论文检索

Online Learning

Zijian Gao, Xingxing Zhang, Kele Xu, Xinjun Mao, Huaimin Wang

Continual learning (CL) empowers pre-trained vision-language (VL) models to efficiently adapt to a sequence of downstream tasks. However, these models often encounter challenges in retaining previously acquired skills due to parameter shifts and limited access to historical data. In response, recent efforts focus on devising specific frameworks and various replay strategies, striving for a typical learning-forgetting trade-off. Surprisingly, both our empirical research and theoretical analysis demonstrate that the stability of the model in consecutive zero-shot predictions serves as a reliable indicator of its anti-forgetting capabilities for previously learned tasks. Motivated by these insights, we develop a novel replay-free CL method named ZAF (Zero-shot Antidote to Forgetting), which preserves acquired knowledge through a zero-shot stability regularization applied to wild data in a plug-and-play manner. To enhance efficiency in adapting to new tasks and seamlessly access historical models, we introduce a parameter-efficient EMA-LoRA neural architecture based on the Exponential Moving Average (EMA). ZAF utilizes new data for low-rank adaptation (LoRA), complemented by a zero-shot antidote on wild data, effectively decoupling learning from forgetting. Our extensive experiments demonstrate ZAF's superior performance and robustness in pre-trained models across various continual VL concept learning tasks, achieving leads of up to 3.70\%, 4.82\%, and 4.38\%, along with at least a 10x acceleration in training speed on three benchmarks, respectively. Additionally, our zero-shot antidote significantly reduces forgetting in existing models by at least 6.37\%. Our code is available at https://github.com/Zi-Jian-Gao/Stabilizing-Zero-Shot-Prediction-ZAF.

Miscellaneous Aspects of Machine Learning · Supervised Learning

Thomas Nagler, Lennart Schneider, Bernd Bischl, Matthias Feurer

Hyperparameter optimization is crucial for obtaining peak performance of machine learning models. The standard protocol evaluates various hyperparameter configurations using a resampling estimate of the generalization error to guide optimization and select a final hyperparameter configuration. Without much evidence, paired resampling splits, i.e., either a fixed train-validation split or a fixed cross-validation scheme, are often recommended. We show that, surprisingly, reshuffling the splits for every configuration often improves the final model's generalization performance on unseen data. Our theoretical analysis explains how reshuffling affects the asymptotic behavior of the validation loss surface and provides a bound on the expected regret in the limiting regime. This bound connects the potential benefits of reshuffling to the signal and noise characteristics of the underlying optimization problem. We confirm our theoretical results in a controlled simulation study and demonstrate the practical usefulness of reshuffling in a large-scale, realistic hyperparameter optimization experiment. While reshuffling leads to test performances that are competitive with using fixed splits, it drastically improves results for a single train-validation holdout protocol and can often make holdout become competitive with standard CV while being computationally cheaper.

Dmitry Shribak, Chen-Xiao Gao, Yitong Li, Chenjun Xiao, Bo Dai

Diffusion-based models have achieved notable empirical successes in reinforcement learning (RL) due to their expressiveness in modeling complex distributions. Despite existing methods being promising, the key challenge of extending existing methods for broader real-world applications lies in the computational cost at inference time, i.e., sampling from a diffusion model is considerably slow as it often requires tens to hundreds of iterations to generate even one sample. To circumvent this issue, we propose to leverage the flexibility of diffusion models for RL from a representation learning perspective. In particular, by exploiting the connection between diffusion models and energy-based models, we develop Diffusion Spectral Representation (Diff-SR), a coherent algorithm framework that enables extracting sufficient representations for value functions in Markov decision processes (MDP) and partially observable Markov decision processes (POMDP). We further demonstrate how Diff-SR facilitates efficient policy optimization and practical algorithms while explicitly bypassing the difficulty and inference cost of sampling from the diffusion model. Finally, we provide comprehensive empirical studies to verify the benefits of Diff-SR in delivering robust and advantageous performance across various benchmarks with both fully and partially observable settings.

Computer Vision · Classification

Kunyu Peng, Di Wen, Kailun Yang, Ao Luo, Yufan Chen, Jia Fu, M. Saquib Sarfraz, Alina Roitberg, Rainer Stiefelhagen

In Open-Set Domain Generalization (OSDG), the model is exposed to both new variations of data appearance (domains) and open-set conditions, where both known and novel categories are present at test time. The challenges of this task arise from the dual need to generalize across diverse domains and accurately quantify category novelty, which is critical for applications in dynamic environments. Recently, meta-learning techniques have demonstrated superior results in OSDG, effectively orchestrating the meta-train and -test tasks by employing varied random categories and predefined domain partition strategies. These approaches prioritize a well-designed training schedule over traditional methods that focus primarily on data augmentation and the enhancement of discriminative feature learning. The prevailing meta-learning models in OSDG typically utilize a predefined sequential domain scheduler to structure data partitions. However, a crucial aspect that remains inadequately explored is the influence brought by strategies of domain schedulers during training. In this paper, we observe that an adaptive domain scheduler benefits more in OSDG compared with prefixed sequential and random domain schedulers. We propose the Evidential Bi-Level Hardest Domain Scheduler (EBiL-HaDS) to achieve an adaptive domain scheduler. This method strategically sequences domains by assessing their reliabilities in utilizing a follower network, trained with confidence scores learned in an evidential manner, regularized by max rebiasing discrepancy, and optimized in a bilevel manner. We verify our approach on three OSDG benchmarks, i.e., PACS, DigitsDG, and OfficeHome. The results show that our method substantially improves OSDG performance and achieves more discriminative embeddings for both the seen and unseen categories, underscoring the advantage of a judicious domain scheduler for the generalizability to unseen domains and unseen categories. The source code is publicly available at https://github.com/KPeng9510/EBiL-HaDS.

Language · Generation

Chufan Shi, Cheng Yang, Xinyu Zhu, Jiahao Wang, Taiqiang Wu, Siheng Li, Deng Cai, Yujiu Yang, Yu Meng

Mixture-of-Experts (MoE) has emerged as a prominent architecture for scaling model size while maintaining computational efficiency. In MoE, each token in the input sequence activates a different subset of experts determined by a routing mechanism. However, the unchosen experts in MoE models do not contribute to the output, potentially leading to underutilization of the model's capacity.In this work, we first conduct exploratory studies to demonstrate that increasing the number of activated experts does not necessarily improve and can even degrade the output quality. Then, we show that output distributions from an MoE model using different routing strategies substantially differ, indicating that different experts do not always act synergistically. Motivated by these findings, we propose **S**elf-**C**ontrast **M**ixture-**o**f-**E**xperts (SCMoE), a training-free strategy that utilizes unchosen experts in a self-contrast manner during inference. In SCMoE, the next-token probabilities are determined by contrasting the outputs from strong and weak activation using the same MoE model.Our method is conceptually simple and computationally lightweight, as it incurs minimal latency compared to greedy decoding. Experiments on several benchmarks (GSM8K, StrategyQA, MBPP and HumanEval) demonstrate that SCMoE can consistently enhance Mixtral 8x7B’s reasoning capability across various domains. For example, it improves the accuracy on GSM8K from 61.79 to 66.94. Moreover, combining SCMoE with self-consistency yields additional gains, increasing major@20 accuracy from 75.59 to 78.31.

Neuroscience, Cognitive Science

Matthijs Pals, A Erdem Sağtekin, Felix Pei, Manuel Gloeckler, Jakob H Macke

A central aim in computational neuroscience is to relate the activity of large populations of neurons to an underlying dynamical system. Models of these neural dynamics should ideally be both interpretable and fit the observed data well. Low-rank recurrent neural networks (RNNs) exhibit such interpretability by having tractable dynamics. However, it is unclear how to best fit low-rank RNNs to data consisting of noisy observations of an underlying stochastic system. Here, we propose to fit stochastic low-rank RNNs with variational sequential Monte Carlo methods. We validate our method on several datasets consisting of both continuous and spiking neural data, where we obtain lower dimensional latent dynamics than current state of the art methods. Additionally, for low-rank models with piecewise linear nonlinearities, we show how to efficiently identify all fixed points in polynomial rather than exponential cost in the number of units, making analysis of the inferred dynamics tractable for large RNNs. Our method both elucidates the dynamical systems underlying experimental recordings and provides a generative model whose trajectories match observed variability.

Computer Vision

Xuan Huang, Hanhui Li, Wanquan Liu, Xiaodan Liang, Yiqiang Yan, Yuhao Cheng, CHENQIANG GAO

In this paper, we propose to create animatable avatars for interacting hands with 3D Gaussian Splatting (GS) and single-image inputs. Existing GS-based methods designed for single subjects often yield unsatisfactory results due to limited input views, various hand poses, and occlusions. To address these challenges, we introduce a novel two-stage interaction-aware GS framework that exploits cross-subject hand priors and refines 3D Gaussians in interacting areas. Particularly, to handle hand variations, we disentangle the 3D presentation of hands into optimization-based identity maps and learning-based latent geometric features and neural texture maps. Learning-based features are captured by trained networks to provide reliable priors for poses, shapes, and textures, while optimization-based identity maps enable efficient one-shot fitting of out-of-distribution hands. Furthermore, we devise an interaction-aware attention module and a self-adaptive Gaussian refinement module. These modules enhance image rendering quality in areas with intra- and inter-hand interactions, overcoming the limitations of existing GS-based methods. Our proposed method is validated via extensive experiments on the large-scale InterHand2.6M dataset, and it significantly improves the state-of-the-art performance in image quality. Code and models will be released upon acceptance.

Theory · Reinforcement Learning and Planning

Fivos Kalogiannis, Jingming Yan, Ioannis Panageas

We study the problem of learning a Nash equilibrium (NE) in Markov games which is a cornerstone in multi-agent reinforcement learning (MARL). In particular, we focus on infinite-horizon adversarial team Markov games (ATMGs) in which agents that share a common reward function compete against a single opponent, *the adversary*. These games unify two-player zero-sum Markov games and Markov potential games, resulting in a setting that encompasses both collaboration and competition. Kalogiannis et al. (2023) provided an efficient equilibrium computation algorithm for ATMGs which presumes knowledge of the reward and transition functions and has no sample complexity guarantees. We contribute a learning algorithm that utilizes MARL policy gradient methods with iteration and sample complexity that is polynomial in the approximation error $\epsilon$ and the natural parameters of the ATMG, resolving the main caveats of the solution by (Kalogiannis et al., 2023). It is worth noting that previously, the existence of learning algorithms for NE was known for Markov two-player zero-sum and potential games but not for ATMGs. Seen through the lens of min-max optimization, computing a NE in these games consists a nonconvex--nonconcave saddle-point problem. Min-max optimization has received an extensive study. Nevertheless, the case of nonconvex--nonconcave landscapes remains elusive: in full generality, finding saddle-points is computationally intractable (Daskalakis et al., 2021). We circumvent the aforementioned intractability by developing techniques that exploit the hidden structure of the objective function via a nonconvex--concave reformulation. However, this introduces a challenge of a feasibility set with coupled constraints. We tackle these challenges by establishing novel techniques for optimizing weakly-smooth nonconvex functions, extending the framework of (Devolder et al., 2014).

Causality

Mohammad Shahverdikondori, Ehsan Mokhtarian, Negar Kiyavash

Causal discovery is essential for understanding relationships among variables of interest in many scientific domains. In this paper, we focus on permutation-based methods for learning causal graphs in Linear Gaussian Acyclic Models (LiGAMs), where the permutation encodes a causal ordering of the variables. Existing methods in this setting are not scalable due to their high computational complexity. These methods are comprised of two main components: (i) constructing a specific DAG, $\mathcal{G}^\pi$, for a given permutation $\pi$, which represents the best structure that can be learned from the available data while adhering to $\pi$, and (ii) searching over the space of permutations (i.e., causal orders) to minimize the number of edges in $\mathcal{G}^\pi$. We introduce QWO, a novel approach that significantly enhances the efficiency of computing $\mathcal{G}^\pi$ for a given permutation $\pi$. QWO has a speed-up of $O(n^2)$ ($n$ is the number of variables) compared to the state-of-the-art BIC-based method, making it highly scalable. We show that our method is theoretically sound and can be integrated into existing search strategies such as GRASP and hill-climbing-based methods to improve their performance.

Reinforcement Learning · Deep RL

Xianghua Zeng, Hao Peng, Angsheng Li

Traditional information theory provides a valuable foundation for Reinforcement Learning (RL), particularly through representation learning and entropy maximiza tion for agent exploration. However, existing methods primarily concentrate on modeling the uncertainty associated with RL’s random variables, neglecting the in herent structure within the state and action spaces. In this paper, we propose a novel Structural Information principles-based Effective Exploration framework, namely SI2E. Structural mutual information between two variables is defined to address the single-variable limitation in structural information, and an innovative embedding principle is presented to capture dynamics-relevant state-action representations. The SI2E analyzes value differences in the agent’s policy between state-action pairs and minimizes structural entropy to derive the hierarchical state-action struc ture, referred to as the encoding tree. Under this tree structure, value-conditional structural entropy is defined and maximized to design an intrinsic reward mechanism that avoids redundant transitions and promotes enhanced coverage in the state-action space. Theoretical connections are established between SI2E and classical information-theoretic methodologies, highlighting our framework’s rationality and advantage. Comprehensive evaluations in the MiniGrid, MetaWorld, and DeepMind Control Suite benchmarks demonstrate that SI2E significantly outperforms state-of-the-art exploration baselines regarding final performance and sample efficiency, with maximum improvements of 37.63% and 60.25%, respectively.

Theory · Reinforcement Learning and Planning

Yujia Jin, Ishani Karmarkar, Aaron Sidford, Jiayi Wang

We provide faster randomized algorithms for computing an $\epsilon$-optimal policy in a discounted Markov decision process with $A_{\text{tot}}$-state-action pairs, bounded rewards, and discount factor $\gamma$. We provide an $\tilde{O}(A_{\text{tot}}[(1 - \gamma)^{-3}\epsilon^{-2} + (1 - \gamma)^{-2}])$-time algorithm in the sampling setting, where the probability transition matrix is unknown but accessible through a generative model which can be queried in $\tilde{O}(1)$-time, and an $\tilde{O}(s + (1-\gamma)^{-2})$-time algorithm in the offline setting where the probability transition matrix is known and $s$-sparse. These results improve upon the prior state-of-the-art which either ran in $\tilde{O}(A_{\text{tot}}[(1 - \gamma)^{-3}\epsilon^{-2} + (1 - \gamma)^{-3}])$ time [Sidford, Wang, Wu, Ye 2018] in the sampling setting, $\tilde{O}(s + A_{\text{tot}} (1-\gamma)^{-3})$ time [Sidford, Wang, Wu, Yang, Ye 2018] in the offline setting, or time at least quadratic in the number of states using interior point methods for linear programming. We achieve our results by building upon prior stochastic variance-reduced value iteration methods [Sidford, Wang, Wu, Yang, Ye 2018]. We provide a variant that carefully truncates the progress of its iterates to improve the variance of new variance-reduced sampling procedures that we introduce to implement the steps. Our method is essentially model-free and can be implemented in $\tilde{O}(A_{\text{tot}})$-space when given generative model access. Consequently, our results take a step in closing the sample-complexity gap between model-free and model-based methods.

Causality

Alexander Levis, Gabriel Loewinger, Francisco Pereira

Optogenetics is widely used to study the effects of neural circuit manipulation on behavior. However, the paucity of causal inference methodological work on this topic has resulted in analysis conventions that discard information, and constrain the scientific questions that can be posed. To fill this gap, we introduce a nonparametric causal inference framework for analyzing "closed-loop" designs, which use dynamic policies that assign treatment based on covariates. In this setting, standard methods can introduce bias and occlude causal effects. Building on the sequentially randomized experiments literature in causal inference, our approach extends history-restricted marginal structural models for dynamic regimes. In practice, our framework can identify a wide range of causal effects of optogenetics on trial-by-trial behavior, such as, fast/slow-acting, dose-response, additive/antagonistic, and floor/ceiling. Importantly, it does so without requiring negative controls, and can estimate how causal effect magnitudes evolve across time points. From another view, our work extends "excursion effect" methods---popular in the mobile health literature---to enable estimation of causal contrasts for treatment sequences greater than length one, in the presence of positivity violations. We derive rigorous statistical guarantees, enabling hypothesis testing of these causal effects. We demonstrate our approach on data from a recent study of dopaminergic activity on learning, and show how our method reveals relevant effects obscured in standard analyses.

Computer Vision

Sirui Xu, ziyin wang, Yu-Xiong Wang, Liangyan Gui

Text-conditioned human motion generation has experienced significant advancements with diffusion models trained on extensive motion capture data and corresponding textual annotations. However, extending such success to 3D dynamic human-object interaction (HOI) generation faces notable challenges, primarily due to the lack of large-scale interaction data and comprehensive descriptions that align with these interactions. This paper takes the initiative and showcases the potential of generating human-object interactions without direct training on text-interaction pair data. Our key insight in achieving this is that interaction semantics and dynamics can be decoupled. Being unable to learn interaction semantics through supervised training, we instead leverage pre-trained large models, synergizing knowledge from a large language model and a text-to-motion model. While such knowledge offers high-level control over interaction semantics, it cannot grasp the intricacies of low-level interaction dynamics. To overcome this issue, we introduce a world model designed to comprehend simple physics, modeling how human actions influence object motion. By integrating these components, our novel framework, InterDreamer, is able to generate text-aligned 3D HOI sequences without relying on paired text-interaction data. We apply InterDreamer to the BEHAVE, OMOMO, and CHAIRS datasets, and our comprehensive experimental analysis demonstrates its capability to generate realistic and coherent interaction sequences that seamlessly align with the text directives.

Deep Learning · Robustness

Yitao Xu, Tong Zhang, Sabine Süsstrunk

Vision Transformers (ViTs) demonstrate remarkable performance in image classification through visual-token interaction learning, particularly when equipped with local information via region attention or convolutions. Although such architectures improve the feature aggregation from different granularities, they often fail to contribute to the robustness of the networks. Neural Cellular Automata (NCA) enables the modeling of global visual-token representations through local interactions, with its training strategies and architecture design conferring strong generalization ability and robustness against noisy input. In this paper, we propose Adaptor Neural Cellular Automata (AdaNCA) for Vision Transformers that uses NCA as plug-and-play adaptors between ViT layers, thus enhancing ViT's performance and robustness against adversarial samples as well as out-of-distribution inputs. To overcome the large computational overhead of standard NCAs, we propose Dynamic Interaction for more efficient interaction learning. Using our analysis of AdaNCA placement and robustness improvement, we also develop an algorithm for identifying the most effective insertion points for AdaNCA. With less than a 3% increase in parameters, AdaNCA contributes to more than 10% absolute improvement in accuracy under adversarial attacks on the ImageNet1K benchmark. Moreover, we demonstrate with extensive evaluations across eight robustness benchmarks and four ViT architectures that AdaNCA, as a plug-and-play module, consistently improves the robustness of ViTs.

Optimization

Wei Dong, Yuan Sun, Yiting Yang, Xing Zhang, Zhijun Lin, Qingsen Yan, Haokui Zhang, Peng Wang, Yang Yang, Hengtao Shen

A common strategy for Parameter-Efficient Fine-Tuning (PEFT) of pre-trained Vision Transformers (ViTs) involves adapting the model to downstream tasks by learning a low-rank adaptation matrix. This matrix is decomposed into a product of down-projection and up-projection matrices, with the bottleneck dimensionality being crucial for reducing the number of learnable parameters, as exemplified by prevalent methods like LoRA and Adapter. However, these low-rank strategies typically employ a fixed bottleneck dimensionality, which limits their flexibility in handling layer-wise variations. To address this limitation, we propose a novel PEFT approach inspired by Singular Value Decomposition (SVD) for representing the adaptation matrix. SVD decomposes a matrix into the product of a left unitary matrix, a diagonal matrix of scaling values, and a right unitary matrix. We utilize Householder transformations to construct orthogonal matrices that efficiently mimic the unitary matrices, requiring only a vector. The diagonal values are learned in a layer-wise manner, allowing them to flexibly capture the unique properties of each layer. This approach enables the generation of adaptation matrices with varying ranks across different layers, providing greater flexibility in adapting pre-trained models. Experiments on standard downstream vision tasks demonstrate that our method achieves promising fine-tuning performance.

Privacy

Xingyu Zhou, Komo(Wei) ZHANG

We study the interplay between local differential privacy (LDP) and robustness to Huber corruption and possibly heavy-tailed rewards in the context of multi-armed bandits (MABs). We consider two different practical settings: LDP-then-Corruption (LTC) where each user's locally private response might be further corrupted during the data collection process, and Corruption-then-LDP (CTL) where each user's raw data may be corrupted such that the LDP mechanism will only be applied to the corrupted data. To start with, we present the first tight characterization of the mean estimation error in high probability under both LTC and CTL settings. Leveraging this new result, we then present an almost tight characterization (up to log factor) of the minimax regret in online MABs and sub-optimality in offline MABs under both LTC and CTL settings, respectively. Our theoretical results in both settings are also corroborated by a set of systematic simulations. One key message in this paper is that LTC is a more difficult setting that leads to a worse performance guarantee compared to the CTL setting (in the minimax sense). Our sharp understanding of LTC and CTL also naturally allows us to give the first tight performance bounds for the most practical setting where corruption could happen both before and after the LDP mechanism. As an important by-product, we also give the first correct and tight regret bound for locally private and heavy-tailed online MABs, i.e., without Huber corruption, by identifying a fundamental flaw in the state-of-the-art.

Optimization · Bilevel optimization

Junyi Li, Heng Huang

Bilevel Optimization has experienced significant advancements recently with the introduction of new efficient algorithms. Mirroring the success in single-level optimization, stochastic gradient-based algorithms are widely used in bilevel optimization. However, a common limitation in these algorithms is the presumption of independent sampling, which can lead to increased computational costs due to the unique hyper-gradient structure in bilevel problems. To address this challenge, we study the example-selection strategy for bilevel optimization in this work. More specifically, we introduce a without-replacement sampling based algorithm which achieves a faster convergence rate compared to its counterparts that rely on independent sampling. Beyond the standard bilevel optimization formulation, we extend our discussion to conditional bilevel optimization and also two special cases: minimax and compositional optimization. Finally, we validate our algorithms over both synthetic and real-world applications. Numerical results clearly showcase the superiority of our algorithms.

Generative Models · Diffusion models

Chaoyi Pan, Zeji Yi, Guanya Shi, Guannan Qu

Recent advances in diffusion models have demonstrated their strong capabilities in generating high-fidelity samples from complex distributions through an iterative refinement process. Despite the empirical success of diffusion models in motion planning and control, the model-free nature of these methods does not leverage readily available model information and limits their generalization to new scenarios beyond the training data (e.g., new robots with different dynamics). In this work, we introduce Model-Based Diffusion (MBD), an optimization approach using the diffusion process to solve trajectory optimization (TO) problems without data. The key idea is to explicitly compute the score function by leveraging the model information in TO problems, which is why we refer to our approach as model-based diffusion. Moreover, although MBD does not require external data, it can be naturally integrated with data of diverse qualities to steer the diffusion process. We also reveal that MBD has interesting connections to sampling-based optimization. Empirical evaluations show that MBD outperforms state-of-the-art reinforcement learning and sampling-based TO methods in challenging contact-rich tasks. Additionally, MBD’s ability to integrate with data enhances its versatility and practical applicability, even with imperfect and infeasible data (e.g., partial-state demonstrations for high-dimensional humanoids), beyond the scope of standard diffusion models. Videos and codes are available in the supplementary materials.

Causality

Yixin Ren, Yewei Xia, Hao Zhang, Jihong Guan, Shuigeng Zhou

We propose a novel method to efficiently learn significant Fourier feature pairs for maximizing the power of Hilbert-Schmidt Independence Criterion~(HSIC) based independence tests. We first reinterpret HSIC in the frequency domain, which reveals its limited discriminative power due to the inability to adapt to specific frequency-domain features under the current inflexible configuration. To remedy this shortcoming, we introduce a module of learnable Fourier features, thereby developing a new criterion. We then derive a finite sample estimate of the test power by modeling the behavior of the criterion, thus formulating an optimization objective for significant Fourier feature pairs learning. We show that this optimization objective can be computed in linear time (with respect to the sample size $n$), which ensures fast independence tests. We also prove the convergence property of the optimization objective and establish the consistency of the independence tests. Extensive empirical evaluation on both synthetic and real datasets validates our method's superiority in effectiveness and efficiency, particularly in handling high-dimensional data and dealing with large-scale scenarios.

Generative Models · In Context Learning

David Samuel

While in-context learning is commonly associated with causal language models, such as GPT, we demonstrate that this capability also 'emerges' in masked language models. Through an embarrassingly simple inference technique, we enable an existing masked model, DeBERTa, to perform generative tasks without additional training or architectural changes. Our evaluation reveals that the masked and causal language models behave very differently, as they clearly outperform each other on different categories of tasks. These complementary strengths suggest that the field's focus on causal models for in-context learning may be limiting – both architectures can develop these capabilities, but with distinct advantages; pointing toward promising hybrid approaches that combine the strengths of both objectives.