Machine Learning · Cross-listed from cs.CV
FIS-DiT: Breaking the Few-Step Video Inference Barrier via Training-Free Frame Interleaved Sparsity
Jian Tang, Jiawei Fan, Qingbin Liu et al. (+3 more)
While the overall inference latency of
Video Diffusion Transformers (DiTs) can be substantially reduced through model
distillation, per-step inference latency remains a critical bottleneck. Existing acceleration paradigms primarily exploit redundancy across the denoising trajectory; however, we identify a limitation where these step-wise strategies encounter diminishing returns in few-step regimes. In such scenarios, fewer intermediate states can limit opportunities for feature reuse or predictive modeling, motivating complementary forms of acceleration. To overcome this, we propose Frame Interleaved Sparsity DiT (FIS-DiT), a training-free and operator-agnostic framework that shifts the optimization focus from the temporal trajectory to the latent frame dimension. Our approach is motivated by an intrinsic duality within this dimension: the existence of frame-wise sparsity that permits reduced computation, coupled with a structural consistency that calls for maintaining coverage of frame positions in the global spatiotemporal context. Leveraging this insight, we implement Frame Interleaved Sparsity (FIS) as an execution strategy that manipulates frame subsets across the model hierarchy, refreshing all latent positions without requiring full-scale block computation. Empirical evaluations on Wan 2.2 and HunyuanVideo 1.5 demonstrate that FIS-DiT reaches 2.4$\times$ speedup while maintaining comparable performance on VBench, providing a scalable and robust pathway toward real-time high-definition
video generation.
Read the full abstract ▾
All authors Jian Tang, Jiawei Fan, Qingbin Liu, Zheng Wei, Jiang Bian, Zang Li
Machine Learning · Cross-listed from stat.ML
Variance-aware Reward Modeling with Anchor Guidance
Shuxing Fang, Ruijian Han, Liangyu Zhang et al. (+1 more)
Standard Bradley–Terry (BT) reward models are limited when human preferences are pluralistic. Although soft preference labels preserve disagreement information, BT can only express it by shrinking reward margins. Gaussian reward models provide an alternative by jointly predicting a reward mean and a reward variance, but suffer from a fundamental non-identifiability from pairwise preferences alone. We propose Anchor-guided Variance-aware Reward Modeling, a framework that resolves this non-identifiability by augmenting preference data with two coarse response-level anchor labels. Building on this, we prove that two anchors are sufficient for identification, develop a joint training objective and establish a non-asymptotic convergence rate for both the estimated reward mean and variance functions. Across simulation studies and four real-world diverging-preference
datasets, our method consistently improves reward modeling performance and downstream RLHF, including PPO training and best-of-$N$ selection.
Read the full abstract ▾
All authors Shuxing Fang, Ruijian Han, Liangyu Zhang, Fan Zhou
Machine Learning
Beyond Parameter Aggregation: Semantic Consensus for Federated Fine-Tuning of LLMs
Amr Abourayya, Jens Kleesiek and Michael Kamp
Federated fine-tuning of large
language models is commonly formulated as a parameter aggregation problem. However, even parameter-efficient methods require transmitting large collections of trainable weights, assume aligned architectures, and rely on white-box access to model parameters. As model sizes continue to grow and deployments become increasingly heterogeneous, these assumptions become progressively misaligned with practical constraints. We consider an alternative formulation in which collaboration is mediated through model behavior rather than parameters. Clients fine-tune local models on private data and exchange generated outputs on a shared, public
prompt set. The server maps these outputs into a semantic representation space, forms a per-
prompt semantic consensus, and returns pseudo-labels for further local fine-tuning. This formulation fundamentally changes the communication scaling of
federated LLM fine-tuning. The amount of information exchanged depends only on the public
prompt budget and the size of the communicated behaviors, independent of model size. As a consequence, the protocol naturally accommodates heterogeneous architectures and applies directly to open-ended text generation. We present a theoretical analysis and empirical results demonstrating that this approach can match strong
federated fine-tuning baselines while substantially reducing communication by orders of magnitude (e.g., analytically by a factor of $1006$ for Llama3.1-405B), as well as reductions in runtime and energy consumption. These results suggest that, for generative foundation models, behavior-level consensus provides a more appropriate abstraction for
federated adaptation than parameter aggregation.
Read the full abstract ▾
Machine Learning
Improving the Performance and Learning Stability of Parallelizable RNNs Designed for Ultra-Low Power Applications
Julien Brandoit, Arthur Fyon, Damien Ernst et al. (+1 more)
Sequence learning is dominated by
Transformers and parallelizable recurrent neural networks (RNNs) such as state-space models, yet learning long-term dependencies remains challenging, and state-of-the-art designs trade power consumption for performance. The Bistable
Memory Recurrent Unit (BMRU) was introduced to enable hardware-software co-design of ultra-low power RNNs: quantized states with hysteresis provide persistent
memory while mapping directly to analog primitives. However, BMRU performance lags behind parallelizable RNNs on complex sequential tasks. In this paper, we identify gradient blocking during state updates as a key limitation and propose a cumulative update formulation that restores gradient flow while preserving persistent
memory, creating skip-connections through time. This leads to the Cumulative
Memory Recurrent Unit (CMRU) and its relaxed variant, the $\alpha$CMRU. Experiments show that the cumulative formulation dramatically improves convergence stability and reduces initialization sensitivity. The CMRU and $\alpha$CMRU match or outperform Linear Recurrent Units (LRUs) and minimal Gated Recurrent Units (minGRUs) across diverse
benchmarks at small model sizes, with particular advantages on tasks requiring discrete long-range retention, while the CMRU retains quantized states, persistent
memory, and noise-resilient dynamics essential for analog implementation.
Read the full abstract ▾
All authors Julien Brandoit, Arthur Fyon, Damien Ernst, Guillaume Drion
Machine Learning
GEAR: Granularity-Adaptive Advantage Reweighting for LLM Agents via Self-Distillation
Sijia Li, Yuchen Huang, Zifan Liu et al. (+7 more)
Reinforcement learning has become a widely used post-training approach for LLM
agents, where training commonly relies on outcome-level rewards that provide only coarse supervision. While finer-grained credit assignment is promising for effective policy updates, obtaining reliable local credit and assigning it to the right parts of the long-horizon trajectory remains an open challenge. In this paper, we propose Granularity-adaptivE Advantage Reweighting (GEAR), an adaptive-granularity credit assignment framework that reshapes the trajectory-level GRPO advantage using
token- and segment-level signals derived from self-
distillation. GEAR compares an on-policy student with a ground-truth-conditioned teacher to obtain a reference-guided divergence signal for identifying adaptive segment boundaries and modulating local advantage weights. This divergence often spikes at the onset of a semantic deviation, while later
tokens in the same autoregressive continuation may return to low divergence. GEAR therefore treats such spikes as anchors for adaptive credit regions: where the student remains aligned with the teacher,
token-level resolution is preserved; where it departs, GEAR groups the corresponding continuation into an adaptive segment and uses the divergence at the departure point to modulate the segment' s advantage. Experiments across eight mathematical
reasoning and agentic tool-use
benchmarks with Qwen3 4B and 8B models show that GEAR consistently outperforms standard GRPO, self-
distillation-only baselines, and
token- or turn-level credit-assignment methods. The gains are especially strong on
benchmarks with lower GRPO baseline accuracy, reaching up to around 20% over GRPO, suggesting that the proposed adaptive reweighting scheme is especially useful in more challenging long-horizon settings.
Read the full abstract ▾
All authors Sijia Li, Yuchen Huang, Zifan Liu, Yanping Li, Jingjing Fu, Li Zhao, Jiang Bian, Ling Zhang, Jun Zhang, Rui Wang
Machine Learning · Cross-listed from math.OC
Constrained Stochastic Spectral Preconditioning Converges for Nonconvex Objectives
Konstantinos Oikonomidis, Jan Quan, Kimon Antonakopoulos et al. (+3 more)
In this work, we develop proximal preconditioned gradient methods with a focus on spectral gradient methods providing a proximal extension to the
Muon and Scion optimizers. We introduce a family of
stochastic algorithms that can handle a wide variety of convex and nonconvex constraints and study its convergence under heavy-tailed noise, through a novel analysis tailored to the geometry of the proposed methods. We further propose a variance-reduced version, which achieves faster convergence under standard noise assumptions. Finally, we show that the
polynomial iterations used in
Muon are more accurately captured by a nonlinear preconditioner than by the ideal
matrix sign, leading to a convergence analysis that more faithfully reflects practical implementations.
Read the full abstract ▾
All authors Konstantinos Oikonomidis, Jan Quan, Kimon Antonakopoulos, Antonio Silveti-Falls, Volkan Cevher, Panagiotis Patrinos
Machine Learning · Cross-listed from cs.ET
A Fast and Energy-Efficient Latch-Based Memristive Analog Content-Addressable Memory
Paul-Philipp Manea, Aishwarya Natarajan, Jim Ignowski et al. (+2 more)
Analog content-addressable
memories (aCAMs) based on memristors provide a promising pathway toward energy-efficient large-scale associative computing for Edge AI and embedded intelligence applications. They have been successfully applied to decision-tree inference and extend the capabilities of compute-in-
memory (CIM) architectures beyond conventional vector-
matrix multiplication. However, conventional designs such as the 6T2M architecture suffer from static search power, limited voltage gain, and pronounced match-line crosstalk, constraining analog precision and scalability. We introduce a strong-arm latched memristor (SALM) aCAM
cell that replaces static voltage division with a dynamic current-race comparator, enabling high regenerative gain, intrinsic result latching, and near-zero static search power. Compared to 6T2M, SALM reduces read energy by 33% at identical latency while eliminating the gain and crosstalk limitations that prevent 6T2M from scaling to large arrays. SALM further enables scalable sequential and parallel latch sharing, and a
dataset-aware optimization framework exposes an explicit energy-latency tradeoff, achieving up to 50% energy reduction at 3x latency across representative workloads. To enable architectural exploration, we develop a
circuit-accurate behavioral model derived from SPICE lookup tables in 22 nm FD-SOI technology, capturing match-line dynamics and crosstalk. Integrated into the X-TIME decision-tree compiler, this framework demonstrates that SALM maintains near-software accuracy for high-dimensional
datasets, whereas baseline designs degrade due to limited gain and cumulative crosstalk.
Read the full abstract ▾
All authors Paul-Philipp Manea, Aishwarya Natarajan, Jim Ignowski, John Paul Strachan, Luca Buonanno
Machine Learning
Martingale-Consistent Self-Supervised Learning
Moritz Gögl, Hanwen Xing and Christopher Yau
Self-supervised learning (SSL) is often deployed under changing information, such as shorter histories, missing features, or partially observed images. In these settings, predictions from coarse and refined views should be coherent: before refinement, the coarse-view prediction should match the average prediction expected after refinement. Martingales formalize this coherence principle, but standard SSL objectives do not enforce it. Unlike invariance objectives that pull views together, martingale consistency constrains only the expected refined prediction, allowing predictions to update as information is revealed while preventing systematic drift. We introduce a martingale-consistent SSL framework that closes this gap, with practical prediction- and latent-space variants and an unbiased two-sample Monte Carlo
estimator based on
stochastic refinement. We evaluate the approach on synthetic and real time-series, tabular, and image
benchmarks under partial-observation regimes, in both semi-self-supervised and fully label-free settings. Across these experiments, our framework improves robustness and calibration under partial observation, yielding more stable representations as information is revealed.
Read the full abstract ▾
Machine Learning · Cross-listed from stat.ML
Minimax Rates and Spectral Distillation for Tree Ensembles
Binh Duc Vu and David S. Watson
Tree ensembles such as random forests (RFs) and gradient boosting machines (GBMs) are among the most widely used supervised learners, yet their theoretical properties remain incompletely understood. We adopt a spectral perspective on these
algorithms, with two main contributions. First, we derive minimax-optimal convergence for RF regression, showing that, under mild regularity conditions on tree growth, the
eigenvalue decay of the induced kernel operator governs the statistical rate. Second, we exploit this spectral viewpoint to develop compression schemes for tree ensembles. For RFs, leading eigenfunctions of the kernel operator capture the dominant predictive directions; for GBMs, leading singular vectors of the smoother
matrix play an analogous role. Learning nonlinear maps for these spectral representations yields distilled models that are orders of magnitude smaller than the originals while maintaining competitive predictive performance. Our methods compare favorably to state of the art
algorithms for forest
pruning and rule extraction, with applications to resource constrained computing.
Read the full abstract ▾
Machine Learning
Gradient Clipping Beyond Vector Norms: A Spectral Approach for Matrix-Valued Parameters
Alexander Yukhimchuk, Mladen Kolar, Martin Takáč et al. (+1 more)
Gradient clipping is a standard safeguard for training neural networks under noisy, heavy-tailed
stochastic gradients; yet, most clipping rules treat all parameters as vectors and ignore the
matrix structure of modern architectures. We show empirically that data outliers often amplify only a small number of leading singular values in layer-wise gradient
matrices, while the rest of the spectrum remains largely unchanged. Motivated by this phenomenon, we propose spectral clipping, which stabilizes training by clamping singular values that exceed a threshold while preserving the singular directions. This framework generalizes classical gradient norm clipping and can be easily integrated into existing optimizers. We provide a convergence analysis for non-convex optimization with spectrally clipped SGD, yielding the optimal $\mathcal{O}\left(K^{\frac{2 - 2\alpha}{3\alpha - 2}}\right)$ rate for heavy-tailed noise. To minimize hyperparameter tuning, we introduce layer-wise adaptive thresholds based on moving averages or sliding-window
quantiles of the top singular values. Finally, we develop efficient implementations that clip only the top $r$ singular values via randomized truncated SVD, avoiding full decompositions for large layers. We demonstrate competitive performance across synthetic heavy-tailed settings and neural network training tasks.
Read the full abstract ▾
All authors Alexander Yukhimchuk, Mladen Kolar, Martin Takáč, Sayantan Choudhury
Machine Learning
More Edits, More Stable: Understanding the Lifelong Normalization in Sequential Model Editing
Xin Ma, Wei Chen, Qi Liu et al. (+4 more)
Lifelong Model Editing aims to continuously update evolving facts in Large
Language Models while preserving unrelated knowledge and general capabilities, yet it remains plagued by catastrophic forgetting and model collapse. Empirically, we find that recent editors resilient over long horizons share the same core strategy: Lifelong Normalization (LN), which normalizes value gradients using running statistics. Removing LN causes immediate performance collapse, and we observe a counter-intuitive positive cumulative effect where early edits can promote the success of future edits. Yet the mechanism of LN remains a "black box", leaving its precise role in lifelong stability poorly understood. In this work, we provide the first theoretical account of LN in the lifelong regime. Our analysis reveals a self-reinforcing stability loop and proves that, when combined with ridge-regularized regression, LN yields parameter updates with asymptotic orthogonality and bounded norms, directly mitigating forgetting and systemic collapse. Based on these insights, we derive StableEdit, which strengthens this stability loop via an explicit warm-up stage and full whitening, improving long-horizon stability at minimal overhead. Extensive experiments validate our theory and demonstrate competitive performance. Our code is available at https://github.com/MINE-USTC/StableEdit.
Read the full abstract ▾
All authors Xin Ma, Wei Chen, Qi Liu, Derong Xu, Zhi Zheng, Tong Xu, Enhong Chen
Machine Learning · Cross-listed from cs.NE
Multi-Timescale Conductance Spiking Networks: A Sparse, Gradient-Trainable Framework with Rich Firing Dynamics for Enhanced Temporal Processing
Alex Fulleda-Garcia, Saray Soldado-Magraner and Josep Maria Margarit-Taulé
Spiking neural networks (SNNs) promise low-power event-driven computation for temporally rich tasks, but commonly used
neuron models often trade off gradient-based trainability, dynamical richness, and high activity sparsity. These limitations are acute in regression, where approximation error, noise and spike discretization can severely degrade continuous-valued outputs. Indeed, many state-of-the-art (SOTA) SNNs rely on simple phenomenological dynamics trained with surrogate gradients and offer limited control over spiking diversity and sparsity. To overcome such limitations, we introduce multi-timescale conductance spiking networks, a gradient-trainable framework in which neural dynamics emerge from shaping the current-voltage (I-V) curve by tuning fast, slow and ultra-slow conductances. This parametrization allows systematic control over excitability, can be implemented efficiently in analog
circuits, and yields rich firing regimes including tonic, phasic and bursting responses within a single model. We derive a discrete-time formulation of these differentiable dynamics, enabling direct backpropagation through time without surrogate-gradient approximations. To probe both trainability and accuracy, we evaluate feedforward networks of these
neurons at the predictability limit of Mackey-Glass time-series regression and compare them to baseline LIF and SOTA AdLIF networks. Our model outperforms LIF and AdLIF networks, while exhibiting substantially sparser activity from both communication and computational perspectives. These results highlight multi-timescale conductance spiking
neurons as a promising building block for energy-aware temporal processing and neuromorphic implementation.
Read the full abstract ▾
Machine Learning · Cross-listed from physics.optics
Bin Latent Transformer (BiLT): A shift-invariant autoencoder for calibration-free spectral unmixing of turbid media
Martin Hohmann
The accurate recovery of constituent-level optical properties from integrating sphere measurements is a central analytical challenge in pharmaceutical analysis, food science, and biomedical diagnostics. Neural network autoencoders can extract spectrally resolved absorption and scattering coefficients for each constituent without prior knowledge, but their fully connected encoders bind learned features to absolute wavelength indices, causing accuracy loss under spectrometer calibration drift or hardware exchange. This work introduces the Bin Latent
Transformer (BiLT)-Autoencoder, in which the dense encoder is replaced by a cross-
attention scanner: 16 learnable probe vectors query a convolutional feature map, aggregating morphological spectral information independently of absolute wavelength position. A physics-constrained linear decoder with enforced absorption/scattering separation and a three-phase curriculum augmentation strategy complete the architecture. On a
liquid phantom
benchmark (intralipid and two ink absorbers; 496 samples), the model achieves $R^2 = 0.979$ and $0.975$ for $\mu_a(\lambda)$ and $\mu_s'(\lambda)$, respectively, on held-out test spectra, maintaining $R^2 > 0.90$ for $\mu_a$ and $R^2 \approx 0.99$ for $\mu_s'$ across the full tested shift range of $\pm 10$ spectral bands. The model generalises to a simulated spectrometer with a broader instrument line shape (${\approx}24$nm FWHM) without retraining, retaining $R^2 \approx 0.96$ and $0.974$ for the two channels.
Attention map analysis reveals a physically interpretable two-component probe strategy: sparse anchor probes at absorption-edge wavelengths combined with a diffuse, SNR-driven ensemble at the high-transmittance long-wavelength region, which recruits additional probes dynamically under noise to provide implicit spectral averaging.
Read the full abstract ▾
Machine Learning
Fed-BAC: Federated Bandit-Guided Additive Clustering in Hierarchical Federated Learning
Satwat Bashir, Tasos Dagiuklas and Muddesar Iqbal
Hierarchical
federated learning (HFL) leverages edge servers for partial aggregation in edge computing. Yet existing FL methods lack mechanisms for jointly optimizing cluster assignment and client selection under data heterogeneity. This paper proposes Fed-BAC, which integrates additive cluster personalization with a two-level bandit framework: contextual bandits at the cloud learn server-to-cluster assignments, while Thompson Sampling at each edge server identifies high-contributing clients. The additive decomposition enables the sharing of knowledge between groups through a globally aggregated network, while cluster-specific networks capture distribution variations. Across three classification
benchmarks (CIFAR-10, SVHN, Fashion-MNIST) under moderate ($\alpha = 0.5$) and severe ($\alpha = 0.1$) Dirichlet non-IID partitioning, Fed-BAC achieves distributed accuracy gains of up to +35.5pp over HierFAVG and +8.4pp over IFCA, while requiring only 80% client participation, converging 1.5 to 4.8$\times$ faster depending on
dataset and accuracy target, and improving cross-server
fairness. These gains are further validated at 5$\times$ deployment scale on CIFAR-10. The advantage of Fed-BAC increases with heterogeneity severity, confirming that additive cluster personalization becomes increasingly valuable as data distributions diverge.
Read the full abstract ▾
Machine Learning
Stop Marginalizing My Dreams: Model Inversion via Laplace Kernel for Continual Learning
Patryk Krukowski, Jacek Tabor, Przemysław Spurek et al. (+2 more)
Data-free continual learning (DFCIL) relies on model inversion to synthesize pseudo-samples and mitigate catastrophic forgetting. However, existing inversion methods are fundamentally limited by a simplifying assumption: they model feature distributions using diagonal covariance, effectively ignoring correlations that define the geometry of learned representations. As a result, synthesized samples often lack fidelity, limiting knowledge retention. In this work, we show that modeling feature dependencies is a key ingredient for effective DFCIL. We introduce REMIX, a structured covariance modeling framework that enables scalable full-covariance modeling without the prohibitive cost of dense
matrix inversion and log-determinant computation. By leveraging a Laplace kernel parameterization, REMIX captures structured feature dependencies using
memory that scales linearly with the feature dimensionality, while requiring only an additional logarithmic factor in computation. Modeling these correlations produces more coherent synthetic samples and consistently improves performance across standard DFCIL
benchmarks. Our results demonstrate that moving beyond diagonal assumptions is essential for effective and scalable data-free continual learning. Our code is available at https://github. com/pkrukowski1/REMIX-Model-Inversion-via-Laplace-Kernel.
Read the full abstract ▾
All authors Patryk Krukowski, Jacek Tabor, Przemysław Spurek, Marek Śmieja, Łukasz Struski
Machine Learning
ROMER: Expert Replacement and Router Calibration for Robust MoE LLMs on Analog Compute-in-Memory Systems
Wenyong Zhou, Yuannuo Feng, Yizhe Chen et al. (+6 more)
Large
language models (LLMs) with mixture-of-experts (MoE) architectures achieve remarkable scalability by sparsely activating a subset of experts per
token, yet their frequent expert switching creates
memory bandwidth bottlenecks that compute-in-
memory (CIM) architectures are well-suited to mitigate. However, analog CIM systems suffer from inherent hardware imperfections that perturb stored weights, and its negative impact on MoE-based LLMs in noisy CIM environments remains unexplored. In this work, we present the first systematic investigation of MoE-based LLMs under noise model calibrated with real chip measurements, revealing that hardware noise critically disrupts expert load balance and renders clean-trained routing decisions consistently suboptimal. Based on these findings, we propose ROMER, a post-training calibration framework that (1) replaces underactivated experts with high-frequency ones to restore load balance, and (2) recalibrates router logits via percentile-based normalization to stabilize routing under noise. Extensive experiments across multiple
benchmarks demonstrate that ROMER achieves up to 58.6%, 58.8%, and 59.8% reduction in perplexity under real-chip noise conditions for DeepSeek-MoE, Qwen-MoE, and OLMoE, respectively, establishing its effectiveness and generalizability across diverse MoE architectures.
Read the full abstract ▾
All authors Wenyong Zhou, Yuannuo Feng, Yizhe Chen, Taiqiang Wu, Wendong Xu, Wenbo Qi, Zhengwu Liu, Wang Kang, Ngai Wong
Machine Learning · Cross-listed from cs.CE
Crash Assessment via Mesh-Based Graph Neural Networks and Physics-Aware Attention
Gabriel Curtosi, Carlos Manuel Ruiz Ruiz, Fabiola Cavaliere et al. (+1 more)
Full-vehicle crash simulations are computationally expensive, limiting their use in iterative design exploration. This work investigates learned hybrid surrogate models (MeshTransolver, MeshGeoTransolver, and MeshGeoFLARE) for predicting time-resolved structural deformation fields in an industrial lateral pole-impact
benchmark. We evaluate whether neural surrogates can reproduce full-field crash kinematics with sufficient accuracy, spatial regularity, and structural plausibility for engineering interpretation. The proposed architectures combine local mesh message passing, geometry-aware global
attention, and sparse contact-aware correction for autoregressive crash rollout. We compare mesh-based graph neural networks,
attention-based
geometric models, and hybrid architectures under a common training and hyperparameter configuration. The hybrid models capture both short-range structural interactions and long-range deformation patterns, while a sparse contact-aware variant assesses the effect of dynamic proximity interactions during rollout. On a 25-sample full-vehicle test set, the best hybrid model achieves a temporal mean root-mean-square error of 3.20 mm. While geometry-aware
attention baselines are quantitatively competitive, qualitative side-view inspection shows they can introduce local spatial noise and deformation irregularities that complicate structural interpretation. In contrast, hybrid mesh-
attention models provide the best balance between scalar accuracy, survival-space consistency, and physically interpretable displacement fields. These results suggest that crash surrogate assessment should combine global error metrics with downstream safety-relevant quantities and qualitative field inspection. The proposed methodology enables fast full-field predictions while preserving essential structural information for industrial crash-engineering analysis.
Read the full abstract ▾
All authors Gabriel Curtosi, Carlos Manuel Ruiz Ruiz, Fabiola Cavaliere, Xabier Larráyoz Izcara
Machine Learning
Entropy Polarity in Reinforcement Fine-Tuning: Direction, Asymmetry, and Control
Jiazheng Zhang, Ziche Fu, Junrui Shen et al. (+17 more)
Policy
entropy has emerged as a fundamental measure for understanding and controlling exploration in reinforcement learning with verifiable rewards (RLVR) for LLMs. However, existing
entropy-aware methods mainly regulate
entropy through global objectives, while the
token-level mechanism by which sampled policy updates reshape policy
entropy remains underexplored. In this work, we develop a theoretical framework of
entropy mechanics in RLVR. Our analysis yields a first-order approximation of the
entropy change, giving rise to
entropy polarity, a signed
token-level quantity that predicts how much a sampled update expands or contracts
entropy. This analysis further reveals a structural asymmetry: reinforcing frequent high-probability
tokens triggers contraction tendencies, whereas expansive tendencies typically require lower-probability samples or stronger distributional correction. Empirically, we show that
entropy polarity reliably predicts
entropy changes, and that positive and negative polarity branches play complementary roles in preserving exploration while strengthening exploitation. Building on these insights, we propose Polarity-Aware Policy Optimization (PAPO), which preserves both polarity branches and implements
entropy control through advantage reweighting. With the empirical
entropy trajectory as an online phase signal, PAPO adaptively reallocates optimization pressure between
entropy-expanding and
entropy-contracting updates. Experiments on mathematical
reasoning and agentic
benchmarks show that PAPO consistently outperforms competitive baselines, while delivering superior training efficiency and substantial reward improvements.
Read the full abstract ▾
All authors Jiazheng Zhang, Ziche Fu, Junrui Shen, Yunbin Zhao, Yunke Zhang, Zhiheng Xi, Long Ma, Chenxin An, Zhihao Zhang, Shichun Liu, Dingwei Zhu, Shihan Dou, Shaofan Liu, Han Li, Wiggin Zhou, Aiden Adams, Tao Gui, Fei Huang, Qi Zhang, Xuanjing Huang
Machine Learning · Cross-listed from cs.CL
From Token to Token Pair: Efficient Prompt Compression for Large Language Models in Clinical Prediction
Mingcheng Zhu, Zhiyao Luo, Yu Liu et al. (+1 more)
By processing electronic health records (EHRs) as natural
language sequences, large
language models (LLMs) have shown potential in clinical prediction tasks such as mortality prediction and phenotyping. However, longitudinal or highly frequent EHRs often yield excessively long
token sequences that result in high computational costs and even reduced performance. Existing solutions either add modules for compression or remove less important
tokens, which introduce additional inference latency or risk losing clinical information. To achieve lossless compression of
token sequences without additional cost or loss of performance, we propose Medical
Token-Pair Encoding (MedTPE), a layered method that extends standard tokenisation for EHR sequences. MedTPE merges frequently co-occurring medical
token pairs into composite
tokens, providing lossless compression while preserving the computational complexity through a dependency-aware replacement strategy. Only the
embeddings of the newly introduced
tokens of merely 0.5-1.0% of the LLM's parameters are fine-tuned via self-supervised learning. Experiments on real-world
datasets for two clinical scenarios demonstrate that MedTPE reduces input
token length by up to 31% and inference latency by 34-63%, while maintaining or even improving both predictive performance and output format compliance across multiple LLMs and four clinical prediction tasks. Furthermore, MedTPE demonstrates robustness across different input context lengths and generalisability to scientific and financial domains and different
languages.
Read the full abstract ▾
All authors Mingcheng Zhu, Zhiyao Luo, Yu Liu, Tingting Zhu
Machine Learning
Is Monotonic Sampling Necessary in Diffusion Models?
Muhammad Haris Khan
Diffusion models generate samples by iteratively denoising a Gaussian prior, traversing a sequence of noise levels that, in every published sampler, decreases monotonically. Six years of intensive work has refined nearly every aspect of this recipe, including the corruption operator, the training objective, the schedule shape, the architecture, and the ODE solver. Yet the assumption of monotonicity itself has never been systematically tested. Here we ask whether monotonic sampling is load-bearing or merely conventional. We design four families of structured nonmonotonic schedules and apply them to three architecturally distinct generative models, DDPM, EDM, and Flow Matching, across NFE budgets ranging from 10 to 200 function evaluations, plus a 42-
cell hyperparameter ablation, on CIFAR-10. Across all 90 tested configurations, no tested nonmonotonic schedule improves on the monotonic baseline. The magnitude of the penalty, however, spans nearly three orders of magnitude: persistent and substantial in DDPM, intermediate in Flow Matching, and indistinguishable from zero in EDM. We show that this variation is not noise but a structural property of each trained denoiser, and we formalize it as the Schedule Sensitivity Coefficient, a cheap, architecture-agnostic diagnostic that provides evidence of non-convergence to the Bayes-optimal denoiser at the critical noise level. Our findings justify the field's tacit reliance on monotonic schedules and supply a new probe of
diffusion model quality complementary to sample-quality metrics such as Frechet Inception Distance.
Read the full abstract ▾
Machine Learning
Decomposing the Generalization Gap in PROTAC Activity Prediction: Variance Attribution and the Inter-Laboratory Ceiling
Thor Klamt, Wolfgang Nejdl and Ming Tang
Machine-learning predictors of biochemical activity often exhibit large random-split-to-leave-one-target-out generalisation gaps that have been documented but not decomposed. We frame this as an evaluation-science question and use targeted
protein degradation as the empirical test bed. PROTACs (proteolysis-targeting chimeras) are heterobifunctional small molecules that induce targeted
protein degradation, with more than forty candidates currently in clinical trials; published predictors report AUROC of 0.85 to 0.91 under random-split cross-validation, while the leave-one-target-out (LOTO) protocol of Ribes et al. reduces performance to approximately 0.67. Random splits reward within-target interpolation, whereas LOTO measures the novel-target prediction that de-novo design depends on. We decompose this gap and identify inter-laboratory measurement variance as the dominant component, anchored by a within-target cross-laboratory cascade bounding the inter-laboratory contribution at 0.124 AUROC, well above the 0.05 contribution from binarisation-threshold choice. Across eight published architectures and ESM-2
protein language models up to 3B parameters, LOTO AUROC plateaus near 0.67, with a comparable plateau under SMILES-level deduplication; a 21-dimensional 2000-trial hyperparameter optimisation cannot break this ceiling, and the rank-1 single-seed configuration regresses by 0.161 AUROC under multi-seed validation, matching a closed-form selection-bias prediction (Bailey and Lopez de Prado, 2014). Few-shot k=5 stratified per-target retraining combined with ADMET features lifts 65-target LOTO AUROC from 0.668 to 0.7050, and post-hoc Platt scaling recovers raw output to within the 0.05 well-calibrated threshold. We release PROTAC-Bench (10,748 measurements, 173 targets, 65 LOTO folds), the variance-decomposition framework, the per-target calibration protocol, and the evaluation code.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CE
A nonlinear extension of parametric model embedding for dimensionality reduction in parametric shape design
Andrea Serani, Giorgio Palma and Matteo Diez
Dimensionality reduction is essential in simulation-based shape design, where high-dimensional parameterizations hinder optimization, surrogate modeling, and systematic design-space exploration. Parametric Model
Embedding (PME) addresses this issue by constructing reduced variables from
geometric information while preserving an explicit backmapping to the original design parameters. However, PME is intrinsically linear and may become inefficient when the sampled design space is governed by nonlinear
geometric variability. This paper introduces a nonlinear extension of PME, denoted NLPME. The proposed framework preserves the defining principle of PME – geometry-driven latent variables and parameter-mediated reconstruction – while replacing the linear reduced subspace with a nonlinear latent representation. Geometry is not reconstructed directly from the latent variables; instead, the latent representation is decoded into admissible design parameters, and the corresponding geometry is recovered through a forward parametric map. The method is assessed on a bio-inspired autonomous
underwater glider with a 32-dimensional parametric shape description and a CAD-based geometry-generation process. NLPME reaches a 5% reconstruction-error threshold with \(N=5\) latent variables, compared with \(N=8\) for linear PME, and a 1% threshold with \(N=9\), compared with \(N=15\) for PME. Comparison with a deep autoencoder shows that most of the nonlinear compression gain can be retained while preserving an explicit backmapping to the original design variables. The results establish NLPME as a compact, admissible, and engineering-compatible nonlinear reduced representation for parametric shape design spaces.
Read the full abstract ▾
Machine Learning
One-Step Generative Modeling via Wasserstein Gradient Flows
Jiaqi Han, Puheng Li, Qiushan Guo et al. (+3 more)
Diffusion models and flow-based methods have shown impressive generative capability, especially for images, but their sampling is expensive because it requires many iterative updates. We introduce W-Flow, a framework for training a generator that transforms samples from a simple reference distribution into samples from a target data distribution in a single step. This is achieved in two steps: we first define an
evolution from the reference distribution to the target distribution through a Wasserstein gradient flow that minimizes an energy functional; second, we train a static neural generator to compress this
evolution into one-step generation. We instantiate the energy functional with the Sinkhorn divergence, which yields an efficient optimal-transport-based update rule that captures global distributional discrepancy and improves coverage of the target distribution. We further prove that the finite-sample training dynamics converge to the continuous-time distributional dynamics under suitable assumptions. Empirically, W-Flow sets a new state of the art for one-step ImageNet 256$\times$256 generation, achieving 1.29 FID, with improved mode coverage and domain transfer. Compared to multi-step
diffusion models with similar FID scores, our method yields approximately 100$\times$ faster sampling. These results show that Wasserstein gradient flows provide a principled and effective foundation for fast and high-fidelity generative modeling.
Read the full abstract ▾
All authors Jiaqi Han, Puheng Li, Qiushan Guo, Renyuan Xu, Stefano Ermon, Emmanuel J. Candès
Machine Learning
Federated Client Selection under Partial Visibility: A POMDP Approach with Spatio-Temporal Attention
Qijun Hou, Yuchen Shi, Pingyi Fan et al. (+1 more)
Federated learning relies on effective client selection to alleviate the performance degradation caused by data heterogeneity. Most existing methods assume full visibility of all clients at each communication round. However, in large-scale or edge-based deployments, the server can only access a subset of clients due to communication, mobility, or availability constraints, resulting in partial visibility where only a subset of clients is observable for aggregation in each communication round. In this paper, we formulate
federated client selection under partial visibility as a Partially Observable Markov Decision Process (POMDP) and propose a Spatial-Temporal
attention-based reinforcement learning framework. By integrating historical global models and client identity
embeddings, the proposed method captures both the temporal contexts of training and the persistent characteristics of clients. Experimental results across multiple
datasets demonstrate that our approach achieves superior performance compared to existing baselines in heterogeneous and partially visible settings, validating its effectiveness in addressing the challenges of incomplete observations in practical
federated learning systems.
Read the full abstract ▾
All authors Qijun Hou, Yuchen Shi, Pingyi Fan, Khaled B. Letaief
Machine Learning
Learning Feature Encoder with Synthetic Anomalies for Weakly Supervised Graph Anomaly Detection
Yingjie Zhou, Yuqin Xie, Fanxing Liu et al. (+3 more)
Weakly supervised graph anomaly detection aims to unveil unusual graph instances, e.g., nodes, whose behaviors significantly differ from normal ones, given only a limited number of annotated anomalies and abundant unlabeled samples. A major challenge is to learn a meaningful latent feature representation that reduces intra-class variance among normal data while remaining highly sensitive to anomalies. Although recent works have applied self-supervised feature learning for graph anomaly detection, their strategies are not specifically tailored to its unique requirements, motivating our exploration of a more domain-specific approach. In this paper, we introduce a weakly supervised graph anomaly detection method that leverages a feature learning strategy tailored for graph anomalies. Our approach is built upon a multi-task learning scheme that extracts robust feature representations through synthesized anomalies. We generate synthetic anomalies by perturbing the normal graph in various ways and assign a dedicated detection head to each anomaly type, ensuring that learned features are sensitive to potential deviations from normal patterns. Although synthetic anomalies may not perfectly replicate real-world patterns, they provide valuable auxiliary data for effective feature learnin, much like features learned from ImageNet classification transfer to downstream
vision tasks. Additionally, we adopt a two-phase learning strategy: an initial warm-up phase using only synthetic samples, followed by a full-training phase integrating both tasks, to balance the influence of synthetic and real data. Extensive experiments on public
datasets demonstrate the superior performance of our method over its competitors. Code is available at https://github.com/yj-zhou/SAWGAD.
Read the full abstract ▾
All authors Yingjie Zhou, Yuqin Xie, Fanxing Liu, Dongjin Song, Ce Zhu, Lingqiao Liu
Machine Learning · Cross-listed from cs.CL
Training-Inference Consistent Segmented Execution for Long-Context LLMs
Xianpeng Shang, Jiang Li, Zehua Duo et al. (+2 more)
Transformer-based large
language models face severe scalability challenges in long-context generation due to the computational and
memory costs of full-context
attention. Under practical computation and
memory constraints, many inference-efficient long-context methods improve efficiency by adopting bounded-context or segment-level execution only during inference, while continuing to train models under full-context
attention, resulting in a mismatch between training and inference execution and state-transition semantics. Based on this insight, we propose a training-inference consistent segment-level generation framework, in which training and inference follow the same segment-level forward execution semantics. During training, consistency with inference is enforced by restricting gradient propagation to KV states carried over from the immediately preceding segment, while permitting head-specific access to past KV states during the forward pass without involving them in gradient propagation. Across long-context
benchmarks, our approach achieves performance comparable to full-context
attention, while achieving competitive latency-
memory trade-offs against strong inference-efficient baselines, and substantially improving scalability at very long context lengths (e.g., approximately 6x lower peak prefill
memory at 128K compared to full-context
attention with FlashAttention).
Read the full abstract ▾
All authors Xianpeng Shang, Jiang Li, Zehua Duo, Qianyi Cai, Xiangdong Su
Machine Learning · Cross-listed from cs.CV
WorldComp2D: Spatio-semantic Representations of Object Identity and Location from Local Views
SeongMin Jin and Doo Seok Jeong
Learning latent representations that capture both semantic and spatial information is central to efficient spatio-semantic
reasoning. However, many existing approaches rely on implicit latent structures combined with dense feature maps or task-specific heads, limiting computational efficiency and flexibility. We propose WorldComp2D, a novel lightweight representation learning framework that explicitly structures latent space geometry according to object identity and spatial proximity using multiscale local receptive fields. This framework consists of (i) a proximity-dependent encoder that maps a given observation into a spatio-semantic latent space and (ii) a localizer that infers the coordinates of objects in the input from the resulting spatio-semantic representation. Using facial landmark localization as a proof-of-concept, we show that, compared to SoTA lightweight models, WorldComp2D reduces the numbers of parameters and FLOPs by up to 4.0X and 2.2X, respectively, while maintaining real-time performance on CPU. These results demonstrate that explicitly structured latent spaces provide an efficient and general foundation for spatio-semantic
reasoning. This framework is open-sourced at https://github.com/JinSeongmin/WorldComp2D.
Read the full abstract ▾
Machine Learning
Online Continual Learning with Dynamic Label Hierarchies
Xinrui Wang, Shao-Yuan Li, Bartłomiej Twardowski et al. (+2 more)
Online Continual Learning (OCL) aims to learn from endless non\text{-}stationary data streams, yet most existing methods assume a flat label space and overlook the hierarchical organization of real\text{-}world concepts that evolves both horizontally (sibling classes) and vertically (coarse or fine categories). To better reflect this context, we introduce a new problem setting, DHOCL (Online Continual Learning from Dynamic Hierarchies), where taxonomies evolve across granularities and each sample provides supervision at a single hierarchical level. In this setting, we find two fundamental issues: (i) partial supervision under mixed granularities provides only point-wise signals over an evolving path-wise hierarchy, which constrains plasticity and undermines cross-level semantic consistency, and (ii) the dynamically evolving hierarchies induce granularity-dependent interference, destabilizing popular replay and regularization mechanisms and thereby exacerbating catastrophic forgetting. To tackle these issues, we propose
HALO (Hierarchical Adaptive Learning with Organized Prototypes), which adaptively combines complementary classification heads, regularized by organized learnable hierarchical prototypes, enabling rapid adaptation, hierarchical consistency, and structured knowledge consolidation as the taxonomy evolves. Extensive experiments on multiple
benchmarks demonstrate that
HALO consistently outperforms existing methods across hierarchical accuracy, mistake severity, and continual performance.
Read the full abstract ▾
All authors Xinrui Wang, Shao-Yuan Li, Bartłomiej Twardowski, Alexandra Gomez-Villa, Songcan Chen
Machine Learning
U-STS-LLM: A Unified Spatio-Temporal Steered Large Language Model for Traffic Prediction and Imputation
Yichen Zhang, Yiyang Ni, Pengcheng Xia et al. (+2 more)
Accurate spatio-temporal cellular traffic analysis is fundamental to the reliable operation of modern mobile networks, as it underpins load
forecasting for congestion avoidance and missing-data imputation for continuity despite transmission errors. While deeply connected,
forecasting and imputation have historically evolved as separate sub-fields. Existing models are typically task-specific, decouple temporal and spatial modeling, and rely on cascaded designs that accumulate errors and latency. Meanwhile, adapting pre-trained LLMs for this domain often leaves spatial structure under-guided and tends to overfit on limited data. To bridge these gaps, we propose U-STS-LLM, a unified framework built on a partially frozen spatio-temporally attentive (PFSTA) LLM backbone. A Dynamic Spatio-Temporal
Attention Bias Generator synthesizes a persistent functional graph with transient nodal states to steer
attention, while a lightweight parallel branch yields a stable local estimate fused via a Gated Adaptive Fusion, confining the LLM to residual correction. Trained under a unified multi-task objective, the model attains state-of-the-art results with low trainable-parameter overhead. Extensive experiments on real-world cellular traffic
datasets demonstrate the effectiveness and transferability of U-STS-LLM, achieving average RMSE improvements of 23.7% over the strongest baselines in
forecasting and 13.1% in imputation, with further gains exceeding 33% in zero-shot transfer to unseen network regions.
Read the full abstract ▾
All authors Yichen Zhang, Yiyang Ni, Pengcheng Xia, Jun Li, Chunguo Li
Machine Learning
Persona-Conditioned Adversarial Prompting: Multi-Identity Red-Teaming for Adversarial Discovery and Mitigation
Cristian Morasso, Anisa Halimi, Muhammad Zaid Hameed et al. (+1 more)
Automated red-teaming for LLMs often discovers narrow attack slices, missing diverse real-world threats, and yielding insufficient data for safety fine-tuning. We introduce Persona-Conditioned
Adversarial Prompting (PCAP), which conditions
adversarial search on diverse attacker personas (e.g., doctors, students, malicious actors) and strategy sets to explore realistic attack scenarios. By running parallel persona-conditioned searches, PCAP discovers transferable jailbreaks across different contexts and generates rich defense
datasets with automatic metadata tracking. On GPT-OSS 120B, PCAP increases attack success from 57% to 97% while producing 2-6$\times$ more diverse
prompts covering varied real-world scenarios. Critically, fine-tuning lightweight adapters on PCAP-generated data significantly improves model robustness (recall: 0.36 $\rightarrow$ 0.99, F1: 0.53 $\rightarrow$ 0.96) with minimal false positives, demonstrating a practical closed-loop approach from vulnerability discovery to automated
alignment.
Read the full abstract ▾
All authors Cristian Morasso, Anisa Halimi, Muhammad Zaid Hameed, Douglas Leith