论文检索

Reinforcement Learning · Deep RL

Oleh Rybkin, Michal Nauman, Preston Fu, Charlie Snell, Pieter Abbeel, Sergey Levine, Aviral Kumar

Scaling data and compute is critical in modern machine learning. However, scaling also demands _predictability_: we want methods to not only perform well with more compute or data, but also have their performance be predictable from low compute or low data runs, without ever running the large-scale experiment. In this paper, we show predictability of value-based off-policy deep RL. First, we show that data and compute requirements to reach a given performance level lie on a _Pareto frontier_, controlled by the updates-to-data (UTD) ratio. By estimating this frontier, we can extrapolate data requirements into a higher compute regime, and compute requirements into a higher data regime. Second, we determine the optimal allocation of total _budget_ across data and compute to obtain given performance and use it to determine hyperparameters that maximize performance for a given budget. Third, this scaling behavior is enabled by first estimating predictable relationships between different _hyperparameters_, which is used to counteract effects of overfitting and plasticity loss unique to RL. We validate our approach using three algorithms: SAC, BRO, and PQL on DeepMind Control, OpenAI gym, and IsaacGym, when extrapolating to higher levels of data, compute, budget, or performance.

Deep Learning · Theory

Noam Levi

Neural scaling laws have garnered significant interest due to their ability to predict model performance as a function of increasing parameters, data, and compute. In this work, we propose a simple statistical ansatz based on memorization to study scaling laws in the context of inference. Specifically, how performance improves with multiple inference attempts. We explore the coverage, or pass@k metric, which measures the chance of success over repeated attempts and provide a motivation for the observed functional form of the inference scaling behavior of the coverage in large language models (LLMs) on reasoning tasks. We then define an "inference loss", which exhibits a power law decay as the number of trials increases, and connect this result with prompting costs. We further test the universality of our construction by conducting experiments on a simple generative model, and find that our predictions are in agreement with the empirical coverage curves in a controlled setting. Our simple framework sets the ground for incorporating inference scaling with other known scaling laws.

General Machine Learning · Supervised Learning

Weiwei Li, Haitao Wu, Yunan Lu, Xiuyi Jia

Label distribution learning (LDL) is a powerful learning paradigm that emulates label polysemy by assigning label distributions over the label space. However, existing LDL evaluation metrics struggle to capture meaningful performance differences due to their insensitivity to subtle distributional changes, and existing LDL learning objectives often exhibit biases by disproportionately emphasizing a small subset of samples with extreme predictions. As a result, the LDL metrics lose their discriminability, and the LDL objectives are also at risk of overfitting. In this paper, we propose DeltaLDL, a percentage of predictions that are approximately correct within the context of LDL, as a solution to the above problems. DeltaLDL can serve as a novel evaluation metric, which is parameter-free and reflects more on real performance improvements. DeltaLDL can also serve as a novel learning objective, which is differentiable and encourages most samples to be predicted as approximately correct, thereby mitigating overfitting. Our theoretical analysis and empirical results demonstrate the effectiveness of the proposed solution.

Applications · Chemistry, Physics, and Earth Sciences

Chuqi CHEN, Yahong Yang, Yang Xiang, Wenrui Hao

Solving partial differential equations (PDEs) using neural networks has become a central focus in scientific machine learning. Training neural networks for singularly perturbed problems is particularly challenging due to certain parameters in the PDEs that introduce near-singularities in the loss function. In this study, we overcome this challenge by introducing a novel method based on homotopy dynamics to effectively manipulate these parameters. From a theoretical perspective, we analyze the effects of these parameters on training difficulty in these singularly perturbed problems and establish the convergence of the proposed homotopy dynamics method. Experimentally, we demonstrate that our approach significantly accelerates convergence and improves the accuracy of these singularly perturbed problems. These findings present an efficient optimization strategy leveraging homotopy dynamics, offering a robust framework to extend the applicability of neural networks for solving singularly perturbed differential equations.

General Machine Learning · Unsupervised and Semi-supervised Learning

Erez Peterfreund, Ofir Lindenbaum, Yuval Kluger, Boris Landa

Embedding and visualization techniques are essential for analyzing high-dimensional data, but they often struggle with complex data governed by multiple latent variables, potentially distorting key structural characteristics. This paper considers scenarios where the observed features can be partitioned into mutually exclusive subsets, each capturing a different smooth substructure. In such cases, visualizing the data based on each feature partition can better characterize the underlying processes and structures in the data, leading to improved interpretability. To partition the features, we propose solving an optimization problem that promotes graph Laplacian-based smoothness in each partition, thereby prioritizing partitions with simpler geometric structures. Our approach generalizes traditional embedding and visualization techniques, allowing them to learn multiple embeddings simultaneously. We establish that if several independent or partially dependent manifolds are embedded in distinct feature subsets in high-dimensional space, then our framework can reliably identify the correct subsets with theoretical guarantees. Finally, we demonstrate the effectiveness of our approach in extracting multiple low-dimensional structures and partially independent processes from both simulated and real data.

Probabilistic Methods · Monte Carlo and Sampling Methods

Aaron Havens, Benjamin Kurt Miller, Bing Yan, Carles Domingo i Enrich, Anuroop Sriram, Daniel S. Levine, Brandon Wood, Bin Hu, Brandon Amos, Brian Karrer 等

We introduce Adjoint Sampling, a highly scalable and efficient algorithm for learning diffusion processes that sample from unnormalized densities, or energy functions. It is the first on-policy approach that allows significantly more gradient updates than the number of energy evaluations and model samples, allowing us to scale to much larger problem settings than previously explored by similar methods.Our framework is theoretically grounded in stochastic optimal control and shares the same theoretical guarantees as Adjoint Matching, being able to train without the need for corrective measures that push samples towards the target distribution.We show how to incorporate key symmetries, as well as periodic boundary conditions, for modeling molecules in both cartesian and torsional coordinates.We demonstrate the effectiveness of our approach through extensive experiments on classical energy functions, and further scale up to neural network-based energy models where we perform amortized conformer generation across many molecular systems.To encourage further research in developing highly scalable sampling methods, we plan to open source these challenging benchmarks, where successful methods can directly impact progress in computational chemistry. Code \& and benchmarks provided at https://github.com/facebookresearch/adjoint_sampling.

Deep Learning · Large Language Models

Songlin Zhai, Yuan Meng, Yongrui Chen, Yiwei Wang, Guilin Qi

Large Language Models (LLMs) have revolutionized various natural language processing tasks with their remarkable capabilities. However, challenges persist in effectively integrating new knowledge into LLMs without compromising their performance, particularly in the Large Language Models (LLMs) have revolutionized various natural language processing tasks with their remarkable capabilities. However, a challenge persists in effectively processing new information, particularly in the area of long-term knowledge updates without compromising model performance. To address this challenge, this paper introduces a novel memory augmentation framework that conceptualizes memory as a peripheral component (akin to physical RAM), with the LLM serving as the information processor (analogous to a CPU). Drawing inspiration from RAM architecture, we design memory as a sequence of memory banks, each modeled using Kolmogorov-Arnold Network (KAN) to ensure smooth state transitions. Memory read and write operations are dynamically controlled by query signals derived from the LLMs' internal states, closely mimicking the interaction between a CPU and RAM. Furthermore, a dedicated memory bank is used to generate a mask value that indicates the relevance of the retrieved data, inspired by the sign bit in binary coding schemes. The retrieved memory feature is then integrated as a prefix to enhance the model prediction. Extensive experiments on knowledge-based model editing validate the effectiveness and efficiency of our peripheral memory.

Probabilistic Methods · Monte Carlo and Sampling Methods

Johannes Hertrich, Robert Gruhlke

In order to sample from an unnormalized probability density function, we propose to combine continuous normalizing flows (CNFs) with rejection-resampling steps based on importance weights. We relate the iterative training of CNFs with regularized velocity fields to a JKO scheme and prove convergence of the involved velocity fields to the velocity field of the Wasserstein gradient flow (WGF). The alternation of local flow steps and non-local rejection-resampling steps allows to overcome local minima or slow convergence of the WGF for multimodal distributions. Since the proposal of the rejection step is generated by the model itself, they do not suffer from common drawbacks of classical rejection schemes. The arising model can be trained iteratively, reduces the reverse Kullback-Leibler (KL) loss function in each step, allows to generate iid samples and moreover allows for evaluations of the generated underlying density. Numerical examples show that our method yields accurate results on various test distributions including high-dimensional multimodal targets and outperforms the state of the art in almost all cases significantly.

Deep Learning · Sequential Models, Time series

Yihang Wang, Yuying Qiu, Peng Chen, Kai Zhao, Yang Shu, Zhongwen Rao, Lujia Pan, Bin Yang, Chenjuan Guo

With the growing availability of multi-domain time series data, there is an increasing demand for general forecasting models pre-trained on multi-source datasets to support diverse downstream prediction scenarios. Existing time series foundation models primarily focus on scaling up pre-training datasets and model sizes to enhance generalization performance. In this paper, we take a different approach by addressing two critical aspects of general forecasting models: (1) how to derive unified representations from heterogeneous multi-domain time series data, and (2) how to effectively capture domain-specific features to enable adaptive transfer across various downstream scenarios. To address the first aspect, we propose Decomposed Frequency Learning as the pre-training task, which leverages frequency-based masking and reconstruction to decompose coupled semantic information in time series, resulting in unified representations across domains. For the second aspect, we introduce the Time Series Register, which captures domain-specific representations during pre-training and enhances adaptive transferability to downstream tasks. Our model achieves the state-of-the-art forecasting performance on seven real-world benchmarks, demonstrating remarkable few-shot and zero-shot capabilities.

Social Aspects · Accountability, Transparency, and Interpretability

Xingyi Yang, Constantin Venhoff, Ashkan Khakzar, Christian Schroeder de Witt, Puneet Dokania, Adel Bibi, Phil Torr

Neurons in large language models often exhibit \emph{polysemanticity}, simultaneously encoding multiple unrelated concepts and obscuring interpretability. Instead of relying on post-hoc methods, we present \textbf{MoE-X}, a mixture-of-experts (MoE) language model designed to be \emph{intrinsically} interpretable. Our approach is motivated by the observation that, in language models, wider networks with sparse activations are more likely to capture interpretable factors. however, directly training such large sparse networks is computationally prohibitive. MoE architectures offer a scalable alternative by activating only a subset of experts for any given input, inherently aligning with interpretability objectives. In MoE-X, we establish this connection by rewriting the MoE layer as an equivalent sparse, large MLP. This approach enables efficient scaling of the hidden size while maintaining sparsity. To further enhance interpretability, we enforce sparse activation within each expert and redesign the routing mechanism to prioritize experts with the highest activation sparsity. These designs ensure that only the most salient features are routed and processed by the experts. We evaluate MoE-X on chess and natural language tasks, showing that it achieves performance comparable to dense models while significantly improving interpretability. MoE-X achieves a perplexity better than GPT-2, with interpretability surpassing even sparse autoencoder (SAE)-based approaches.

Applications · Computer Vision

Zehong Ma, Shiliang Zhang, Longhui Wei, Qi Tian

Traditional approaches to adapting multi-modal large language models (MLLMs) to new tasks have relied heavily on fine-tuning. This paper introduces Efficient Multi-Modal Long Context Learning (EMLoC), a novel training-free alternative that embeds demonstration examples directly into the model input. EMLoC offers a more efficient, flexible, and scalable solution for task adaptation. Because extremely lengthy inputs introduce prohibitive computational and memory overhead, EMLoC contributes a chunk-wise compression mechanism combined with layer-wise adaptive pruning. It condenses long-context multimodal inputs into compact, task-specific memory representations. By adaptively pruning tokens at each layer under a Jensen-Shannon divergence constraint, our method achieves a dramatic reduction in inference complexity without sacrificing performance. This approach is the first to seamlessly integrate compression and pruning techniques for multi-modal long-context learning, offering a scalable and efficient solution for real-world applications. Extensive experiments on diverse vision-language benchmarks demonstrate that EMLoC achieves performance on par with or superior to naive long-context approaches. Our results highlight the potential of EMLoC as a groundbreaking framework for efficient and flexible adaptation of multi-modal models in resource-constrained environments. Codes are publicly available at https://github.com/Zehong-Ma/EMLoC.

Deep Learning · Large Language Models

Shashwat Goel, Joschka Strüber, Ilze Amanda Auzina, Karuna Chandra, Ponnurangam Kumaraguru, Douwe Kiela, Ameya Pandurang Prabhu, Matthias Bethge, Jonas Geiping

As Language Model (LM) capabilities advance, evaluating and supervising them at scale is getting harder for humans. There is hope that other language models can automate both these tasks, which we refer to as *AI Oversight*. We study how model similarity affects both aspects of AI oversight by proposing *Chance Adjusted Probabilistic Agreement (CAPA)*--a metric for LM similarity based on overlap in model mistakes. Using CAPA, we first show that *LLM-as-a-judge* scores favor models similar to the judge, generalizing recent self-preference results. Then, we study training on LM annotations, and find complementary knowledge between the weak supervisor and strong student model plays a crucial role in gains from *weak-to-strong generalization*. As model capabilities increase, it becomes harder to find their mistakes, and we might defer more to AI oversight. However, we observe a concerning trend--model mistakes are becoming more similar with increasing capabilities, pointing to risks from correlated failures. Our work underscores the importance of reporting and correcting for model similarity, especially in the emerging paradigm of AI oversight.

Social Aspects · Robustness

Harvineet Singh, Fan Xia, Alexej Gossmann, Andrew Chuang, Julian Hong, Jean Feng

Machine learning (ML) models frequently experience performance degradation when deployed in new contexts. Such degradation is rarely uniform: some subgroups may suffer large performance decay while others may not. Understanding where and how large differences in performance arise is critical for designing *targeted* corrective actions that mitigate decay for the most affected subgroups while minimizing any unintended effects. Current approaches do not provide such detailed insight, as they either (i) explain how *average* performance shifts arise or (ii) identify adversely affected subgroups without insight into how this occurred. To this end, we introduce a **S**ubgroup-scanning **H**ierarchical **I**nference **F**ramework for performance drif**T** (SHIFT). SHIFT first asks "Is there any subgroup with unacceptably large performance decay due to covariate/outcome shifts?" (*Where?*) and, if so, dives deeper to ask "Can we explain this using more detailed variable(subset)-specific shifts?" (*How?*). In real-world experiments, we find that SHIFT identifies interpretable subgroups affected by performance decay, and suggests targeted actions that effectively mitigate the decay.

Applications · Computer Vision

Jiaqi Wang, Jingtao Li, Weiming Zhuang, Chen Chen, Lingjuan Lyu, Fenglong Ma

Vision foundation models (FMs) like CLIP have exhibited exceptional capabilities in visual and linguistic understanding, particularly in zero-shot inference tasks. However, these models struggle with data that significantly deviates from their training samples, necessitating fine-tuning, which is often infeasible in centralized settings due to data privacy concerns. Federated learning (FL) combined with parameter-efficient fine-tuning (PEFT) offers a potential solution, yet existing methods face issues with domain-specific characteristics and out-of-domain generalization. We propose a cross-silo Federated Adapter Generalization (FedAG), a novel federated fine-tuning approach that leverages multiple fine-grained adapters to capture domain-specific knowledge while enhancing out-of-domain generalization. Our method uses quality-aware in-domain mutual learning and attention-regularized cross-domain learning to integrate domain-specific insights effectively. Experiments of the CLIP model on three domain-shifting datasets, ImageCLEF-DA, Office-Home, and DomainNet, demonstrate the superior performance of FedAG in both in-domain and out-of-domain scenarios. We envision this work as a milestone for generalizing CLIP to handle the challenge of out-of-domain knowledge under federated learning setting.

Social Aspects · Security

Chenyang Zhang, Xiaoyu Zhang, Jian Lou, KAI WU, Zilong Wang, Xiaofeng Chen

Vision-Language Retrieval-Augmented Generation (VLRAG) systems have been widely applied to Large Vision-Language Models (LVLMs) to enhance their generation ability. However, the reliance on external multimodal knowledge databases renders VLRAG systems vulnerable to malicious poisoning attacks. In this paper, we introduce PoisonedEye, the first knowledge poisoning attack designed for VLRAG systems. Our attack successfully manipulates the response of the VLRAG system for the target query by injecting only one poison sample into the knowledge database. To construct the poison sample, we follow two key properties for the retrieval and generation process, and identify the solution by satisfying these properties. Besides, we also introduce a class query targeted poisoning attack, a more generalized strategy that extends the poisoning effect to an entire class of target queries. Extensive experiments on multiple query datasets, retrievers, and LVLMs demonstrate that our attack is highly effective in compromising VLRAG systems.

Theory · Learning Theory

Hongyi Wan, Shiyuan Ren, Wei Huang, Miao Zhang, Xiang Deng, Yixin Bao, Liqiang Nie

Continual learning (CL) is crucial for advancing human-level intelligence, but its theoretical understanding, especially regarding factors influencing forgetting, is still relatively limited. This work aims to build a unified theoretical framework for understanding CL using feature learning theory. Different from most existing studies that analyze forgetting under linear regression model or lazy training, we focus on a more practical two-layer convolutional neural network (CNN) with polynomial ReLU activation for sequential tasks within a signal-noise data model. Specifically, we theoretically reveal how the angle between task signal vectors influences forgetting that: *acute or small obtuse angles lead to benign forgetting, whereas larger obtuse angles result in harmful forgetting*. Furthermore, we demonstrate that the replay method alleviates forgetting by expanding the range of angles corresponding to benign forgetting. Our theoretical results suggest that mid-angle sampling, which selects examples with moderate angles to the prototype, can enhance the replay method's ability to mitigate forgetting. Experiments on synthetic and real-world datasets confirm our theoretical results and highlight the effectiveness of our mid-angle sampling strategy.

Social Aspects · Fairness

Zichong Wang, Wenbin Zhang

Graph generation models have shown significant potential across various domains. However, despite their success, these models often inherit societal biases, limiting their adoption in real-world applications. Existing research on fairness in graph generation primarily addresses structural bias, overlooking the critical issue of feature bias. To address this gap, we propose FDGen, a novel approach that defines and mitigates both feature and structural biases in graph generation models. Furthermore, we provide a theoretical analysis of how bias sources in graph data contribute to disparities in graph generation tasks. Experimental results on four real-world datasets demonstrate that FDGen outperforms state-of-the-art methods, achieving notable improvements in fairness while maintaining competitive generation performance.

Deep Learning · Large Language Models

Valentyn Boreiko, Alexander Panfilov, Václav Voráček, Matthias Hein, Jonas Geiping

A plethora of jailbreaking attacks have been proposed to obtain harmful responses from safety-tuned LLMs. These methods largely succeed in coercing the target output in their original settings, but their attacks vary substantially in fluency and computational effort. In this work, we propose a unified threat model for the principled comparison of these methods.Our threat model checks if a given jailbreak is likely to occur in the distribution of text. For this, we build an N-gram language model on 1T tokens, which, unlike model-based perplexity, allows for an LLM-agnostic, nonparametric, and inherently interpretable evaluation. We adapt popular attacks to this threat model, and, for the first time, benchmark these attacks on equal footing with it. After an extensive comparison, we find attack success rates against safety-tuned modern models to be lower than previously presented and that attacks based on discrete optimization significantly outperform recent LLM-based attacks. Being inherently interpretable, our threat model allows for a comprehensive analysis and comparison of jailbreak attacks. We find that effective attacks exploit and abuse infrequent bigrams, either selecting the ones absent from real-world text or rare ones, e.g., specific to Reddit or code datasets.

Deep Learning · Large Language Models

Xintong Sun, Chi Wei, Minghao Tian, Shiwen Ni

Large Language Models (LLMs) have shown remarkable capabilities, yet ensuring their outputs conform to strict structural or grammatical constraints remains challenging, which is critical in function calls and domain-specific language (DSL) generation. Constrained decoding with context-free grammar is a flexible approach to guarantee LLMs' adherence to a specific format by dynamically building a token logits mask. However, creating this mask requires checking the validity of all tokens in the LLM vocabulary at every decoding step, which often incurs significant overheads in existing constrained decoding engines. To address this challenge, we propose $\textbf{ZapFormat}$, a novel $\textbf{dynamic pruning}$ strategy based on the Earley algorithm that identifies and eliminates invalid or redundant Earley states in real-time, significantly reducing memory occupation of the Earley algorithm's states. This further enables us to use a state cache to speed up structured generations on a large number of queries. We implemented ZapFormat in a new constrained decoding engine called Formatron which also incorporates existing optimizations. Through comprehensive experiments on structured generation tasks, including JSON generation, JSON Schema validation, and semantic parsing, we demonstrate that Formatron not only $\textbf{consistently maintains}$ high-precision compliant outputs but also achieves $\textbf{significant improvements}$ in inference speed up to 2x compared to state-of-the-art implementations. More importantly, Formatron is generally applicable across various LLM architectures. We release Formatron as open source at https://github.com/Dan-wanna-M/formatron.

Deep Learning · Large Language Models

Yurun Yuan, Tengyang Xie

Leveraging more test-time computation has proven to be an effective way to boost the reasoning capabilities of large language models (LLMs). Among various methods, the verify-and-improve paradigm stands out for enabling dynamic solution exploration and feedback incorporation. However, existing approaches often suffer from restricted feedback spaces and lack of coordinated training of different parties, leading to suboptimal performance. To address this, we model this multi-turn refinement process as a Markov Decision Process and introduce DPSDP (**D**irect **P**olicy **S**earch by **D**ynamic **P**rogramming), a reinforcement learning algorithm that trains an actor-critic LLM system to iteratively refine answers via direct preference learning on self-generated data. Theoretically, DPSDP can match the performance of any policy within the training distribution. Empirically, we instantiate DPSDP with various base models and show improvements on both in- and out-of-distribution benchmarks. For example, on benchmark MATH 500, majority voting over five refinement steps increases first-turn accuracy from 58.2% to 63.2% with Ministral-based models. An ablation study further confirms the benefits of multi-agent collaboration and out-of-distribution generalization.