arXiv subjects

Liangze Jiang

Publications and source records attributed to Liangze Jiang.

12 recordsLinked to original sources

arXiv2026

Can Transformers Really Do It All? On the Compatibility of Inductive Biases Across Tasks

Transformers are remarkably versatile and their design is largely consistent across a variety of applications. But are they optimal for any given task or dataset? The answer may be key for pushing AI beyond merely scaling current designs. *Method.* We present a method to optimize a transformer architecture for a given dataset, which we use as a tool to study optimal task-specific inductive biases. This method replaces the most important non-linearities (GeLUs,;softmax) with functions learned on held-out data. We then train the resulting architectures on other datasets, as a way to evaluate the compatibility between pairs of tasks. Findings. On algorithmic toy tasks, we identify new architectures with dramatic improvements in learning speed, in- and out-of-distribution generalization, and stability across seeds. The new designs prove very task-specific however, and indicate that these tasks require inductive biases very different from those of standard transformers. On code and language modeling datasets, we also find architectures with consistent, yet smaller improvements. These designs transfer much better across datasets and domains (English & computer code). Implications. Our results show that standard transformers are rarely a local optimum in the space of architectures. Simple alternatives can perform much better but sacrifice universality. This suggests that there may be room for improved architectures that better support multiple capabilities simultaneously, such as fluency and robust reasoning.

cs.LG↗

arXiv2026

Procedural Pretraining: Warming Up Language Models with Abstract Data

Pretraining language models directly on web-scale corpora is the de facto paradigm. We study an alternative where the model is initially exposed to abstract structured data to ease the subsequent acquisition of rich semantic knowledge, much like humans learning simple logic and mathematics before higher reasoning. We focus on procedural data, generated by formal languages and other simple algorithms, as such abstract data. We first diagnose the algorithmic skills that different forms of procedural data can improve, often significantly. For example, the accuracy of context recall (Needle-in-a-haystack) jumps from 10 to 98% when a model is pretrained on Dyck sequences (balanced brackets). Second, we study how these gains are reflected in pretraining larger models (up to 1.3B). We find that front-loading as little as 0.1 to 0.3% procedural data significantly outperforms standard pretraining on natural language, code, and informal mathematics (C4, CodeParrot, and DeepMind-Math datasets). Notably, this also enables the models to reach the same loss value with only 55/67/86% of the original data and thus a comparable reduction in FLOPs. Third, we explore the mechanisms behind the benefits and find that procedural pretraining instills non-trivial structure in both attention and MLP layers. The former is particularly important for structured domains (e.g. code), and the latter for language. Finally, we lay a path for combining multiple forms of procedural data. Our results show that procedural pretraining is a simple, lightweight means of improving performance and accelerating language model pretraining, ultimately suggesting the promise of disentangling knowledge acquisition from reasoning in LLMs.

cs.CL↗

arXiv2025

Meta-RL Induces Exploration in Language Agents

Reinforcement learning (RL) has enabled the training of large language model (LLM) agents to interact with the environment and to solve multi-turn long-horizon tasks. However, the RL-trained agents often struggle in tasks that require active exploration and fail to efficiently adapt from trial-and-error experiences. In this paper, we present LaMer, a general Meta-RL framework that enables LLM agents to actively explore and learn from the environment feedback at test time. LaMer consists of two key components: (i) a cross-episode training framework to encourage exploration and long-term rewards optimization; and (ii) in-context policy adaptation via reflection, allowing the agent to adapt their policy from task feedback signal without gradient update. Experiments across diverse environments show that LaMer significantly improves performance over RL baselines, with 11%, 14%, and 19% performance gains on Sokoban, MineSweeper and Webshop, respectively. Moreover, LaMer also demonstrates better generalization to more challenging or previously unseen tasks compared to the RL-trained agents. Overall, our results demonstrate that Meta-RL provides a principled approach to induce exploration in language agents, enabling more robust adaptation to novel environments through learned exploration strategies.

cs.LG↗

arXiv2025

Can You Learn to See Without Images? Procedural Warm-Up for Vision Transformers

Transformers are remarkably versatile, suggesting the existence of generic inductive biases beneficial across modalities. In this work, we explore a new way to instil such biases in vision transformers (ViTs) through pretraining on procedurally generated data devoid of visual or semantic content. We generate this data with simple algorithms such as formal grammars, so the results bear no relationship to either natural or synthetic images. We use this procedurally generated data to pretrain ViTs in a warm-up phase that bypasses their visual patch embedding mechanisms, thus encouraging the models to internalise abstract computational priors. When followed by standard image-based training, this warm-up significantly improves data efficiency, convergence speed, and downstream performance. On ImageNet-1K, for example, allocating just 1% of the training budget to procedural data improves final accuracy by over 1.7%. In terms of its effect on performance, 1% procedurally generated data is thus equivalent to 28% of the ImageNet-1K data. These findings suggest a promising path toward new data-efficient and domain-agnostic pretraining strategies.

cs.CV↗

arXiv2025

OOD-Chameleon: Is Algorithm Selection for OOD Generalization Learnable?

Out-of-distribution (OOD) generalization is challenging because distribution shifts come in many forms. Numerous algorithms exist to address specific settings, but choosing the right training algorithm for the right dataset without trial and error is difficult. Indeed, real-world applications often involve multiple types and combinations of shifts that are hard to analyze theoretically. Method. This work explores the possibility of learning the selection of a training algorithm for OOD generalization. We propose a proof of concept (OOD-Chameleon) that formulates the selection as a multi-label classification over candidate algorithms, trained on a dataset of datasets representing a variety of shifts. We evaluate the ability of OOD-Chameleon to rank algorithms on unseen shifts and datasets based only on dataset characteristics, i.e., without training models first, unlike traditional model selection. Findings. Extensive experiments show that the learned selector identifies high-performing algorithms across synthetic, vision, and language tasks. Further inspection shows that it learns non-trivial decision rules, which provide new insights into the applicability of existing algorithms. Overall, this new approach opens the possibility of better exploiting and understanding the plethora of existing algorithms for OOD generalization.

cs.LG↗

arXiv2025

Transformers Pretrained on Procedural Data Contain Modular Structures for Algorithmic Reasoning

Pretraining on large, semantically rich datasets is key for developing language models. Surprisingly, recent studies have shown that even synthetic data, generated procedurally through simple semantic-free algorithms, can yield some of the same benefits as natural language pretraining. It is unclear what specific capabilities such simple synthetic data instils in a model, where these capabilities reside in the architecture, and how they manifest within its weights. In this short paper, we identify several beneficial forms of procedural data, together with specific algorithmic reasoning skills that improve in small transformers. Our core finding is that different procedural rules instil distinct but complementary inductive structures in the model. With extensive ablations and partial-transfer experiments, we discover that these structures reside in different parts of the model. Attention layers often carry the most transferable information, but some pretraining rules impart useful structure to MLP blocks instead. Most interestingly, the structures induced by multiple rules can be composed to jointly reinforce multiple capabilities. These results suggest an exciting possibility of disentangling the acquisition of knowledge from reasoning in language models, with the goal of improving their robustness and data efficiency.

cs.LG↗

arXiv2025

Do We Always Need the Simplicity Bias? Looking for Optimal Inductive Biases in the Wild

Neural architectures tend to fit their data with relatively simple functions. This "simplicity bias" is widely regarded as key to their success. This paper explores the limits of this principle. Building on recent findings that the simplicity bias stems from ReLU activations [96], we introduce a method to meta-learn new activation functions and inductive biases better suited to specific tasks. Findings: We identify multiple tasks where the simplicity bias is inadequate and ReLUs suboptimal. In these cases, we learn new activation functions that perform better by inducing a prior of higher complexity. Interestingly, these cases correspond to domains where neural networks have historically struggled: tabular data, regression tasks, cases of shortcut learning, and algorithmic grokking tasks. In comparison, the simplicity bias induced by ReLUs proves adequate on image tasks where the best learned activations are nearly identical to ReLUs and GeLUs. Implications: Contrary to popular belief, the simplicity bias of ReLU networks is not universally useful. It is near-optimal for image classification, but other inductive biases are sometimes preferable. We showed that activation functions can control these inductive biases, but future tailored architectures might provide further benefits. Advances are still needed to characterize a model's inductive biases beyond "complexity", and their adequacy with the data.

cs.LG↗

arXiv2024

Distributed Adaptive Time-Varying Optimization with Global Asymptotic Convergence

In this note, we study distributed time-varying optimization for a multi-agent system. We first focus on a class of time-varying quadratic cost functions, and develop a new distributed algorithm that integrates an average estimator and an adaptive optimizer, with both bridged by a Dead Zone Algorithm. Based on a composite Lyapunov function and finite escape-time analysis, we prove the closed-loop global asymptotic convergence to the optimal solution under mild assumptions. Particularly, the introduction of the estimator relaxes the requirement for the Hessians of cost functions, and the integrated design eliminates the waiting time required in the relevant literature for estimating global parameter during algorithm implementation. We then extend this result to a more general class of time-varying cost functions. Two examples are used to verify the proposed designs.

eess.SY↗

arXiv2024

Distributed Continuous-Time Optimization with Uncertain Time-Varying Quadratic Cost Functions

This paper studies distributed continuous-time optimization for time-varying quadratic cost functions with uncertain parameters. We first propose a centralized adaptive optimization algorithm using partial information of the cost function. It can be seen that even if there are uncertain parameters in the cost function, exact optimization can still be achieved. To solve this problem in a distributed manner when different local cost functions have identical Hessians, we propose a novel distributed algorithm that cascades the fixed-time average estimator and the distributed optimizer. We remove the requirement for the upper bounds of certain complex functions by integrating state-based gains in the proposed design. We further extend this result to address the distributed optimization where the time-varying cost functions have nonidentical Hessians. We prove the convergence of all the proposed algorithms in the global sense. Numerical examples verify the proposed algorithms.

eess.SY↗

arXiv2024

Unraveling the Key Components of OOD Generalization via Diversification

Supervised learning datasets may contain multiple cues that explain the training set equally well, i.e., learning any of them would lead to the correct predictions on the training data. However, many of them can be spurious, i.e., lose their predictive power under a distribution shift and consequently fail to generalize to out-of-distribution (OOD) data. Recently developed "diversification" methods (Lee et al., 2023; Pagliardini et al., 2023) approach this problem by finding multiple diverse hypotheses that rely on different features. This paper aims to study this class of methods and identify the key components contributing to their OOD generalization abilities. We show that (1) diversification methods are highly sensitive to the distribution of the unlabeled data used for diversification and can underperform significantly when away from a method-specific sweet spot. (2) Diversification alone is insufficient for OOD generalization. The choice of the used learning algorithm, e.g., the model's architecture and pretraining, is crucial. In standard experiments (classification on Waterbirds and Office-Home datasets), using the second-best choice leads to an up to 20\% absolute drop in accuracy. (3) The optimal choice of learning algorithm depends on the unlabeled data and vice versa i.e. they are co-dependent. (4) Finally, we show that, in practice, the above pitfalls cannot be alleviated by increasing the number of diverse hypotheses, the major feature of diversification methods. These findings provide a clearer understanding of the critical design factors influencing the OOD generalization abilities of diversification methods. They can guide practitioners in how to use the existing methods best and guide researchers in developing new, better ones.

cs.LG↗

arXiv2023

TF-GNN: Graph Neural Networks in TensorFlow

TensorFlow-GNN (TF-GNN) is a scalable library for Graph Neural Networks in TensorFlow. It is designed from the bottom up to support the kinds of rich heterogeneous graph data that occurs in today's information ecosystems. In addition to enabling machine learning researchers and advanced developers, TF-GNN offers low-code solutions to empower the broader developer community in graph learning. Many production models at Google use TF-GNN, and it has been recently released as an open source project. In this paper we describe the TF-GNN data model, its Keras message passing API, and relevant capabilities such as graph sampling and distributed training.

cs.LG↗

arXiv2023

Test-Time Robust Personalization for Federated Learning

Federated Learning (FL) is a machine learning paradigm where many clients collaboratively learn a shared global model with decentralized training data. Personalized FL additionally adapts the global model to different clients, achieving promising results on consistent local training and test distributions. However, for real-world personalized FL applications, it is crucial to go one step further: robustifying FL models under the evolving local test set during deployment, where various distribution shifts can arise. In this work, we identify the pitfalls of existing works under test-time distribution shifts and propose Federated Test-time Head Ensemble plus tuning(FedTHE+), which personalizes FL models with robustness to various test-time distribution shifts. We illustrate the advancement of FedTHE+ (and its computationally efficient variant FedTHE) over strong competitors, by training various neural architectures (CNN, ResNet, and Transformer) on CIFAR10 andImageNet with various test distributions. Along with this, we build a benchmark for assessing the performance and robustness of personalized FL methods during deployment. Code: https://github.com/LINs-lab/FedTHE.

cs.LG↗