Overview

Meta-learning studies how experience across related tasks can prepare a model to learn a new task from a small support set, brief feedback, or context alone. Common approaches include gradient-based initialization methods such as MAML, metric and prototype methods, Bayesian methods, learned optimizers and hypernetworks, and episodic training for in-context or test-time adaptation. Applications range from few-shot vision and domain-shifted control to AutoML, scientific system identification, and language or robotic agents, where fast adaptation can reduce labeling, retraining, or optimization costs. The central challenge is transferring useful shared structure without negative transfer: task diversity, distribution shift, uncertainty, costly bilevel training, and fair evaluation of adaptation remain active concerns.

Latest papers

174 papers11 papers in the last four weeks, down 58% on the four weeks before. 0.1% of all new papers.

Nov 18, 2026

Meta-LSQ: A Closed-Form Sobolev Inner Solver for Few-Shot Meta-Learning

Gradient-based meta-learners such as model-agnostic meta-learning (MAML) adapt to a new task with an inner loop of gradient descent that has to be differentiated through, and they meta-learn every network weight. Meta-LSQ instead fits each task with one regularised least-squares solve over a fixed random Fourier basis and meta-learns only three scalars, a bandwidth, a ridge and a derivative weight. Since the derivative of a sinusoid is again a sinusoid, the same solve can also take derivative observations, and the fitted model has exact closed-form derivatives of any order. On 10-shot sinusoid regression Meta-LSQ trains in 14 s and ends with a 5.5 times lower error than a first-order MAML baseline that trains 36 times longer, and a Gaussian-process control suggests that most of this gain comes from sharing one kernel across tasks. In simulated beam tuning at DESY's ARES facility with model derivatives it is slightly more accurate than an ALPaCA-style model and well ahead of gradient descent. Over DESY's full range of target beams, again in simulation and from measurements alone, its best setting has a median error of 7.3 μμm, against 14.4 for the published final beams of their reinforcement-learning agent, which has the lower median on their 300 trials of small targets. On real sensor-calibration data a learned prior mean puts it ahead of MAML on two of four targets, but an exact Gaussian process with the same change and an ALPaCA-style model stay more accurate on most targets, and it breaks down on non-smooth targets. On noiseless targets outside the span of the basis, in-sample criteria such as generalised cross-validation choose too small a ridge.
Nov 12, 2026

Linking Rapid Learning to Compositional Structure in Neural Networks: Evidence from Additive Analogies in Meta-Learned Parameter Spaces

Humans can learn rapidly from limited experience, while neural networks typically require extensive training. In this work, we investigate what properties of a neural network can enable it to learn rapidly. We study neural networks that have been optimized via meta-learning, a technique that exposes a network to many related functions to facilitate its subsequent learning of similar functions. Prior work has shown that these meta-learned networks can learn new functions rapidly, but how they do so remains unclear. To investigate their internal structure, we take inspiration from classic work demonstrating additive relationships among the learned vectors that neural networks use to represent words (e.g., "king" minus "man" plus "woman" approximates "queen"). Analogously, we look for additive relationships among a meta-learned network's "representations" of different functions; when a meta-learned network rapidly learns to implement a specific function, all of its adapted parameter values can be concatenated into a single vector that we treat as representing the function. We find that such parameter vectors can be added and subtracted according to symbolic relationships among the functions they represent, yielding new parameter vectors that successfully represent the functions predicted by the corresponding symbolic manipulations: the networks specified by these artificially recombined vectors actually implement the intended functions without any direct training on them. Our results suggest that meta-learning can tune a network's starting point to organize subsequent learning along systematic, functionally meaningful trajectories through the space of possible parameter vectors, providing a geometric basis for realizing inductive biases in neural networks.
Nov 9, 2026

ARMOR: Adaptive Curriculum Meta-Learning for Noise-Robust RAG Reasoning

Retrieval-Augmented Generation (RAG) systems have demonstrated remarkable effectiveness in mitigating hallucinations by incorporating external knowledge. However, the retrieval process inevitably introduces noise, posing significant challenges to RAG robustness. Fundamentally, noise robustness is a composite competency encompassing noise identification, retrieval filtering, and robust reasoning. Unfortunately, existing methods often fail to capture this holistic nature, typically restricted to single-dimensional enhancements or hampered by noise-induced optimization interference. To address these limitations, we decompose RAG robustness into three sub-tasks: knowledge boundary perception, retrieval verification, and robust reasoning. Accordingly, we propose ARMOR, an Adaptive Curriculum Meta-Learning for Noise Robust RAG Reasoning. Inspired by the teaching process, ARMOR introduces an adaptive task scheduler as a teacher to curate training plans according to the absolute learning progress of RAG generators (Student). Distinct from GRPO, ARMOR utilizes inter-group advantage estimation during learning to effectively mitigate mode collapse caused by limited output spaces. Experiments on three complex reasoning datasets (2WikiMultiHopQA, MusiQue, and HotPotQA) indicate that ARMOR achieves synergistic improvements in retrieval noise robustness, significantly outperforming baselines in output quality.
Nov 5, 2026

Simple Meta-Learning for HPO and CASH Using Performance Representations

Bayesian optimization for Hyperparameter Optimization (HPO) and Combined Algorithm Selection and Hyperparameter Optimization (CASH) often relies on explicit hyperparameter encodings as surrogate inputs. In high-dimensional search spaces, however, such representations can make it difficult for standard surrogates to learn useful performance models. We instead study an alternative meta-learning representation that describes each configuration directly as its historical performance across prior tasks. This representation can be paired directly with standard surrogate models. Our scope is the iterative selection component of AutoML, deciding which configuration to evaluate next from a fixed candidate set, rather than a complete framework with features such as runtime estimation, cost-aware budgeting and ensembling. We further apply this representation to maximal marginal relevance (MMR), a diversity-based re-ranking criterion from recommender systems, as a posterior-free acquisition rule. When selecting and optimising across multiple models on the TabArena benchmark, the performance-based representation paired with a linear Bayesian ridge regression model outperforms dedicated meta-learning baselines, achieving the same tuned performance in roughly half the number of trials. For single-model HPO, it remains a competitive second. Across both benchmarks this method closely matches or improves on regret at a fraction of the complexity and cost, requiring no pre-training or tuning and scaling independently of search space dimensionality. Code for reproducing our results is available at https://github.com/EWitting/simple-metalearning.
Nov 3, 2026

Meta2^2: Precomputing Reusable Knowledge for Online Harness Evolution in Long-Horizon Agents

Harness optimization improves long-horizon agents by refining the prompts, tools, memory, and hooks around a frozen model. Offline methods can learn from completed trajectories, but deploy one shared harness that may overfit training tasks and cannot react to evidence revealed during a new run. Online evolution can adapt before, during, or after one run through an evolver, but may have to reason about every intervention from scratch or rely on verifier feedback that is usually unavailable at test time. We introduce Meta2^2, which moves offline optimization one level up: rather than optimizing a shared harness, it learns reusable knowledge for harness evolution and applies it within each task online. Offline, Meta2^2 learns a textual Meta-Skill MM by contrasting training rollouts of the same task agent under three configurations: without an evolver, with an evolver guided by the current version of MM, and with an evolver given privileged verifier information available only in training. Contrasts in trajectories and outcomes are aggregated across tasks to refine MM into reusable guidance on when and how to evolve, when to leave the harness unchanged, and which interventions to avoid. Online, an evolver combines the frozen MM with emerging trajectory evidence to decide whether to KEEP the harness unchanged or EVOLVE it through a task-local edit. Each edit applies to the rest of the same run without resetting its environment or progress. Thus, evolution guidance transfers across tasks while harness edits remain task-local, and held-out verifier feedback is never available to the evolver. On two long-horizon benchmarks, Terminal-Bench 2.1 and DeepSWE 1.1, Meta2^2 achieves the highest performance among evaluated methods with GPT-5.6-Terra, reaching 81.00% and 44.44%, respectively. Ablations support the value of contrastive evolution guidance and keeping online harness edits task-local.
Nov 2, 2026

MetaPOCS: Meta-Learned Multi-Exit Unfolding for Budget-Aware Seismic Data Interpolation

Seismic data interpolation must reconcile reconstruction fidelity with the computational budgets of acquisition quality control and subsequent imaging. Fixed-depth unfolding networks are optimized for a single operating point, whereas their intermediate estimates may be insufficiently trained for early deployment. We present MetaPOCS, a meta-learned multi-exit unfolding framework for budget-aware seismic interpolation. Building on a spatial--spectral POCS reconstruction backbone, MetaPOCS supervises intermediate stages and preserves the acquired traces at every exit. A Meta-GF-based optimization procedure constructs exit-oriented parameter displacements from a common initialization and learns their fusion weights through an equal-exit outer objective. The weights vary across parameter tensors and respect the nested support of the reconstruction stages. We derive the resulting fusion hypergradient to characterize how the outer objective reallocates competing updates, and distinguish this mechanism from scalar loss reweighting. At inference, one checkpoint provides a set of reconstruction budgets through a latency-calibrated exit policy. The evaluation framework couples matched-cost interpolation comparisons with finite-update diagnostics and a fixed reverse-time-migration operator, connecting multi-exit optimization to waveform and imaging fidelity. MetaPOCS thus provides a unified formulation for studying the reconstruction--computation trade-off of seismic unfolding across deployment budgets.
Oct 27, 2026

Meta Models: Models that Model Learning

Every neural-network training run produces not only a model, but also evidence about how models learn. Can these accumulated experiments enable us to predict and compare new training runs before executing them? We study meta-models: models that predict the learning behavior of other models. From this perspective, final-performance prediction, learning-curve prediction, and hyperparameter selection are different queries to the same training response function. We take a first, humble empirical step toward this goal: treating real language-model training records as native numeric data, we find that an off-the-shelf tabular in-context learner, TabPFN, can match or outperform a fine-tuned LLM meta-model in multiple evaluated settings. Whereas recent work fine-tunes LLMs on tokenized configurations at a cost of tens to hundreds of GPU-hours, TabPFN requires seconds of fit-and-predict time without task-specific weight updates. We argue that meta-modeling is better treated as its own numeric modality rather than as a text task for LLMs. Going one step further, augmenting TabPFN with Kolmogorov-Arnold Networks (KANs) further improves accuracy on selected prediction tasks. Finally, we pre-registered a forecast of the Marin Hero Run, a 535B-parameter public training run, whose ground truth is revealed point by point as training progresses, putting meta-model predictions to the test against training outcomes not yet observed at the time of forecasting.
Oct 25, 2026

MetaARIMA: Configuration-Level Transfer Learning for ARIMA Models

Time Series Foundation Models have demonstrated that pretraining on diverse datasets enables effective transfer to new forecasting tasks. Classical methods like ARIMA, however, remain configured and trained independently for each series, discarding knowledge from previous experiments. Consequently, selecting and training an ARIMA configuration remains challenging and often relies on heuristic approaches based on in-sample goodnessof-fit measures. We propose MetaARIMA, a pretrained configuration policy based on metalearning that brings transfer learning to classical statistical forecasting at the configuration level rather than the model level. The policy learns from historical experiments which ARIMA configurations achieve top-quantile performance, and recommends promising candidates for new series while preserving interpretability and local estimation. At inference, it selects a diverse set of candidates via maximal marginal relevance and efficiently identifies the final configuration using successive halving. Experiments in a transfer learning setting, training on M4 and evaluating on M1, M3, and Tourism, show that MetaARIMA outperforms AutoARIMA and other classical methods while being more dataefficient. In a comparison with recent zero-shot foundation models, MetaARIMA remains competitive with Chronos2, Moirai2, and TimesFM. The experiments are publicly available at https://github.com/vcerqueira/experiments-metaarima.
Oct 22, 2026

SyRA-CS: Synthetic-to-Real Adaptation for Code-Switched ASR with Meta-Learning

Code-switched (CS) automatic speech recognition (ASR) remains difficult because real CS speech is scarce and synthetic speech does not fully match it. We investigate whether synthetic CS speech is more useful for learning an adaptation-friendly initialization than for ordinary continual fine-tuning. We propose SyRA-CS, a pair-specific synthetic-to-real adaptation framework based on Reptile meta-learning. For each language pair and pretrained model, we meta-train on the corresponding synthetic CS-FLEURS subset and then adapt using one hour of labeled real target speech. We evaluate four architectural settings: Encoder-CTC, FastConformer-RNN-T, Attention Encoder-Decoder (AED) and Speech-LLM. Experiments cover on Arabic-, French-, German-, and Hindi-English, against synthetic-only transfer, real-only fine-tuning, and continual fine-tuning. SyRA-CS improves over continual fine-tuning on WER in all 16 model-pair comparisons and on CER in 15 of 16, with relative macro WER reductions of 17.7% for XLS-R and 5.3-5.8% for the other three models. The gains hold across the tested models but vary in size across architectures, language pairs and metrics.
Oct 20, 2026

MuPACE: Learning to Pace Muon for LoRA Fine-Tuning

Learning rate schedules are typically fixed before training begins, yet when fine-tuning large language models with low-rank adapters (LoRA), the appropriate scale varies substantially with the model, task, and adapter configuration. Practitioners re-tune a schedule for every new problem, yet a tuned schedule still cannot react to the training dynamics it encounters. We instead amortize learning rate selection across LoRA fine-tuning problems. We leave the base optimizer, Muon, untouched, so that its normalized update fixes the direction, and meta-learn only the global step size. Our method MuPACE is a small recurrent policy, meta-trained across a distribution of fine-tuning runs, that conditions on the observed trajectory and emits bounded adjustments to an interpretable analytic schedule. On 64 fine-tuning problems, MuPACE outperforms standard fixed schedules at a strong shared default on 64-78% of problems and is on par with the same schedules with per-task tuned peaks. Meta-trained only on models with up to 16 layers, MuPACE transfers one-shot to unseen tasks on models as deep as 64 layers, such as Qwen3-32B and Qwen3.8-27B, where a single run remains competitive with swept schedules.
Oct 19, 2026

Meta-iLaD: Identifiable Latent Dynamics via Meta-Learning of Dynamics Environments

Learning latent dynamics is central to assessing the current states and forecasting the future trajectories of high-dimensional time series. For locally-stationary latent dynamics F(z<t;c)\mathcal{F}(\mathbf{z}_{<t}; \mathbf{c}) with latent dynamics state zt\mathbf{z}_t and environment variable c\mathbf{c}, prior identifiability results have largely focused on zt\mathbf{z}_t when conditioned on predefined label uu of dynamics environments. This leaves two limitations: reliance on predefined labels that hinder generalization to unseen environments, and a limited understanding of the identifiability of F\mathcal{F} and c\mathbf{c} which -- although offering important structural properties for the identifiability of zt\mathbf{z}_t -- are learned jointly with zt\mathbf{z}_t. We address these limitations with Meta-iLaD, a novel identifiable latent dynamics framework attained by meta-learning across dynamics environments. Meta-iLaD introduces a novel conditional prior of c\mathbf{c}, modeled as a feedforward meta-learner to rapidly extract c\mathbf{c} from few-shot examples. Meta-iLaD further establishes identifiability for zt\mathbf{z}_t, c\mathbf{c}, and F\mathcal{F} simultaneously, for a general formulation of F(z<t;c)\mathcal{F}(\mathbf{z}_{< t};\mathbf{c}) without restricting the dimension of c\mathbf{c} or how it modulates F\mathcal{F}. On synthetic and real data, we provide strong empirical evidence that 1) conditioning on few-shot examples enables generalization to out-of-distribution environments, and 2) identifiability for c\mathbf{c} and F\mathcal{F} is critical for accurate forecasting beyond reconstructing observed trajectories. Code is available at https://github.com/GoForIt-YYB/Meta-iLaD.
Oct 19, 2026

Cognitively Decoupled Meta-Induction for Out-of-Distribution Generalization

In-context learning (ICL) tackles downstream tasks conditioned on demonstrations. This paradigm relies heavily on neighbor-based matching between queries and demonstrations, limiting out-of-distribution (OOD) performance. For better generalization, rule-based inductive reasoning derives task-level regularities, which we term task-level inductive representations (TIR), before directly answering. However, existing methods lack generalizable inductive ability across diverse tasks. Inspired by meta-learning, we propose Cognitive-Decoupling meta-Induction (CoDI) for TIR-based inductive reasoning on general tasks. CoDI comprises a Cognitive-Decoupling Framework for query-blind TIR induction, and a Dual-Loop Meta-Inductive Learning method for cross-task inducer optimization. Empirically, CoDI achieves substantial performance gains on five ID and OOD benchmarks, and preserves OOD gains without explicit induction at inference. Code is available at https://github.com/Yuqian-Wang2277/codi-meta-induction.
Oct 16, 2026

LUMINR: Intensity-conditioned Meta-learned INRs for Accelerated Medical Image Registration

Implicit neural representations (INRs) have been applied to deformable image registration, where a lightweight network maps a spatial coordinate to a deformation vector. Because an INR encodes one specific signal in its parameters, a fresh network must be optimized for every image pair at test time, which is slow. Existing methods reduce this cost by meta-learning an initialization across a population of image pairs, and condition the weights of the INR globally, so the same conditioning applies at every coordinate while the deformation field it initializes varies across the domain. We propose LUMINR, a meta-learning framework that accelerates this per-pair fitting and in which the conditioning is local instead. LUMINR uses an INR that maps a coordinate, together with local intensity and gradient features, to a deformation vector, so the conditioning varies over the domain at the resolution of the field being modelled. Its objective is to find an initialization of the INR parameters from which the network can be fine-tuned to register any unseen pair in only a few optimization iterations. We evaluate LUMINR on 3D deformable registration of brain MRI from the OASIS dataset, where local conditioning and the meta-learned initialization prove complementary: early in fitting they reach a markedly higher accuracy than either does alone. Fine-tuned from that initialization, LUMINR matches the accuracy of the same conditioned INR fit from scratch in roughly a tenth of the gradient steps and a tenth of the wall-clock time, while the deformation fields stay smooth with negligible folding.
Oct 14, 2026

Active Generative Simulation: Meta-Learning What to Simulate for Sim-and-Real Policy Co-Training

Sim-and-real co-training is an effective way to scale robot learning by augmenting limited real-world demonstrations with simulation data. However, existing approaches typically treat simulators as fixed data sources, relying on uniform generation, offline retrieval, or manual parameter alignment by domain experts. We formulate Active Generative Simulation: optimizing a simulator's data-generation distribution to maximize downstream real-world performance. To solve this problem, we propose MetaGenSim, a meta-learning framework that trains an auxiliary rater network via meta-gradients to automatically discover high-utility data-generation distributions within a user-specified simulation parameter space, without manual parameter alignment. Across simulation and real-world manipulation tasks, MetaGenSim outperforms uniform generation and retrieval baselines, achieves a 150% relative improvement in real-world full-task success over uniform generation, and matches expert manual alignment. We further show that distributions curated with a lightweight proxy policy transfer to VLA fine-tuning, and that MetaGenSim can exploit partial alignment through atomic skill stitching and object-receptacle composition when exact sim-and-real task alignment is unavailable. Project website: https://meta-gensim.github.io
Oct 8, 2026

MetaOPD: Meta-Learned Token Weighting for On-Policy Distillation

On-policy distillation (OPD) trains a student on its own generated responses using token-level teacher supervision. However, uniform weighting overlooks differences in token learning value, while existing weighting methods rely on predefined mappings from prediction signals to token weights. These mappings are not learned from the effectiveness of the resulting student updates, limiting their ability to adapt to evolving learning needs. In this paper, we propose MetaOPD, a bilevel optimization framework that jointly learns the student model and a lightweight token-weighting network. The inner objective updates the student through weighted OPD, while the outer objective optimizes the weighting network using validation loss on reference solutions after a virtual student update. Differentiating through this update connects weighting decisions to their effects on post-update performance, allowing the mapping from prediction signals to token weights to evolve alongside the student. Experiments on six mathematical reasoning and three out-of-domain datasets, covering two student scales and seven baselines, demonstrate the effectiveness of MetaOPD, with Avg@8/Pass@8 gains over OPD of 1.99/5.97 percentage points for the 0.6B student and 2.25/6.41 points for the 1.7B student.
Oct 7, 2026

MotherTree: Meta-learning on synthetic data improves decision tree training

Conventional decision tree algorithms produce effective, transparent models that can be audited, communicated, and deployed independently of the training data, but require learning every new task from scratch. In contrast, tabular foundation models demonstrate that meta-learning from a synthetic prior distribution enables strong in-context prediction for previously unseen tasks, especially in small-sample regimes. However, this approach does not produce a standalone model that can be inspected in isolation. We introduce MotherTree, a tabular transformer that meta-learns decision tree induction: given a training set for a new task, it outputs a hard, axis-aligned decision tree, equivalent in form to classically trained trees, in a single forward pass. MotherTree is pre-trained on a synthetic prior using stochastic gradient descent without requiring reference trees for supervision. On established benchmarks with controlled sample size, the approach is competitive with size-matched trees from common algorithms: recursive partitioning, gradient-based tree learning, globally optimal trees, and distillation from tabular foundation models. Notably, MotherTree consistently improves over from-scratch gradient-based learning and acts as a strong initializer: task-specific tuning of the generated tree outperforms the corresponding from-scratch learner on all benchmarks and sample sizes. These results show that meta-learning can provide effective inductive biases for learning stand-alone, small decision tree classifiers.
Oct 6, 2026

MetaLearnNCA: Few-Shot Offline Meta-Learning via Interacting Neural Cellular Automata

Few-shot meta-learning traditionally formulates task adaptation either as analytical gradient descent through unrolled computational graphs or as metric-based distance comparisons over flattened 1D fea- ture vectors, which either incur costly test-time backpropagation or discard native 2D spatial geometry. In this work, we propose METALEARNNCA, a decentralized framework that achieves few-shot adapta- tion through the dynamical interaction of coupled Neural Cellular Automata (NCAs) without computing analytical gradients during inference. MetaLearnNCA decomposes task adaptation into an Active- NCA, which executes task inference conditioned on a continuous 2D spatial memory grid termed the spatial program, and a learned Meta-NCA, which acts as a decentralized cellular optimizer by diffusing spatial error residuals across local neighborhoods to dynamically update this program. METALEARN- NCA is competitive against canonical meta-learners in-distribution (96.12% on Omniglot) with Out-Of- Distribution transfer gains on MNIST, KMNIST, and Fashion-MNIST transfer across 10 independent testing seeds across 1-, 5-, and 10-shot regimes (e.g., surpassing Prototypical Networks by +10.54% on 10-shot MNIST and a +3.87% gain on 10-shot Fashion-MNIST over FOMAML). Our results establish that robust, gradient-free learning-to-learn can emerge from decentralized cellular dynamics on non-von Neumann substrates.
Oct 6, 2026

Learning PDE solution operators with variable initial conditions via Latent Dynamics Networks

In many-query scenarios, data-driven surrogate models provide an efficient alternative to high-fidelity solvers for simulating physical systems governed by Partial Differential Equations (PDEs). In this context, the Latent Dynamics Network (LDNet) has recently demonstrated remarkable performance in predicting the response of spatio-temporal systems, combining Neural Ordinary Differential Equations with nonlinear dimensionality reduction. However, the original formulation assumes a fixed initial condition, limiting its applicability to many real-world applications where a system evolves from varying starting states. In this work, we overcome this limitation while keeping the end-to-end training procedure of the original LDNet and its encoder-free nature, which preserves its intrinsic independence from spatial resolution and grid topology. We infer the initial latent state directly from a small set of early-time observations, treating latent-state initialization as an adaptation problem, and investigate two strategies: an auto-decoding formulation and a meta-learning approach in which the initial latent state acts as a task-specific context variable. We demonstrate the accuracy of the proposed methods across diverse physical phenomena, spanning advection-diffusion, fluid dynamics, and solid mechanics. Meta-learning markedly accelerates latent-state inference and induces smoother, better-conditioned optimization landscapes, and spontaneously organizes the latent space into a structured representation that reflects physically meaningful features of the underlying dynamics. The coordinate-based decoder enables training from spatially subsampled data while recovering high-resolution solution fields at inference. The resulting approach provides an efficient and resolution-independent surrogate modeling framework for many-query simulations of time-dependent PDEs with varying initial conditions.
Oct 6, 2026

Optimization Encoders: Rethinking Second-Order Meta-Learning for Neural Fields

Conditional neural fields represent signals continuously, but their effectiveness depends on how the conditional latent representations are inferred from observed data. In meta-learning, this encoding occurs through gradient updates induced by the decoder, tying representation learning directly to decoder design. We formalize this connection by interpreting latent optimization as an optimization encoder, unifying the roles of second-order differentiation, latent parameterization, and task supervision. This concept enables second-order meta-learning for end-to-end training of the encoding procedure alongside the decoder, and clarifies which learning pathway first-order approximations discard. Guided by this view, we introduce Attentive Latent Fields (MetaLF), an equivariant transformer-based neural field that contextualizes a latent pointcloud through self-attention. These interactions shape both field predictions and the updates that construct their representation, allowing local observations to inform coherent non-local structure. Disentangling the inner encoding objective from outer task supervision unifies reconstruction, classification, and segmentation within an end-to-end meta-learning framework, using reconstruction-only latent adaptation at test time. Controlled experiments on polynomial fields link latent coordination to lower effective rank and stronger alignment with the underlying function space. Across image and 3D shape reconstruction, MetaLF improves fidelity within three to five gradient updates, while supporting semantic prediction across images, shapes, and volumes. Together, these findings position the optimization encoder perspective as a unified basis for designing neural fields around how representations are constructed, coordinated, and used.
Oct 5, 2026

Few-Shot Bioactivity Prediction with Meta-Learning under Assay Heterogeneity

Accurate bioactivity prediction is a central challenge in early-stage drug discovery, as individual assays often contain too few measurements to train reliable models independently. Meta-learning offers a principled approach to this few-shot setting, but assay heterogeneity may limit its effectiveness. Here, we test this hypothesis and show that meta-learning performance degrades as meta-training tasks become more heterogeneous. To address this, we introduce MetaHeta, a meta-learning framework that accounts for assay heterogeneity by conditioning predictions on auxiliary data from related assays, with relatedness defined flexibly from available assay information. The architecture of MetaHeta combines linear attention over large auxiliary datasets with exact attention over scarce task-specific context, enabling efficient scaling to the former without compromising exact attention over the latter. We demonstrate the benefits of our approach on assays from ChEMBL and BindingDB, improving few-shot bioactivity prediction and downstream compound prioritization in retrospective Bayesian optimization.
Oct 5, 2026

Learning to Learn a Language

We present the Prior-Fitted Language Model (PFLM), a 300M-parameter byte-level transformer pretrained only on samples from a synthetic non-linguistic prior. Given a prefix of real text, it learns to predict the language in context with frozen weights, having never seen a word of any real language. Every training sequence is generated by a recurrent structural causal model drawn fresh from a distribution over such models. The model never sees the same language twice during training, so the only way to predict the continuation is to infer the language from the prefix. Samples from this prior share the statistical signatures of natural text: Zipfian frequencies, slow entropy-rate convergence, and long-range dependence. On Wikipedia in six languages, bits per byte fall from the uniform eight to between 0.9 and 2.4 at one million bytes of context. Given numerals instead of text, PFLM learns to count, to compare magnitudes, and to add approximately. It predicts deterministic sequences like Rudin-Shapiro or the prime indicator, and it compresses six non-text domains, from source code to speech, below gzip and PPMd. The model has not learned a language. It has learned to learn one.
Oct 5, 2026

RepICL: Reusable In-Context Prediction Across Heterogeneous Representation Spaces

Frozen representations are widely reused for downstream classification, yet each new task typically requires fitting a new predictor. We ask whether the few-shot prediction procedure itself can instead be learned once and reused across datasets and representation spaces. To study this question, we introduce RepShiftBench, comprising 1,218 encoder--dataset tasks across text, image, and audio, with separate evaluation of generalization to unseen datasets, unseen encoders, jointly unseen datasets and encoders, and unseen modalities. The benchmark exposes a substantial gap: Logistic Regression fitted independently on each episode outperforms every evaluated in-context learner across all settings. We introduce RepICL, a meta-trained in-context learner that canonicalizes each episode through episodic whitening before prediction. Its inductive variant, RepICL-I, surpasses Logistic Regression in all 12 benchmark settings, while RepICL-T substantially outperforms existing transductive methods. Ablations identify episodic whitening as the primary source of these gains, while showing that it is not a universally beneficial preprocessing step. Across both variants, the gains concentrate on queries for which simple support prototypes favor the wrong class or provide little separation between the true class and competing classes. Transduction provides its largest additional gains when limited support coverage gives a misleading view of class separation. Together, these results demonstrate that a shared few-shot prediction procedure can generalize beyond the representation spaces observed during training.
Oct 3, 2026

TMAML: Temporal Model-Agnostic Meta-Learning for Cold-Start Time Series Forecasting

Cold-start forecasting, the task of forecasting a time series with little to no historical data, is a common challenge. Addressing it requires approaches that learn quickly from few datapoints and leverage information from related series, typically through static covariates, to generalize well to unseen series. While some global forecasting models can generate cold-start predictions by leveraging information shared across multiple series, they are not optimized for out-of-train-set generalization or adaptation from short histories. In this work, we formulate cold-start forecasting as a few-window learning problem and introduce Temporal Model-Agnostic Meta-Learning (TMAML), which tailors the model-agnostic meta-learning algorithm, originally developed for few-shot adaptation of neural networks, to deep time series forecasting. TMAML constructs meta-tasks as temporally consistent support-query windows and pairs them with a temporal meta-training and meta-testing procedure, yielding forecasting models that are explicitly optimized for cold-start forecasting. We instantiate TMAML on the Temporal Fusion Transformer (TFT) and present an initial empirical analysis of forecast accuracy and calibration across three cold-start scenarios: TMAML consistently outperforms or matches a standard ERM-trained TFT, yields better-calibrated forecasts than naive on two of the three scenarios, but does not consistently outperform naive on probabilistic forecast accuracy.
Oct 1, 2026

A Hybrid Approach to Malware Detection: Integrating Few-Shot Model-Agnostic Meta-Learning with Autoencoders

Ransomware has emerged as a major cybersecurity threat, with incidents increasing in frequency and impact across critical sectors. These attacks are typically launched through phishing emails, malicious downloads, or exploitation of software vulnerabilities to gain system access. Once inside, the malware encrypts files and demands a ransom, often in cryptocurrency, for the decryption key. Conventional detection methods often struggle with novel or scarce samples, leaving systems vulnerable. To address these challenges, this paper proposes a hybrid deep learning framework that combines an Autoencoder Feature Extractor (AFE) with a Model Agnostic Meta Learning (MAML) classifier for few shot malware detection. The AFE generates compact latent features that reduce noise and dimensionality, while the MAML classifier rapidly adapts to new threats using limited labeled data. Experiments conducted on the Ransomware Dataset 2024 demonstrate the effectiveness of the framework in binary classification tasks. Across one to fifty shot settings, the proposed model consistently achieves high accuracy, F1 score, and Matthews Correlation Coefficient values, maintaining reliable classification even under extreme scarcity. These results highlight the model's robustness and effectiveness in adapting to limited data scenarios, demonstrating the potential of combining feature extraction with meta learning to enhance resilience against malware, particularly in sectors such as healthcare, manufacturing, and public infrastructure, where cyberattacks can cause significant operational and financial disruption.
Sep 30, 2026

LARC: Low-Rank Adaptive Residual Connections for Learning in Frozen Models

Low-Rank Adaptive Residual Connections (LARC) give a frozen model a compact numerical state that can learn from feedback. The map h+BAhh+BAh changes the hidden input to a pretrained mixer. An episode copies a learned initialization \ρ\ρ into a private state \Φ\Φ, updates its factors when feedback arrives, and resets to the bound \ρ\ρ. We analyze the factor-space updates and trace this input-side policy carrier into Orbit, the post-training realization of Memory-Mediated Learning Architecture studied here. In Orbit's released default, the learned residual shapes predictions while memory adaptation changes the retained evidence.\n\nThe controlled studies use a rank-4 residual with 12,288 factor entries on frozen MiniCPM5-1B-SFT. Two feedback updates reduce expected query execution error by 24.65 and 36.65 percentage points relative to the respective trained reset states in a four-candidate development task with 16 parameter groups and three paired seeds. A direct support-loss rule reaches 0.78125\% error. In repository-balanced chronological replay of public continuous-integration jobs, retaining updates raises half-Brier loss from 0.1274 to 0.1808. A follow-up intervention observes same-batch non-descent and inconsistent future benefit from shrinking steps. These comparisons measure when numerical state helps a later read and when the trained reference is more effective.
Sep 29, 2026

Compiling Learning Problems into Adaptation Programs for Language Models

Model adaptation is typically governed by a fixed recipe, even though different update programs can produce substantially different behavioral outcomes. We introduce adaptation compilation, which reframes where, how, and to what extent a model should adapt as a joint prediction and decision problem. Rather than searching over candidate programs anew for each learning episode, a compiler learns from prior adaptations to predict a vector-valued counterfactual response surface over candidate programs---their expected effects on acquisition, transfer, boundedness, and preservation---and selects a program before adaptation begins. Because this predicted geometry captures multiple behavioral consequences rather than a single winner or scalar score, it can be reused under different downstream priorities without retraining. Across five learning types, preferred programs vary meaningfully across episodes, and this variation is predictable from pre-adaptation information. On Llama-3.1-8B, compiler-selected programs approach exhaustive search while outperforming global and objective-specific defaults. Replication on Gemma-2-9B preserves program heterogeneity and selection headroom, but shows that exploiting this headroom requires accounting for uncertainty when departing from strong defaults. Together, these results show that adaptation search can be amortized across related learning problems, turning prior adaptation experience into a basis for deciding how future learning should occur.
Sep 28, 2026

Harness Learning Enables Generalizable Test-Time Adaptation

A language-model agent is jointly defined by its model and its harness, the executable program that organizes model calls, tool use, and information flow. Because different tasks call for different ways of organizing these operations, the harness needs to be adapted using feedback from the task at hand. We introduce harness learning, which trains a proposer model to revise a solver's harness using execution feedback. We formulate this process as meta-learning over executable programs, with harness revisions playing the role of weight updates in gradient-based adaptation. We train the proposer with reinforcement learning, using the task performance of revised harnesses as the reward. At test time, the proposer uses feedback from successive executions on a new task to refine the harness, without performing any parameter-space update. Experiments on reasoning and multi-hop question answering show that harness learning improves revision quality and that the ability to adapt at test time transfers to unseen tasks. Policies trained on individual revisions can continue improving harnesses over multiple rounds, while the benefits of training on revision sequences vary across settings. These findings suggest a path towards continually learning agents that turn accumulated experience into generalizable improvements.
Sep 27, 2026

DGS-MLDG: Domain Gradient Surgery Guided Meta-Learning for Domain Generalization in Speech Deepfake Detection

Speech deepfake detection faces significant challenges due to domain shifts. Domain generalization (DG), particularly meta-learning for domain generalization (MLDG), offers a promising solution by simulating and mitigating domain shifts. However, MLDG is often hindered by conflicting gradients between its meta-train and meta-test objectives, leading to suboptimal performance. To address this problem, we propose domain gradient surgery (DGS), a meta-learning method that resolves conflicts through an asymmetric projection strategy. DGS removes the destructive component from the meta-test gradient, ensuring a conflict-free optimization trajectory versus the meta-train gradient. Furthermore, we introduce layer-wise DGS (LW-DGS), an efficient variant of DGS that dynamically identifies and intervenes only conflict-prone layers. Extensive experiments on challenging benchmarks demonstrate that DGS-MLDG and LW-DGS-MLDG achieve an average relative EER reduction of 5.29% and 4.04%, respectively.
Sep 27, 2026

ABO-Med: Accelerated Bilevel Optimization for Few-Shot Medical Image Classification

In recent years, bilevel optimization has been widely used in a variety of machine learning tasks. However, prior bilevel optimization algorithms generally require the computation of second-order information, which limits their practical scalability. Only recently has a first-order paradigm for bilevel optimization been established, attaining near-optimal theoretical guarantees for solving bilevel optimization problems. In this paper, we propose ABO-Med, a scalable instantiation of this paradigm for few-shot learning, by incorporating it into the model-agnostic meta-learning (MAML) framework and tailoring it to medical image classification. We also introduce Medical Adaptive RandomAugment (MedRAug), a modality-aware augmentation strategy designed for medical images. Theoretically, ABO-Med establishes the optimality of MAML-type meta-learning approaches. Empirically, ABO-Med outperforms prior baselines on several public medical datasets, with gains of 1.99% to 18.76%, while MedRAug further improves the average accuracy by 2.20% to 6.34%. Additional cross-domain experiments, augmentation ablation studies, backbone ablation studies, and training efficiency analysis further validate the effectiveness and efficiency of the proposed method.
Sep 26, 2026

Structuring Relations Among Learning Paradigms via Protocol--Objective--Resource Reductions

Modern machine learning spans supervised, transfer, continual, meta-learning, and related regimes that often reuse the same hypothesis classes, architectures, and optimizers but differ in information access, objectives, memory, adaptation, and sample accounting. This makes it difficult to determine whether one paradigm is genuinely distinct, a special case of another, or part of a broader structural hierarchy. We introduce a protocol-objective-resource (POR) framework that separates representational capacity from these design choices. A paradigm is specified by an environment class, observation protocol, admissible learners, performance functional, and resource accounting rule. POR reductions combine environment embeddings, learner compilers, threshold maps, and calibrated resource overheads. Our main theorem shows that such reductions imply worst-case complexity domination on embedded comparison classes, transferring upper bounds forward and lower bounds backward; under labeled-example accounting, this yields sample complexity domination. We also show that calibrated nontrivial accuracy regimes are necessary to avoid vacuous comparisons, and that strengthening the objective can strictly increase minimax sample complexity even with unchanged protocols and learner classes. Instantiating the framework for supervised, transfer, continual, and meta-learning yields canonical special-case relations: continual contains transfer, transfer contains supervised, and meta-learning contains supervised under aligned raw-example accounting. We further derive a non-exact episode-to-example reduction for episodic meta-learning and capture within-paradigm refinements such as replay memory and task identifiers. The framework thus provides a unified language for structuring learning paradigms and transferring complexity guarantees across them.