ICML 2025 Papers

Millions of abandoned oil and gas wells are scattered across the world, leaching methane into the atmosphere and toxic compounds into the groundwater. Many of these locations are unknown, preventing the wells from being plugged and their polluting effects averted. Remote sensing is a relatively unexplored tool for pinpointing abandoned wells at scale. We introduce the first large-scale Benchmark dataset for this problem, leveraging high-resolution multi-spectral satellite imagery from Planet Labs. Our curated Dataset comprises over 213,000 wells (abandoned, suspended, and active) from Alberta, a region with especially high well density, sourced from the Alberta Energy Regulator and verified by domain experts. We evaluate baseline algorithms for well detection and segmentation, showing the promise of computer vision approaches but also significant room for improvement.

The quest for Continual Learning (CL) seeks to empower neural networks with the ability to learn and adapt incrementally. Central to this pursuit is addressing the stability-plasticity dilemma, which involves striking a balance between two conflicting objectives: preserving previously learned knowledge and acquiring new knowledge. While numerous CL methods aim to achieve this trade-off, they often overlook the impact of network architecture on stability and plasticity, restricting the trade-off to the parameter level. In this paper, we delve into the conflict between stability and plasticity at the architectural level. We reveal that under an equal parameter constraint, deeper networks exhibit better plasticity, while wider networks are characterized by superior stability. To address this architectural-level dilemma, we introduce a novel framework denoted Dual-Arch, which serves as a plug-in component for CL. This framework leverages the complementary strengths of two distinct and independent networks: one dedicated to plasticity and the other to stability. Each network is designed with a specialized and lightweight architecture, tailored to its respective objective. Extensive experiments demonstrate that Dual-Arch enhances the performance of existing CL methods while being up to 87% more compact in terms of parameters.

The emergence of multimodal large language models (MLLMs) advances multimodal emotion recognition (MER) to the next level—from naive discriminative tasks to complex emotion understanding with advanced video understanding abilities and natural language description. However, the current community suffers from a lack of large-scale datasets with intensive, descriptive emotion annotations, as well as a multimodal-centric framework to maximize the potential of MLLMs for emotion understanding. To address this, we establish a new benchmark for MLLM-based emotion understanding with a novel dataset (MER-Caption) and a new model (AffectGPT). Utilizing our model-based crowd-sourcing data collection strategy, we construct the largest descriptive emotion dataset to date (by far), featuring over 2K fine-grained emotion categories across 115K samples. We also introduce the AffectGPT model, designed with pre-fusion operations to enhance multimodal integration. Finally, we present MER-UniBench, a unified benchmark with evaluation metrics tailored for typical MER tasks and the free-form, natural language output style of MLLMs. Extensive experimental results show AffectGPT's robust performance across various MER tasks. We have released both the code and the dataset to advance research and development in emotion understanding: https://github.com/zeroQiaoba/AffectGPT.

We introduce PaperBench, a benchmark evaluating the ability of AI agents to replicate state-of-the-art AI research. Agents must replicate 20 ICML 2024 Spotlight and Oral papers from scratch, including understanding paper contributions, developing a codebase, and successfully executing experiments. For objective evaluation, we develop rubrics that hierarchically decompose each replication task into smaller sub-tasks with clear grading criteria. In total, PaperBench contains 8,316 individually gradable tasks. Rubrics are co-developed with the author(s) of each ICML paper for accuracy and realism. To enable scalable evaluation, we also develop an LLM-based judge to automatically grade replication attempts against rubrics, and assess our judge's performance by creating a separate benchmark for judges. We evaluate several frontier models on PaperBench, finding that the best-performing tested agent, Claude 3.5 Sonnet (New) with open-source scaffolding, achieves an average replication score of 21.0%. Finally, we recruit top ML PhDs to attempt a subset of PaperBench, finding that models do not yet outperform the human baseline. We open-source our code (https://github.com/openai/preparedness) to facilitate future research in understanding the AI engineering capabilities of AI agents.

Decision trees (DTs) and their random forest (RF) extensions are workhorses of classification and regression in Euclidean spaces. However, algorithms for learning in non-Euclidean spaces are still limited. We extend DT and RF algorithms to product manifolds: Cartesian products of several hyperbolic, hyperspherical, or Euclidean components. Such manifolds handle heterogeneous curvature while still factorizing neatly into simpler components, making them compelling embedding spaces for complex datasets. Our novel angular reformulation respects manifold geometry while preserving the algorithmic properties that make decision trees effective. In the special cases of single-component manifolds, our method simplifies to its Euclidean or hyperbolic counterparts, or introduces hyperspherical DT algorithms, depending on the curvature. In benchmarks on a diverse suite of 57 classification, regression, and link prediction tasks, our product RFs ranked first on 29 tasks and came in the top 2 for 41. This highlights the value of product RFs as straightforward yet powerful new tools for data analysis in product manifolds. Code for our method is available at https://github.com/pchlenski/manify.

Dataset contamination, where evaluation datasets overlap with pre-training corpora, inflates performance metrics and undermines the reliability of model evaluations. Measuring dataset contamination thus becomes essential to ensure that performance evaluations genuinely reflect a model's ability to generalize to unseen data, rather than relying on memorized examples. To address this problem, we propose Kernel Divergence Score (KDS), a novel method that evaluates dataset contamination by computing the divergence between the kernel similarity matrix of sample embeddings, before and after fine-tuning on the benchmark dataset. Leveraging the insight that fine-tuning affects unseen samples more significantly than seen ones, KDS provides a reliable measure of contamination. Through extensive experiments on controlled contamination scenarios, KDS demonstrates a near-perfect correlation with contamination levels and outperforms existing baselines. Additionally, we perform comprehensive ablation studies to analyze the impact of key design choices, providing deeper insights into the components and effectiveness of KDS. These ablations highlight the importance of leveraging fine-grained kernel-based information and confirm the reliability of the proposed framework across diverse datasets and settings. Code is released in https://github.com/deeplearning-wisc/kernel-divergence-score.

The Platonic Representation Hypothesis claims that recent foundation models are converging to a shared representation space as a function of their downstream task performance, irrespective of the objectives and data modalities used to train these models (Huh et al., 2024). Representational similarity is generally measured for individual datasets and is not necessarily consistent across datasets. Thus, one may wonder whether this convergence of model representations is confounded by the datasets commonly used in machine learning. Here, we propose a systematic way to measure how representational similarity between models varies with the set of stimuli used to construct the representations. We find that the objective function is a crucial factor in determining the consistency of representational similarities across datasets. Specifically, self-supervised vision models learn representations whose relative pairwise similarities generalize better from one dataset to another compared to those of image classification or image-text models. Moreover, the correspondence between representational similarities and the models' task behavior is dataset-dependent, being most strongly pronounced for single-domain datasets. Our work provides a framework for analyzing similarities of model representations across datasets and linking those similarities to differences in task behavior.

Random defense represents a promising strategy to protect neural networks from adversarial attacks. Most of these methods enhance robustness by injecting randomness into the data, increasing uncertainty for attackers. However, this randomness could reduce the generalization capacity of defense, as defense performance could be sensitive to the hyperparameters of noise added to the data, making it difficult to generalize across different datasets. Additionally, the involvement of randomness always comes with a reduction of natural accuracy, which leads to a delicate trade-off between them, which is seldom studied in random defense. In this work, we propose incorporating randomness into the network structure instead of data input by designing stochastic deformable convolution, where a random mask replaces the convolutional offset. This process promotes data independence, enhancing generalization across datasets. To study the trade-off, we conduct a theoretical analysis of both robust and clean accuracy, from a perspective of gradient cosine similarity and natural inference. Based on the analysis, we reformulate the adversarial training in our random defense framework. Extensive experiments show that our method achieves SOTA adversarial robustness and clean accuracy compared with other random defense methods.

Aligning language models with human preferences relies on pairwise preference datasets. While some studies suggest that on-policy data consistently outperforms off-policy data for preference learning, others indicate that the advantages of on-policy data are task-dependent, highlighting the need for a systematic exploration of their interplay. In this work, we show that on-policy and off-policy data offer complementary strengths: on-policy data is particularly effective for reasoning tasks like math and coding, while off-policy data performs better on subjective tasks such as creative writing and making personal recommendations. Guided by these findings, we introduce SimpleMix, an approach to combine the complementary strengths of on-policy and off-policy preference learning by simply mixing these two data sources. Our empirical results across diverse tasks and benchmarks demonstrate that SimpleMix substantially improves language model alignment. Specifically, SimpleMix improves upon on-policy DPO and off-policy DPO by an average of 6.03 on Alpaca Eval 2.0. Moreover, it surpasses prior approaches that are much more complex in combining on- and off-policy data, such as HyPO and DPO-Mix-P, by an average of 3.05. These findings validate the effectiveness and efficiency of SimpleMix for enhancing preference-based alignment.

Spatio-temporal forecasting provides potential for discovering evolutionary patterns in geographical scientific data. However, geographical scientific datasets are often manually collected across studies, resulting in limited time spans and data scales. This hinders existing methods that rely on rich historical data for individual entities. In this paper, we argue that heterogeneous datasets from different studies can provide complementary insights into the same underlying system, helping improve predictions for geographical entities with limited historical data. To this end, we propose a Segment Quadtree Geographical Embedding Framework (SQGEF). SQGEF integrates knowledge from datasets with varied target entities, time spans, and observation variables to learn unified representations for multi-granularity entities—including those absent during training. Specifically, we propose a novel data structure, Segment Quadtree, that flexibly accommodates entities of varying granularities. SQGEF not only captures multi-level interactions from grid data but also extracts nested relationships and human-defined boundaries from diverse entities, enabling a comprehensive understanding of complex geographical structures. Experiments on real-world datasets demonstrate that SQGEF effectively represents unseen geographical entities and enhances performance for various models.

We propose PieClam (Prior Inclusive Exclusive Cluster Affiliation Model): a graph autoencoder, where nodes are embedded into a code space by an algorithm that maximizes the log-likelihood of the decoded graph. PieClam is a community affiliation model that extends well-known methods like BigClam in two main manners. First, instead of the decoder being defined via pairwise interactions between the nodes in the code space, we also incorporate a learned prior on the distribution of nodes in the code space, turning our method into a graph generative model. Secondly, we generalize the notion of communities by allowing not only sets of nodes with strong connectivity, which we call inclusive communities, but also sets of nodes with strong disconnection, which we call exclusive communities. By introducing a new graph similarity measure, called the log cut distance, we show that PieClam is a universal autoencoder, able to uniformly approximately reconstruct any graph. Our method is shown to obtain competitive performance in graph anomaly detection and link prediction benchmarks.

The propensity of large language models (LLMs) to generate hallucinations and non-factual content undermines their reliability in high-stakes domains, where rigorous control over Type I errors (the conditional probability of incorrectly classifying hallucinations as truthful content) is essential. Despite its importance, formal verification of LLM factuality with such guarantees remains largely unexplored. In this paper, we introduce FactTest, a novel framework that statistically assesses whether an LLM can provide correct answers to given questions with high-probability correctness guarantees. We formulate hallucination detection as a hypothesis testing problem to enforce an upper bound of Type I errors at user-specified significance levels. Notably, we prove that FactTest also ensures strong Type II error control under mild conditions and can be extended to maintain its effectiveness when covariate shifts exist. Our approach is distribution-free and works for any number of human-annotated samples. It is model-agnostic and applies to any black-box or white-box LM. Extensive experiments on question-answering (QA) benchmarks demonstrate that FactTest effectively detects hallucinations and enable LLMs to abstain from answering unknown questions, leading to an over 40% accuracy improvement.

With recent dramatic increases in AI system capabilities, there has been growing interest in utilizing machine learning for reasoning-heavy, quantitative tasks, particularly mathematics. While there are many resources capturing mathematics at the high-school, undergraduate, and graduate level, there are far fewer resources available that align with the level of difficulty and open endedness encountered by professional mathematicians working on open problems. To address this, we introduce a new collection of datasets, the Algebraic Combinatorics Dataset Repository (ACD Repo), representing either foundational results or open problems in algebraic combinatorics, a subfield of mathematics that studies discrete structures arising from abstract algebra. Further differentiating our dataset collection is the fact that it aims at the conjecturing process. Each dataset includes an open-ended research level question and a large collection of examples (up to 10M in some cases) from which conjectures should be generated. We describe all nine datasets, the different ways machine learning models can be applied to them (e.g., training with narrow models followed by interpretability analysis or program synthesis with LLMs), and discuss some of the challenges involved in designing datasets like these.

Multimodal Continual Instruction Tuning (MCIT) aims to enable Multimodal Large Language Models (MLLMs) to incrementally learn new tasks without catastrophic forgetting, thus adapting to evolving requirements. In this paper, we explore the forgetting caused by such incremental training, categorizing it into superficial forgetting and essential forgetting. Superficial forgetting refers to cases where the model’s knowledge may not be genuinely lost, but its responses to previous tasks deviate from expected formats due to the influence of subsequent tasks’ answer styles, making the results unusable. On the other hand, essential forgetting refers to situations where the model provides correctly formatted but factually inaccurate answers, indicating a true loss of knowledge. Assessing essential forgetting necessitates addressing superficial forgetting first, as severe superficial forgetting can conceal the model’s knowledge state. Hence, we first introduce the Answer Style Diversification (ASD) paradigm, which defines a standardized process for data style transformations across different tasks, unifying their training sets into similarly diversified styles to prevent superficial forgetting caused by style shifts. Building on this, we propose RegLoRA to mitigate essential forgetting. RegLoRA stabilizes key parameters where prior knowledge is primarily stored by applying regularization to LoRA’s weight update matrices, enabling the model to retain existing competencies while remaining adaptable to new tasks. Experimental results demonstrate that our overall method, SEFE, achieves state-of-the-art performance.

We study the relationship between gradient-based optimization of parametric models (e.g., neural networks) and optimization of linear combinations of random features. Our main result shows that if a parametric model can be learned using mini-batch stochastic gradient descent (bSGD) without making assumptions about the data distribution, then with high probability, the target function can also be approximated using a polynomial-sized combination of random features. The size of this combination depends on the number of gradient steps and numerical precision used in the bSGD process. This finding reveals fundamental limitations of distribution-free learning in neural networks trained by gradient descent, highlighting why making assumptions about data distributions is often crucial in practice. Along the way, we also introduce a new theoretical framework called average probabilistic dimension complexity (adc), which extends the probabilistic dimension complexity developed by Kamath et al. (2020). We prove that adc has a polynomial relationship with statistical query dimension, and use this relationship to demonstrate an infinite separation between adc and standard dimension complexity.

Language model pretraining involves training on extensive corpora, where data quality plays a pivotal role. In this work, we aim to directly estimate the contribution of data during pretraining and select pretraining data in an efficient manner. Specifically, we draw inspiration from recent findings showing that compression efficiency (i.e., normalized loss) of diverse models on certain text correlates strongly with their downstream performance, when the text domain aligns with the downstream benchmarks (Huang et al., 2024). Building on this observation, we hypothesize that data on which model losses are predictive of downstream abilities also contribute effectively to learning, which shares similar intuition with Thrush et al. (2024). To leverage this insight, we introduce predictive data selection (PreSelect), a lightweight and efficient data selection method that requires training and deploying only a fastText-based scorer. Through comprehensive experiments with 1B and 3B parameter models, we demonstrate that models trained on 30B tokens selected with PreSelect surpass the performance of the vanilla baseline trained on 300B tokens, achieving a 10x reduction in compute requirements. Furthermore, PreSelect significantly outperforms other competitive data selection baselines, such as DCLM and FineWeb-Edu on a scale of 3B models trained on 100B tokens. We open-source our trained data selection scorer along with the curated datasets at https://github.com/hkust-nlp/PreSelect.

Exemplar-Free Class-Incremental Learning (EFCIL) aims to sequentially learn from distinct categories without retaining exemplars but easily suffers from catastrophic forgetting of learned knowledge. While existing EFCIL methods leverage knowledge distillation to alleviate forgetting, they still face two critical challenges: semantic shift and decision bias. Specifically, the embeddings of old tasks shift in the embedding space after learning new tasks, and the classifier becomes biased towards new tasks due to training solely with new data, hindering the balance between old and new knowledge. To address these issues, we propose the Dual-Projection Shift Estimation and Classifier Reconstruction (DPCR) approach for EFCIL. DPCR effectively estimates semantic shift through a dual-projection, which combines a learnable transformation with a row-space projection to capture both task-wise and category-wise shifts. Furthermore, to mitigate decision bias, DPCR employs ridge regression to reformulate a classifier reconstruction process. This reconstruction exploits previous in covariance and prototype of each class after calibration with estimated shift, thereby reducing decision bias. Extensive experiments demonstrate that, on various datasets, DPCR effectively balances old and new tasks, outperforming state-of-the-art EFCIL methods. Our codes are available at https://github.com/RHe502/ICML25-DPCR.

Probabilistic verification problems of neural networks are concerned with formally analysing the output distribution of a neural network under a probability distribution of the inputs. Examples of probabilistic verification problems include verifying the demographic parity fairness notion or quantifying the safety of a neural network. We present a new algorithm for solving probabilistic verification problems of neural networks based on an algorithm for computing and iteratively refining lower and upper bounds on probabilities over the outputs of a neural network. By applying state-of-the-art bound propagation and branch and bound techniques from non-probabilistic neural network verification, our algorithm significantly outpaces existing probabilistic verification algorithms, reducing solving times for various benchmarks from the literature from tens of minutes to tens of seconds. Furthermore, our algorithm compares favourably even to dedicated algorithms for restricted probabilistic verification problems. We complement our empirical evaluation with a theoretical analysis, proving that our algorithm is sound and, under mildly restrictive conditions, also complete when using a suitable set of heuristics.

Continual learning is crucial for applying machine learning in challenging, dynamic, and often resource-constrained environments. However, catastrophic forgetting — overwriting previously learned knowledge when new information is acquired — remains a major challenge. In this work, we examine the intermediate representations in neural network layers during continual learning and find that such representations are less prone to forgetting, highlighting their potential to accelerate computation. Motivated by these findings, we propose to use auxiliary classifiers~(ACs) to enhance performance and demonstrate that integrating ACs into various continual learning methods consistently improves accuracy across diverse evaluation settings, yielding an average 10\% relative gain. We also leverage the ACs to reduce the average cost of the inference by 10-60\% without compromising accuracy, enabling the model to return the predictions before computing all the layers. Our approach provides a scalable and efficient solution for continual learning.