbstract The advancement of autonomous driving is increasingly reliant on high-quality annotated datasets, especially in the task of 3D occupancy prediction, where the occupancy labels require dense 3D annotation with significant human effort. In this paper, we propose SytheOcc , which de... | [
{
"id": "3pB1AfdZX0",
"initial_rating": 5,
"confidence": 5,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "This paper presents SyntheOcc, a method for generating multi-camera images and videos of driving scenarios, using occupancy and text prompt as guiding inputs. The... | {
"rating": "5;5;5;5",
"rating_avg": 5,
"confidence": "5;4;5;5",
"confidence_avg": 4.75,
"soundness": "2;3;3;2",
"soundness_avg": 2.5,
"contribution": "2;3;2;2",
"contribution_avg": 2.25,
"presentation": "2;3;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": "2026-01-14T22:16:01.304043"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
DtFCIfvAFc | 2410.01404v1 | Gaussian-Det: Learning Closed-Surface Gaussians for 3D Object Detection | {
"content": "## Abstract\n\nAbstract Skins wrapping around our bodies, leathers covering over the sofa, sheet metal coating the car – it suggests that objects are enclosed by a series of continuous surfaces, which provides us with informative geometry prior for objectness deduction.\nIn this paper, we propose Gaussi... | [
{
"id": "eylfgIWNck",
"initial_rating": 3,
"confidence": 5,
"soundness": 2,
"contribution": 1,
"presentation": 1,
"summary": "This paper proposed a 3D object detection method based on the Gaussian splatting. With the Gaussian splatting as input, the proposed deep model extracts the 3D ob... | {
"rating": "3;5;6;6",
"rating_avg": 5,
"confidence": "5;3;3;4",
"confidence_avg": 3.75,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "1;2;3;3",
"contribution_avg": 2.25,
"presentation": "1;1;2;3",
"presentation_avg": 1.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.304792"
} | {
"id": "Ouu0sQKshJ",
"metareview": "The proposed method, Gaussian-Det, uses Gaussian Splatting (3DGS) to represent the surfaces for multiview 3D object detection. Gaussian-Det formulates the input Gaussians as feature descriptors on partial surfaces. Outliers derived from 3DGS are reduced by the Closure Inferring ... | {
"decision": "Accept (Poster)"
} |
DwiwOcK1B7 | 2409.18850v1 | Two Sparse Matrices are Better than One: Sparsifying Neural Networks with Double Sparse Factorization | {
"content": "## Abstract\n\nAbstract Neural networks are often challenging to work with due to their large size and complexity. To address this, various methods aim to reduce model size by sparsifying or decomposing weight matrices, such as magnitude pruning and low-rank or block-diagonal factorization. In this work... | [
{
"id": "hT1i54iZqw",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper proposes Double Sparse Factorization (DSF) of the weight matrices to prune them effectively. They formulate it as an alternating optimization and optimi... | {
"rating": "3;6;6",
"rating_avg": 5,
"confidence": "4;4;4",
"confidence_avg": 4,
"soundness": "2;3;3",
"soundness_avg": 2.6666666666666665,
"contribution": "2;2;3",
"contribution_avg": 2.3333333333333335,
"presentation": "2;3;3",
"presentation_avg": 2.6666666666666665
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.305383"
} | {
"id": "okhaGCHOfR",
"metareview": "The authors propose a method for sparsifying neural network parameter matrices by reparameterizing them as the product of two sparse matrices. This is accomplished via a heuristic that seeks to minimize the error of the factorized approximation relative to the original weights ... | {
"decision": "Accept (Poster)"
} |
DwqoBkj2Mw | 2410.06186v2 | The Last Iterate Advantage: Empirical Auditing and Principled Heuristic Analysis of Differentially Private SGD | {
"content": "## Abstract\n\nAbstract We propose a simple heuristic privacy analysis of noisy clipped stochastic gradient descent (DP-SGD) in the setting where only the last iterate is released and the intermediate iterates remain hidden. Namely, our heuristic assumes a linear structure for the model. We show experim... | [
{
"id": "tdKSpJGz0s",
"initial_rating": 5,
"confidence": 5,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The paper introduces a heuristic privacy analysis for DP-SGD when only the final model is released, and intermediate updates are hidden. This heuristic assumes li... | {
"rating": "5;5;5;8",
"rating_avg": 5.75,
"confidence": "4;5;5;4",
"confidence_avg": 4.5,
"soundness": "3;2;3;4",
"soundness_avg": 3,
"contribution": "2;2;2;3",
"contribution_avg": 2.25,
"presentation": "2;3;3;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.305976"
} | {
"id": "6YiC7Lqi8B",
"metareview": "The paper investigates the privacy of DP-SGD when only the last iterate is provided (and intermediate model states are hidden). It shows a heuristic analysis for linear functions which is found to be predictive of the privacy obtained via auditing on various training procedures.... | {
"decision": "Accept (Poster)"
} |
DxT3e2f1jc | 2406.16260v1 | Video-Infinity: Distributed Long Video Generation | {
"content": "## Abstract\n\nAbstract Diffusion models have recently\nachieved remarkable results\nfor video generation.\nDespite the encouraging performances,\nthe generated videos are typically\nconstrained to a small number of frames,\nresulting in clips lasting merely a few seconds.\nThe primary challenges in pro... | [
{
"id": "ZaQU1OFbVk",
"initial_rating": 5,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The paper proposes Video-Infinity, a distributed inference pipeline designed for long-form video generation using diffusion models. The framework leverages two ma... | {
"rating": "3;3;5;6",
"rating_avg": 4.25,
"confidence": "4;5;3;4",
"confidence_avg": 4,
"soundness": "2;3;3;3",
"soundness_avg": 2.75,
"contribution": "1;2;2;3",
"contribution_avg": 2,
"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": "2026-01-14T22:16:01.306641"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
DydCqKa6AH | 2410.07500v1 | Learning to Generate Diverse Pedestrian Movements from Web Videos with Noisy Labels | {
"content": "## Abstract\n\nAbstract Understanding and modeling pedestrian movements in the real world is crucial for applications like motion forecasting and scene simulation. Many factors influence pedestrian movements, such as scene context, individual characteristics, and goals, which are often ignored by the ex... | [
{
"id": "gT56JPCyRS",
"initial_rating": 6,
"confidence": 4,
"soundness": 2,
"contribution": 3,
"presentation": 2,
"summary": "This paper introduces a context-aware generative model for realistic pedestrian movement prediction. It leverages a conditional diffusion framework that uses 3D p... | {
"rating": "5;5;6;8",
"rating_avg": 6,
"confidence": "4;4;4;3",
"confidence_avg": 3.75,
"soundness": "3;4;2;4",
"soundness_avg": 3.25,
"contribution": "2;3;3;3",
"contribution_avg": 2.75,
"presentation": "2;3;2;4",
"presentation_avg": 2.75
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.307303"
} | {
"id": "SwU5KPQdLl",
"metareview": "This paper introduces a context-aware generative model for realistic pedestrian movement prediction. This paper proposes a valuable idea, supported by clear visualizations and thorough ablation studies. The results demonstrate performance improvements over baselines, and compreh... | {
"decision": "Accept (Poster)"
} |
Dyo2tS5A8b | 2403.02580v1 | What do we learn from inverting CLIP models? | {
"content": "## Abstract\n\nAbstract We employ an inversion-based approach to examine CLIP models. Our examination reveals that inverting CLIP models results in the generation of images that exhibit semantic alignment with the specified target prompts. We leverage these inverted images to gain insights into various ... | [
{
"id": "r7gvHDARw9",
"initial_rating": 3,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The paper investigates CLIP models using image inversion techniques to analyze their learned representations and biases. Their analysis reveals three findings: (a... | {
"rating": "3;3;5;6",
"rating_avg": 4.25,
"confidence": "4;4;3;4",
"confidence_avg": 3.75,
"soundness": "1;2;3;3",
"soundness_avg": 2.25,
"contribution": "2;2;2;3",
"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": "2026-01-14T22:16:01.307915"
} | {
"id": "",
"metareview": "",
"additional_comments": ""
} | {
"decision": ""
} |
DyyLUUVXJ5 | 2411.02397v2 | Adaptive Caching for Faster Video Generation with Diffusion Transformers | {
"content": "## Abstract\n\nAbstract Generating temporally-consistent high-fidelity videos can be computationally expensive, especially over longer temporal spans. More-recent Diffusion Transformers (DiTs)— despite making significant headway in this context— have only heightened such challenges as they rely on large... | [
{
"id": "FYO6KrCulY",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 2,
"presentation": 2,
"summary": "The paper proposes training-free Diffusion Transformer (DiT) acceleration named AdaCache. The method is motivated by the fact that different videos require differ... | {
"rating": "5;5;6;8",
"rating_avg": 6,
"confidence": "5;4;4;5",
"confidence_avg": 4.5,
"soundness": "4;2;2;4",
"soundness_avg": 3,
"contribution": "4;2;3;3",
"contribution_avg": 3,
"presentation": "4;2;2;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.308597"
} | {
"id": "aFrdQygiM3",
"metareview": "The paper introduces a novel adaptive algorithm but has several weaknesses. The choice of MSE as a metric is questioned, and the reviewer asks whether alternative metrics, like cosine similarity, would be more suitable. The compatibility with large T2I models like FLUX is unclea... | {
"decision": "Reject"
} |
E040QmNETN | 2410.12957v1 | MuVi: Video-to-Music Generation with Semantic Alignment and Rhythmic Synchronization | {
"content": "## Abstract\n\nAbstract Generating music that aligns with the visual content of a video has been a challenging task, as it requires a deep understanding of visual semantics and involves generating music whose melody, rhythm, and dynamics harmonize with the visual narratives. This paper presents MuVi, a ... | [
{
"id": "JGn5A2TtbV",
"initial_rating": 6,
"confidence": 3,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "The paper proposed a method for video to music generation. There exist not many previous works that utilizes the concept of (video)sequence-to-(music)sequence gen... | {
"rating": "3;6;6",
"rating_avg": 5,
"confidence": "5;3;3",
"confidence_avg": 3.6666666666666665,
"soundness": "2;3;3",
"soundness_avg": 2.6666666666666665,
"contribution": "2;3;2",
"contribution_avg": 2.3333333333333335,
"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": "2026-01-14T22:16:01.309428"
} | {
"id": "e1mlsogTA5",
"metareview": "This paper, MuVi, proposes a new video-to-music generation framework focusing on two primary goals: semantic alignment (music content changing in pair with the video’s mood or scene) and rhythmic synchronization (musical beats matching a video’s pacing). \n\n## Reviewers’ Feedba... | {
"decision": "Reject"
} |
E0dTlxy1T4 | 2409.05840v3 | MMEvol: Empowering Multimodal Large Language Models with Evol-Instruct | {
"content": "## Abstract\n\nAbstract The development of Multimodal Large Language Models (MLLMs) has seen significant advancements with increasing demands in various fields (e.g., multimodal agents, embodied intelligence). While model-driven approaches attempt to enhance MLLMs capabilities through diverse architectu... | [
{
"id": "RhC2eDRqf9",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper introduce a method to systematically evolve seed instruction fine-tuning data for VLM to larger scale and enhance vision centric capabilities for fine-... | {
"rating": "5;5;5;6;6",
"rating_avg": 5.4,
"confidence": "4;4;4;5;4",
"confidence_avg": 4.2,
"soundness": "3;2;3;3;3",
"soundness_avg": 2.8,
"contribution": "2;2;3;3;3",
"contribution_avg": 2.6,
"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": "2026-01-14T22:16:01.310164"
} | {
"id": "Rypbd8nA3u",
"metareview": "This paper introduces a framework to improve multimodal models by evolving image-text instruction data. It shows some performance improvements with less data. The approach is interesting and shows promising results. However, concerns were raised about the comparison with other m... | {
"decision": "Reject"
} |
E1EHO0imOb | 2409.12517v1 | Scaling FP8 training to trillion-token LLMs | {
"content": "## Abstract\n\nAbstract We train, for the first time, large language models using FP8 precision on datasets up to 2 trillion tokens — a 20-fold increase over previous limits. Through these extended training runs, we uncover critical instabilities in FP8 training that were not observable in earlier works... | [
{
"id": "3JogXSWYQe",
"initial_rating": 8,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "The paper presents new findings and addresses key challenges in large-scale FP8 training for LLMs on modern hardware that supports FP8 operations. The authors ide... | {
"rating": "5;6;8;8",
"rating_avg": 6.75,
"confidence": "3;3;2;4",
"confidence_avg": 3,
"soundness": "3;3;3;3",
"soundness_avg": 3,
"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 Spotlight",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.310944"
} | {
"id": "6VUtIu3dLe",
"metareview": "All reviewers agreed this paper should be accepted: it addresses an important problem, the method is thoughtfully-designed, and the paper is clearly written. A clear accept. Authors: you've already indicated that you've updated the submission to respond to reviewer changes, if y... | {
"decision": "Accept (Spotlight)"
} |
E1ML0nEReb | 2410.21211v1 | Exploring contextual modeling with linear complexity for point cloud segmentation | {
"content": "## Abstract\n\nAbstract Point cloud segmentation is an important topic in 3D understanding. Traditionally, this task has been tackled using either the CNN or Transformer. Recently, a newcomer, Mamba, has emerged as a promising alternative, offering efficient long-range contextual modeling capabilities w... | [
{
"id": "m11hXJnMtV",
"initial_rating": 8,
"confidence": 5,
"soundness": 3,
"contribution": 3,
"presentation": 4,
"summary": "This paper focuses on a simple yet practical target in 3D understanding: how to enhance the accuracy of a Mamba-based framework while preserving its efficiency ro... | {
"rating": "1;5;5;6;8",
"rating_avg": 5,
"confidence": "5;4;4;4;5",
"confidence_avg": 4.4,
"soundness": "1;3;3;3;3",
"soundness_avg": 2.6,
"contribution": "1;3;3;3;3",
"contribution_avg": 2.6,
"presentation": "1;3;3;3;4",
"presentation_avg": 2.8
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.311887"
} | {
"id": "E5OmJT5hxS",
"metareview": "The paper receives 4 positive and 1 negative rating after rebuttal. Although the paper has some merits like competitive results with faster runtime and lower memory cost, the reviewers pointed out a few critical concerns about 1) technical contributions compared to other Mamba-b... | {
"decision": "Reject"
} |
E1N1oxd63b | 2406.02540v2 | ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation | {
"content": "## Abstract\n\nAbstract Diffusion transformers have exhibited remarkable performance in visual generation tasks, such as generating realistic images or videos based on textual instructions. However, larger model sizes and multi-frame processing for video generation lead to increased computational and me... | [
{
"id": "oqszlQVvSm",
"initial_rating": 6,
"confidence": 2,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper introduces ViDiT-Q, a novel quantization method designed to address the unique challenges faced by diffusion transformers (DiTs) in text-to-image and v... | {
"rating": "5;6;6",
"rating_avg": 5.666666666666667,
"confidence": "5;2;2",
"confidence_avg": 3,
"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": "2026-01-14T22:16:01.312629"
} | {
"id": "fUmxJ3r6gX",
"metareview": "Summary:\n\nThe paper presents a post-quantization method for Diffusion Transformer models for image and video generation. It provides a detailed analysis of the source of degradation and proposes a tailored quantization method, achieving W8A8 and W4A8 with negligible degradatio... | {
"decision": "Accept (Poster)"
} |
E1Tr7wTlIt | 2312.04920v2 | $\lambda$-SecAgg: Partial Vector Freezing for Lightweight Secure Aggregation in Federated Learning | {
"content": "## Abstract\n\nAbstract Secure aggregation of user vectors has become a critical issue in the field of federated learning. Many Secure Aggregation Protocols (SAP) face exorbitant computation costs, which severely limit their applicability.\nWe uncover that current endeavors to reduce computation costs t... | [
{
"id": "r3lBZOFK1o",
"initial_rating": 1,
"confidence": 4,
"soundness": 1,
"contribution": 2,
"presentation": 1,
"summary": "This paper introduces $\\lambda$-SecAgg, a secure aggregation protocol for federated learning (FL) designed to reduce computational and communication overhead thr... | {
"rating": "1;1;3;5;5;6",
"rating_avg": 3.5,
"confidence": "4;4;4;2;3;3",
"confidence_avg": 3.3333333333333335,
"soundness": "1;1;2;3;2;2",
"soundness_avg": 1.8333333333333333,
"contribution": "1;2;2;2;2;3",
"contribution_avg": 2,
"presentation": "2;1;3;2;2;2",
"presentation_avg": 2
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.313367"
} | {
"id": "7Bq7WRqDhB",
"metareview": "The paper proposes $\\lambda$-SecAgg, a secure aggregation protocol for federated learning (FL), to reduce computational and communication overhead using Partial Vector Freezing (PVF). However, most reviewers raised their concerns that PVF significantly reduces this privacy. Des... | {
"decision": "Reject"
} |
E2RyjrBMVZ | 2406.10229v1 | Quantifying Variance in Evaluation Benchmarks | {
"content": "## Abstract\n\nAbstract Evaluation benchmarks are the cornerstone of measuring capabilities of large language models (LLMs), as well as driving progress in said capabilities.\nOriginally designed to make claims about capabilities (or lack thereof) in fully pretrained models, evaluation benchmarks are no... | [
{
"id": "Vacn7oIcoH",
"initial_rating": 3,
"confidence": 3,
"soundness": 3,
"contribution": 1,
"presentation": 3,
"summary": "This work aims to quantify evaluation benchmark variance across a range of settings (from pretraining intermediate checkpoints, to the largest frontier LLMs) usin... | {
"rating": "3;3;3;5;5;6",
"rating_avg": 4.166666666666667,
"confidence": "4;2;3;3;3;3",
"confidence_avg": 3,
"soundness": "3;1;3;3;2;2",
"soundness_avg": 2.3333333333333335,
"contribution": "2;3;1;2;2;3",
"contribution_avg": 2.1666666666666665,
"presentation": "2;2;3;3;2;3",
"presentation_avg": 2.5... | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.314540"
} | {
"id": "ZtfNcu90Ge",
"metareview": "This work investigates the variance in LLM evaluation benchmarks, primarily looking at sources of variance that manifest during training rather than at test time. To this end, a set of 7B Llama-style models are trained and experiments are conducted to determine how the conclusio... | {
"decision": "Reject"
} |
E3LDsbUSRZ | 2406.09923v2 | CliBench: A Multifaceted and Multigranular Evaluation of Large Language Models for Clinical Decision Making | {
"content": "## Abstract\n\nAbstract The integration of Artificial Intelligence (AI), especially Large Language Models (LLMs), into the clinical diagnosis process offers significant potential to improve the efficiency and accessibility of medical care. While LLMs have shown some promise in the medical domain, their ... | [
{
"id": "2JhFHcDC3H",
"initial_rating": 5,
"confidence": 4,
"soundness": 2,
"contribution": 3,
"presentation": 3,
"summary": "This paper presents a new benchmark CliBench developed from the MIMIC IV dataset, offering a comprehensive and realistic assessment of LLMs’ capabilities in clini... | {
"rating": "3;3;5;5;6",
"rating_avg": 4.4,
"confidence": "3;4;4;4;4",
"confidence_avg": 3.8,
"soundness": "2;1;2;2;2",
"soundness_avg": 1.8,
"contribution": "1;2;3;3;2",
"contribution_avg": 2.2,
"presentation": "3;2;3;3;3",
"presentation_avg": 2.8
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.315475"
} | {
"id": "MNTmJdOX2t",
"metareview": "The paper introduces CliBench, a benchmark using MIMIC-IV to evaluate LLMs in clinical decision-making tasks, providing insights into their potential, limitations, and suitability for real-world healthcare applications.\n\nStrengths\n- CliBench introduces a benchmark for evaluat... | {
"decision": "Reject"
} |
E3PgLQzPob | 2408.16766v2 | CSGO: Content-Style Composition in Text-to-Image Generation | {
"content": "## Abstract\n\nAbstract The diffusion model has shown exceptional capabilities in controlled image generation, which has further fueled interest in image style transfer.\nExisting works mainly focus on training free-based methods (e.g., image inversion) due to the scarcity of specific data.\nIn this stu... | [
{
"id": "2JyVEXZFfJ",
"initial_rating": 5,
"confidence": 5,
"soundness": 1,
"contribution": 3,
"presentation": 3,
"summary": "This paper was well written with clear structure and easy to understand. The paper firstly proposes a high quality and carefully cleaned dataset with 210k Content... | {
"rating": "3;5;5;6;6",
"rating_avg": 5,
"confidence": "5;5;5;4;4",
"confidence_avg": 4.6,
"soundness": "2;3;1;2;3",
"soundness_avg": 2.2,
"contribution": "3;2;3;2;3",
"contribution_avg": 2.6,
"presentation": "2;3;3;2;3",
"presentation_avg": 2.6
} | {
"primary_area": "",
"track": "main",
"venue": "Submitted to ICLR 2025",
"venueid": "ICLR.cc/2025/Conference/Rejected_Submission",
"processed_at": "2026-01-14T22:16:01.316237"
} | {
"id": "IoLh6KM7g6",
"metareview": "This work introduced a style encoder and an image encoder for style transfer task by using feature injection through selected layers, and collected a new 210K dataset for training such a model. While reviewers appreciate good results and the effort of dataset collection, there a... | {
"decision": "Reject"
} |
E4LAVLXAHW | 2405.20777v2 | Black-Box Detection of Language Model Watermarks | {
"content": "## Abstract\n\nAbstract Watermarking has emerged as a promising way to detect LLM-generated text.\nTo apply a watermark an LLM provider, given a secret key, augments generations with a signal that is later detectable by any party with the same key.\nRecent work has proposed three main families of waterm... | [
{
"id": "iPD812JhCw",
"initial_rating": 8,
"confidence": 3,
"soundness": 3,
"contribution": 3,
"presentation": 4,
"summary": "The paper shows that it is possible to detect the presence of most existing watermarks using black-box interaction with the model, without knowing the watermarkin... | {
"rating": "3;5;6;8",
"rating_avg": 5.5,
"confidence": "5;4;3;3",
"confidence_avg": 3.75,
"soundness": "2;3;4;3",
"soundness_avg": 3,
"contribution": "2;3;2;3",
"contribution_avg": 2.5,
"presentation": "1;3;4;4",
"presentation_avg": 3
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.316847"
} | {
"id": "CE76YId6Vf",
"metareview": "Summary: This paper studies the problem of detecting LLM watermarks in a black-box way, without even knowing the watermark key. Extensive experiments across three families of LLM watermarks, Red-Green, Fixed-Sampling and Cache-Augmented, verify the effectiveness of the proposed ... | {
"decision": "Accept (Poster)"
} |
E4NShSRRDP | 2406.00262v2 | Contrastive Learning Via Equivariant Representation | {
"content": "## Abstract\n\nAbstract Invariant Contrastive Learning (ICL) methods have achieved impressive performance across various domains. However, the absence of latent space representation for distortion (augmentation)-related information in the latent space makes ICL sub-optimal regarding training efficiency ... | [
{
"id": "hye6N224jd",
"initial_rating": 5,
"confidence": 4,
"soundness": 3,
"contribution": 2,
"presentation": 3,
"summary": "This paper proposes CLeVER, an ECL framework which adds regularization to the decoupled contrastive learning representations that is previously proposed, encourag... | {
"rating": "3;3;5;5",
"rating_avg": 4,
"confidence": "4;5;3;4",
"confidence_avg": 4,
"soundness": "3;2;2;3",
"soundness_avg": 2.5,
"contribution": "2;1;2;2",
"contribution_avg": 1.75,
"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": "2026-01-14T22:16:01.317492"
} | {
"id": "FgArFCGrkV",
"metareview": "This paper proposes an equivariant-based contrastive learning framework to address limitations in invariant contrastive learning (ICL). The authors claim that the absence of augmentation-related information in ICL leads to inefficiencies and reduced robustness. CLeVER introduces... | {
"decision": "Reject"
} |
E5YmIBvOqV | 2403.00269v2 | Large Convolutional Model Tuning via Filter Subspace | {
"content": "## Abstract\n\nAbstract Efficient fine-tuning methods are critical to address the high computational and parameter complexity while adapting large pre-trained models to downstream tasks.\nOur study is inspired by prior research that represents each convolution filter as a linear combination of a small s... | [
{
"id": "xAs75awy45",
"initial_rating": 6,
"confidence": 4,
"soundness": 3,
"contribution": 3,
"presentation": 3,
"summary": "This paper presents a new way to decompose convolutional layers and experimented a new way to fine-tune large models with those layers by adjusting a small number... | {
"rating": "5;5;6",
"rating_avg": 5.333333333333333,
"confidence": "5;3;4",
"confidence_avg": 4,
"soundness": "2;2;3",
"soundness_avg": 2.3333333333333335,
"contribution": "2;2;3",
"contribution_avg": 2.3333333333333335,
"presentation": "2;2;3",
"presentation_avg": 2.3333333333333335
} | {
"primary_area": "",
"track": "main",
"venue": "ICLR 2025 Poster",
"venueid": "ICLR.cc/2025/Conference",
"processed_at": "2026-01-14T22:16:01.318113"
} | {
"id": "j8heNwbaDi",
"metareview": "The paper presents the idea of filter subspace tuning, i.e. to fine-tune convolutional neural networks by only adapting spatial filter atoms while leaving the linear combination across channels frozen. The paper presents a clear motivation for the idea, which allows parameter ef... | {
"decision": "Accept (Poster)"
} |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.