ICML 2024 Papers

Cross-modality transfer aims to leverage large pretrained models to complete tasks that may not belong to the modality of pretraining data. Existing works achieve certain success in extending classical finetuning to cross-modal scenarios, yet we still lack understanding about the influence of modality gap on the transfer. In this work, a series of experiments focusing on the source representation quality during transfer are conducted, revealing the connection between larger modality gap and lesser knowledge reuse which means ineffective transfer. We then formalize the gap as the knowledge misalignment between modalities using conditional distribution $P(Y|X)$. Towards this problem, we present **Mo**dality k**N**owledge **A**lignment (MoNA), a meta-learning approach that learns target data transformation to reduce the modality knowledge discrepancy ahead of the transfer. Experiments show that the approach significantly improves upon cross-modal finetuning methods, and most importantly leads to better reuse of source modality knowledge.

Known learning rules tend to fall near one of two extremes: single-pass associative learning with low complexity and capacity, and multi-pass iterative learning with high complexity and capacity. In this work we investigate the mathematical feasibility of learning rules that are both single-pass and achieve the theoretical upper bound on capacity. We consider a fairly broad family of learning rules we call ``span rules,'' which include known rules such as Hebbian learning, perceptron learning, and backpropagation as special cases. To our knowledge, previous work has not determined whether single-pass, full-capacity span rules exist, even in the most fundamental case of a linear threshold neuron with binary input vectors, which is the focus of this study. We derive a necessary condition for the existence of such learning rules, which takes the form of a linear program, and show that the linear program is infeasible. This establishes an impossibility result that span rules can not be both single-pass and full-capacity.

I introduce a novel associative memory model named *Correlated Dense Associative Memory* (CDAM), which integrates both auto- and hetero-association in a unified framework for continuous-valued memory patterns. Employing an arbitrary graph structure to semantically link memory patterns, CDAM is theoretically and numerically analysed, revealing four distinct dynamical modes: auto-association, narrow hetero-association, wide hetero-association, and neutral quiescence. Drawing inspiration from inhibitory modulation studies, I employ anti-Hebbian learning rules to control the range of hetero-association, extract multi-scale representations of community structures in graphs, and stabilise the recall of temporal sequences. Experimental demonstrations showcase CDAM's efficacy in handling real-world data, replicating a classical neuroscience experiment, performing image retrieval, and simulating arbitrary finite automata.

Knowledge distillation (KD) has been widely used for model compression and deployment acceleration. Nonetheless, the statistical insight of the remarkable performance of KD remains elusive, and methods for evaluating the uncertainty of the distilled model/student model are lacking. To address these issues, we establish a close connection between KD and a Bayesian model. In particular, we develop an innovative method named Bayesian Knowledge Distillation (BKD) to provide a transparent interpretation of the working mechanism of KD, and a suite of Bayesian inference tools for the uncertainty quantification of the student model. In BKD, the regularization imposed by the teacher model in KD is formulated as a teacher-informed prior for the student model's parameters. Consequently, we establish the equivalence between minimizing the KD loss and estimating the posterior mode in BKD. Efficient Bayesian inference algorithms are developed based on the stochastic gradient Langevin Monte Carlo and examined with extensive experiments on uncertainty ranking and credible intervals construction for predicted class probabilities.

Incorporating spectral information to enhance Graph Neural Networks (GNNs) has shown promising results but raises a fundamental challenge due to the inherent ambiguity of eigenvectors. Various architectures have been proposed to address this ambiguity, referred to as spectral invariant architectures. Notable examples include GNNs and Graph Transformers that use spectral distances, spectral projection matrices, or other invariant spectral features. However, the potential expressive power of these spectral invariant architectures remains largely unclear. The goal of this work is to gain a deep theoretical understanding of the expressive power obtainable when using spectral features. We first introduce a novel message-passing framework for designing spectral invariant GNNs, called Eigenspace Projection GNN (EPNN). Our comprehensive analysis shows that EPNN essentially unifies all prior spectral invariant architectures, in that they are either strictly less expressive or equivalent to EPNN. A fine-grained expressiveness hierarchy among different architectures is also established. On the other hand, we present a surprising result that EPNN itself is bounded by a recently proposed class of Subgraph GNNs, implying that all these spectral invariant architectures are strictly less expressive than 3-WL. Finally, we demonstrate that these spectral features offer no additional advantage when combined with more expressive GNNs.

The popularity of federated learning comes from the possibility of better scalability and the ability for participants to keep control of their data, improving data security and sovereignty. Unfortunately, sharing model updates also creates a new privacy attack surface. In this work, we characterize the privacy guarantees of decentralized learning with random walk algorithms, where a model is updated by traveling from one node to another along the edges of a communication graph. Using a recent variant of differential privacy tailored to the study of decentralized algorithms, namely Pairwise Network Differential Privacy, we derive closed-form expressions for the privacy loss between each pair of nodes where the impact of the communication topology is captured by graph theoretic quantities. Our results further reveal that random walk algorithms tends to yield better privacy guarantees than gossip algorithms for nodes close from each other. We supplement our theoretical results with empirical evaluation on synthetic and real-world graphs and datasets.

We systematically investigate the expressive power of path-based graph neural networks. While it has been shown that path-based graph neural networks can achieve strong empirical results, an investigation into their expressive power is lacking. Therefore, we propose PATH-WL, a general class of color refinement algorithms based on paths and shortest path distance information. We show that PATH-WL is incomparable to a wide range of expressive graph neural networks, can count cycles, and achieves strong empirical results on the notoriously difficult family of strongly regular graphs. Our theoretical results indicate that PATH-WL forms a new hierarchy of highly expressive graph neural networks.

Time series data across scientific domains are often collected under distinct states (e.g., tasks), wherein latent processes (e.g., biological factors) create complex inter- and intra-state variability. A key approach to capture this complexity is to uncover fundamental interpretable units within the data, Building Blocks (BBs), which modulate their activity and adjust their structure across observations. Existing methods for identifying BBs in multi-way data often overlook inter- vs. intra-state variability, produce uninterpretable components, or do not align with properties of real-world data, such as missing samples and sessions of different duration. Here, we present a framework for Similarity-driven Building Block Inference using Graphs across States (SiBBlInGS). SiBBlInGS offers a graph-based dictionary learning approach for discovering sparse BBs along with their temporal traces, based on co-activity patterns and inter- vs. intra-state relationships. Moreover, SiBBlInGS captures per-trial temporal variability and controlled cross-state structural BB adaptations, identifies state-specific vs. state-invariant components, and accommodates variability in the number and duration of observed sessions across states. We demonstrate SiBBlInGS's ability to reveal insights into complex phenomena as well as its robustness to noise and missing samples through several synthetic and real-world examples, including web search and neural data.

Learning Continuous-Time Dynamic Graphs (C-TDGs) requires accurately modeling spatio-temporal information on streams of irregularly sampled events. While many methods have been proposed recently, we find that most message passing-, recurrent- or self-attention-based methods perform poorly on *long-range* tasks. These tasks require correlating information that occurred "far" away from the current event, either spatially (higher-order node information) or along the time dimension (events occurred in the past). To address long-range dependencies, we introduce Continuous-Time Graph Anti-Symmetric Network (CTAN). Grounded within the ordinary differential equations framework, our method is designed for efficient propagation of information. In this paper, we show how CTAN's (i) long-range modeling capabilities are substantiated by theoretical findings and how (ii) its empirical performance on synthetic long-range benchmarks and real-world benchmarks is superior to other methods. Our results motivate CTAN's ability to propagate long-range information in C-TDGs as well as the inclusion of long-range tasks as part of temporal graph models evaluation.

Graph condensation aims to reduce the size of a large-scale graph dataset by synthesizing a compact counterpart without sacrificing the performance of Graph Neural Networks (GNNs) trained on it, which has shed light on reducing the computational cost for training GNNs. Nevertheless, existing methods often fall short of accurately replicating the original graph for certain datasets, thereby failing to achieve the objective of lossless condensation. To understand this phenomenon, we investigate the potential reasons and reveal that the previous state-of-the-art trajectory matching method provides biased and restricted supervision signals from the original graph when optimizing the condensed one. This significantly limits both the scale and efficacy of the condensed graph. In this paper, we make the first attempt toward *lossless graph condensation* by bridging the previously neglected supervision signals. Specifically, we employ a curriculum learning strategy to train expert trajectories with more diverse supervision signals from the original graph, and then effectively transfer the information into the condensed graph with expanding window matching. Moreover, we design a loss function to further extract knowledge from the expert trajectories. Theoretical analysis justifies the design of our method and extensive experiments verify its superiority across different datasets. Code is released at https://github.com/NUS-HPC-AI-Lab/GEOM.

Multi-agent interactions between Large Language Model (LLM) agents have shown major improvements on diverse reasoning tasks. However, these involve long generations from multiple models across several rounds, making them expensive. Moreover, these multi-agent approaches fail to provide a final, single model for efficient inference. To address this, we introduce MAGDi, a new method for structured distillation of the reasoning interactions between multiple LLMs into smaller LMs. MAGDi teaches smaller models by representing multi-agent interactions as graphs, augmenting a base student model with a graph encoder, and distilling knowledge using three objective functions: next-token prediction, a contrastive loss between correct and incorrect reasoning, and a graph-based objective to model the interaction structure. Experiments on seven widely used commonsense and math reasoning benchmarks show that MAGDi improves the reasoning capabilities of smaller models, outperforming several methods that distill from a single teacher and multiple teachers. Moreover, MAGDi also demonstrates an order of magnitude higher efficiency over its teachers. We conduct extensive analyses to show that MAGDi (1) enhances the generalizability to out-of-domain tasks, (2) scales positively with the size and strength of the base student model, and (3) obtains larger improvements (via our multi-teacher training) when applying self-consistency – an inference technique that relies on model diversity.

Predictions over graphs play a crucial role in various domains, including social networks and medicine. Graph Neural Networks (GNNs) have emerged as the dominant approach for learning on graph data. Although a graph-structure is provided as input to the GNN, in some cases the best solution can be obtained by ignoring it. While GNNs have the ability to ignore the graph-structure in such cases, it is not clear that they will. In this work, we show that GNNs actually tend to overfit the given graph-structure in the sense that they use it even when a better solution can be obtained by ignoring it. We analyze the implicit bias of gradient-descent learning of GNNs and prove that when the ground truth function does not use the graphs, GNNs are not guaranteed to learn a solution that ignores the graph, even with infinite data. We examine this phenomenon with respect to different graph distributions and find that regular graphs are more robust to this overfitting. We also prove that within the family of regular graphs, GNNs are guaranteed to extrapolate when learning with gradient descent. Finally, based on our empirical and theoretical findings, we demonstrate on real-data how regular graphs can be leveraged to reduce graph overfitting and enhance performance.

Planning is a fundamental property of human intelligence. Reasoning about asynchronous plans is challenging since it requires sequential and parallel planning to optimize time costs. Can large language models (LLMs) succeed at this task? Here, we present the first large-scale study investigating this question. We find that a representative set of closed and open-source LLMs, including GPT-4 and LLaMA-2, behave poorly when not supplied with illustrations about the task-solving process in our benchmark AsyncHow. We propose a novel technique called *Plan Like a Graph* (PLaG) that combines graphs with natural language prompts and achieves state-of-the-art results. We show that although PLaG can boost model performance, LLMs still suffer from drastic degradation when task complexity increases, highlighting the limits of utilizing LLMs for simulating digital devices. We see our study as an exciting step towards using LLMs as efficient autonomous agents. Our code and data are available at https://github.com/fangru-lin/graph-llm-asynchow-plan.

Graph augmentation has received great attention in recent years for graph contrastive learning (GCL) to learn well-generalized node/graph representations. However, mainstream GCL methods often favor randomly disrupting graphs for augmentation, which shows limited generalization and inevitably leads to the corruption of high-level graph information, i.e., the graph community. Moreover, current knowledge-based graph augmentation methods can only focus on either topology or node features, causing the model to lack robustness against various types of noise. To address these limitations, this research investigated the role of the graph community in graph augmentation and figured out its crucial advantage for learnable graph augmentation. Based on our observations, we propose a community-invariant GCL framework to maintain graph community structure during learnable graph augmentation. By maximizing the spectral changes, this framework unifies the constraints of both topology and feature augmentation, enhancing the model's robustness. Empirical evidence on 21 benchmark datasets demonstrates the exclusive merits of our framework. Code is released on Github (https://github.com/ShiyinTan/CI-GCL.git).

Pre-trained language models (LMs) are able to perform complex reasoning without explicit fine-tuning. To understand how pre-training with a next-token prediction objective contributes to the emergence of such reasoning capability, we propose that we can view an LM as deriving new conclusions by aggregating indirect reasoning paths seen at pre-training time. We found this perspective effective in two important cases of reasoning: logic reasoning with knowledge graphs (KGs) and chain-of-thought (CoT) reasoning. More specifically, we formalize the reasoning paths as random walk paths on the knowledge/reasoning graphs. Analyses of learned LM distributions suggest that a weighted sum of relevant random walk path probabilities is a reasonable way to explain how LMs reason. Experiments and analysis on multiple KG and CoT datasets reveal the effect of training on random walk paths and suggest that augmenting unlabeled random walk reasoning paths can improve real-world multi-step reasoning performance.

Graph Neural Networks (GNNs) have gained significant attention as a powerful modeling and inference method, especially for homophilic graph-structured data. To empower GNNs in heterophilic graphs, where adjacent nodes exhibit dissimilar labels or features, Signed Message Passing (SMP) has been widely adopted. However, there is a lack of theoretical and empirical analysis regarding the limitations of SMP. In this work, we unveil the potential pitfalls of SMP and their remedies. We first identify two limitations of SMP: undesirable representation update for multi-hop neighbors and vulnerability against oversmoothing issues. To overcome these challenges, we propose a novel message-passing function called Multiset to Multiset GNN (M2M-GNN). Our theoretical analyses and extensive experiments demonstrate that M2M-GNN effectively alleviates the limitations of SMP, yielding superior performance in comparison.

This paper studies clustering algorithms for dynamically evolving graphs $\{G_t\}$, in which new edges (and potential new vertices) are added into a graph, and the underlying cluster structure of the graph can gradually change. The paper proves that, under some mild condition on the cluster-structure, the clusters of the final graph $G_T$ of $n_T$ vertices at time $T$ can be well approximated by a dynamic variant of the spectral clustering algorithm. The algorithm runs in amortised update time $O(1)$ and query time $o(n_T)$. Experimental studies on both synthetic and real-world datasets further confirm the practicality of our designed algorithm.

In this work, we study the impact of QA fine-tuning data on downstream factuality. We show that fine-tuning on lesser-known facts that are poorly stored during pretraining yields significantly worse factuality than fine-tuning on well-known facts, even when all facts are seen during pretraining. We prove this phenomenon theoretically, showing that training on lesser-known facts can lead the model to ignore subject entity names and instead output a generic plausible response even when the relevant factual knowledge is encoded in the model. On three question answering benchmarks (PopQA, Entity Questions, and MMLU) and two language models (Llama-2-7B and Mistral-7B), we find that (i) finetuning on a completely factual but lesser-known subset of the data deteriorates downstream factuality (5-10%) and (ii) finetuning on a subset of better-known examples matches or outperforms finetuning on the entire dataset. Ultimately, our results shed light on the interaction between pretrained knowledge and finetuning data and demonstrate the importance of taking into account how facts are stored in the pretrained model when fine-tuning for knowledge-intensive tasks.

In this paper, we propose a novel generalization of rested and restless bandits where the evolution of the arms' expected rewards is governed by a graph defined over the arms. An edge connecting a pair of arms $(i,j)$ represents the fact that a pull of arm $i$ *triggers* the evolution of arm $j$, and vice versa. Interestingly, rested and restless bandits are both special cases of our model for some suitable (degenerate) graphs. Still, the model can represent way more general and interesting scenarios. We first tackle the problem of computing the optimal policy when no specific structure is assumed on the graph, showing that it is NP-hard. Then, we focus on a specific structure forcing the graph to be composed of a set of fully connected subgraphs (i.e., cliques), and we prove that the optimal policy can be easily computed in closed form. Then, we move to the learning problem presenting regret minimization algorithms for deterministic and stochastic cases. Our regret bounds highlight the complexity of the learning problem by incorporating instance-dependent terms that encode specific properties of the underlying graph structure. Moreover, we illustrate how the knowledge of the underlying graph is not necessary for achieving the no-regret property.