Machine Learning
Towards Fast GNN Surrogates for CO2 Migration in Complex Geological Formations
Rodrigo S. Luna, Thiago H. N. Coelho, Luiz S. L. Neto et al. (+8 more)
This chapter discusses how a data-driven machine learning approach can reproduce key aspects of the physical behavior of multiphase flows in complex geological formations. We propose an end-to-end graph neural surrogate tailored to CO$_2$ plume migration
forecasting in geological storage. The method is evaluated on the SPE11A
benchmark, a well-known industry test case designed to assess CO$_2$ storage scenarios and characterized by sharp gas-water interfaces, strong advective transport, and rapid convective mixing with fingering development. The
benchmark is reformulated as a graph in which nodes represent computational
cells and edges encode transmissibility-based interactions enriched with
geometric attributes. Directional transport arising from grid geometry, permeability contrasts, and geological heterogeneity is captured through an anisotropic message-passing mechanism, where interaction weights are computed via geometry-conditioned edge
embeddings, biasing message aggregation toward physically relevant transport directions. Temporal
evolution is modeled in latent space using an autoregressive residual formulation trained with multi-step supervision. The proposed model produces competitive forecasts of gas saturation and
liquid-phase density, which are key indicators for CO$_2$ storage monitoring, with cumulative errors that remain moderate over extended
forecasting horizons.
Read the full abstract ▾
All authors Rodrigo S. Luna, Thiago H. N. Coelho, Luiz S. L. Neto, Roberto M. Velho, Adriano M. A. Cortes, Renato N. Elias, Alexandre G. Evsukoff, Fernando A. Rochinha, Mauricio Araya-Polo, Herve Gross, Alvaro L. G. A. Coutinho
Machine Learning · Cross-listed from q-bio.QM
Agentic Discovery of Non-Canonical Antimicrobial Peptides with AMPGAN v3
Jay Jung, Xiaohan Zhang, Shenghan Song et al. (+8 more)
Antimicrobial resistance causes to over a million deaths annually. Antimicrobial peptides (AMPs) are a promising solution, but generative AMP models are not yet ready to design peptides with non-natural amino acids and/or chemical modifications, which are essential for real-world peptide drugs. We present AMPGAN v3, a multi-objective conditional GAN that expands the generative vocabulary to D-amino acids and N/C-terminus modifications such as amidation. By separating
adversarial and activity-aware supervision across two specialized discriminators, AMPGAN v3 substantially improves training stability and outperforms prior generative AMP models on external classifiers. We validated five candidates spanning three structural classes in vitro; two showed activity against Gram-positive strains, with the best candidate reaching MIC 8 {\mu}g/mL against B. subtilis. To support downstream curation, we further present PepCraft, a multi-
agent framework for end-to-end AMP discovery in which a Planning
Agent orchestrates specialized executors for generation, filtering, and verification. Its prioritization recommendations align with our in vitro outcomes. Together, these contributions let us examine, on a small but real scale, how generative and agentic AI compose in therapeutic peptide discovery. Code: https://github.com/marszzibros/AMPGANv3
Read the full abstract ▾
All authors Jay Jung, Xiaohan Zhang, Shenghan Song, Mahmoud Sayedahmed, Chijian Xiang, Yunong Xu, Ahmed AbdelKhalek, Severin T. Schneebeli, Matthew J. Wargo, Jianing Li, Safwan Wshah
Machine Learning · Cross-listed from cs.CR
TrustErase: Auditable Instant Machine Unlearning with Passport-Embedded Representations
Rutger Hendrix, Leonardo G. Russo, Concetto Spampinato et al. (+2 more)
The demand for
privacy-compliant AI has amplified the need for machine unlearning; yet, existing retraining or
distillation-based methods remain unverifiable and computationally costly. We introduce TrustErase, a verifiable, data-free unlearning framework leveraging passport-embedded representations for instant, modular, and auditable forgetting. By treating passports as cryptographic keys within parameter-efficient adaptation layers, TrustErase enables the removal of specific classes or
datasets through simple deactivation, without retraining, fine-tuning, or access to the original data. A singular value based decomposition conceals passports within model weights, ensuring that unlearning actions remain transparent and provably compliant. Evaluations on MNIST, CIFAR10 and CIFAR100 show that TrustErase matches or exceeds state-of-the-art
benchmarks such as DELETE, L2UL, and Boundary Shrink, while operating in a strictly data-free regime. Ultimately, TrustErase establishes a new paradigm for trustworthy, accountable, and instantly forgettable AI systems.
Read the full abstract ▾
All authors Rutger Hendrix, Leonardo G. Russo, Concetto Spampinato, Matteo Pennisi, Giovanni Bellitto
Machine Learning · Cross-listed from stat.AP
Regularized Machine Learning for System Identification of Ship Free-Running Manoeuvres from CFD-Based Synthetic Data: A Comparative Study
R.F. Suárez, J.C. Berndt and M. Abdel-Maksoud
This study investigates supervised machine learning techniques for identifying ship hydrodynamic coefficients from CFD-generated data from free-running simulations. Specifically, ordinary least squares and regularized regression methods are applied to Abkowitz-type manoeuvring models. Training and validation
datasets are derived from URANS simulations of zig-zag and turning circle manoeuvres, which are validated against experimental
benchmark data. The analysis evaluates the effects of coefficient set size, minimum training length required for predictive model training, and manoeuvre combinations on model performance. Results demonstrate the suitability of large-angle zig-zag manoeuvres for hydrodynamic system identification, provided that multicollinearity is addressed through appropriate coefficient selection, regression models, or input data variability. Larger coefficient sets offer greater model flexibility for variable conditions but are more prone to multicollinearity. Regularized regression techniques effectively mitigate multicollinearity and notably enhance prediction accuracy, as does incorporating more diverse manoeuvring data. Among tested models, Ridge regression provided the best compromise between computational efficiency and prediction accuracy.
Read the full abstract ▾
Machine Learning
Noise-Driven Escape from Metastable Phases explains Grokking in Deep Neural Networks
Ibrahim Talha Ersoy and Karoline Wiesner
Deep neural networks (DNNs) exhibit first order phase transitions under variations of the L2 regularization strength, with each transition marking the onset of a new learnable feature. Below a critical regularization strength, all features are in principle learnable, but coexisting metastable states, separated by energy barriers, can trap the network and impede convergence. A strength of DNNs is their ability to generalize. But many open questions remain, among them the origin of so called grokking: the abrupt, delayed onset of generalization after prolonged apparent overfitting. We show for linear DNNs that grokking is consistent with hysteresis in first-order L2 phase transitions: using L2 regularization to engineer deliberate trapping, we demonstrate that a model in a low-accuracy metastable state escapes only when SGD noise drives it across an energy barrier, with escape times following Arrhenius scaling. We reproduce grokking-like delayed convergence across two orders of magnitude in escape time by deliberately trapping models in metastable phases. Using sparse sub-sampling we also reproduce the canonical grokking curve where test error eventually approaches the final training error. Our work suggests that the number of metastable states equals the number of learnable features – one per singular value of the data covariance – the potential for hysteresis grows naturally with task complexity. We provide evidence that the same mechanism likely operates in general nonlinear DNNs. Our results provide routes toward more efficient learning schemes.
Read the full abstract ▾
Machine Learning
MODE: Modality-Decomposed Expert-Level Mixed-Precision Quantization for MoE Multimodal LLMs
Yuanteng Chen, Nanxin Zeng, Peisong Wang et al. (+9 more)
Mixture-of-Experts
Multimodal Large
Language Models (MoE-MLLMs) offer remarkable performance but incur prohibitive GPU
memory costs, making compression essential. Among PTQ methods, expert-level mixed-precision quantization has proven effective for MoE-LLMs, yet suffers notable degradation on MoE-MLLMs due to two overlooked biases in expert importance estimation. (1) At the cross-modal level, the numerical dominance of
vision tokens causes expert selection frequency to be dominated by
vision tokens, masking experts that are critical to the text modality; (2) at the intra-
vision level, the large proportion of redundant
vision tokens further skew frequency statistics, obscuring experts critical for informative visual content. To bridge gaps, we propose MODE, a modality-decomposed expert-level mixed-precision quantization framework for MoE-MLLMs that decomposes expert selection frequency by modality, filters redundant
vision tokens to obtain denoised visual frequency, and further evaluates quantization sensitivity per modality as a complementary signal to frequency-based estimation. These signals are integrated into an Integer Linear Programming formulation to assign per-expert bit-widths under a given budget. Extensive experiments show that MODE is particularly well-suited for MoE-MLLMs, limiting average performance loss to within 2.9% at W3A16, with larger gains at the extreme 2-bit setting.
Read the full abstract ▾
All authors Yuanteng Chen, Nanxin Zeng, Peisong Wang, Zhilei Liu, Yuantian Shao, Shiqiang Lang, Tao Liu, Chuangyi Li, Qinghao Hu, Gang Li, Jing Liu, Jian Cheng
Machine Learning
Probing, Fusion, and Trustworthiness: A Systematic Evaluation of Foundation Model Representations for Multimodal Cancer Analysis
Jingyu Hu, Giuseppe Tripodi, Reed Naidoo et al. (+2 more)
Foundation models (FMs) have emerged as powerful representation extractors for medical data, yet their generalizability to
datasets under distribution shift remains underexplored. This work systematically evaluates FM-based representations on a suite of computational pathology tasks across two real-world commercial cohorts, IH-BC and IH-NSCLC, drawn from the licensed in-house (IH) oncology
dataset. The analysis focuses on two modalities, whole-slide images and transcriptomic profiles, drawn from the IH
multimodal data. We first
benchmark unimodal probing performance across five FMs on eight downstream classification tasks, and find that image and omics representations carry complementary predictive signals. Then we investigate whether
multimodal fusion can yield additional gains over unimodal baselines by comparing three image-omics fusion strategies built on paired representations. The trustworthiness of selected unimodal and
multimodal pipelines is further assessed through conformal prediction. Our results show that FM representations achieve competitive performance on out-of-distribution data and that
multimodal fusion helps mainly when no single modality dominates the signal. Conformal prediction reveals that in the majority of cases where a point prediction fails, the true diagnosis remains recoverable within the prediction set, reinforcing the value of uncertainty-aware inference for clinical support.
Read the full abstract ▾
All authors Jingyu Hu, Giuseppe Tripodi, Reed Naidoo, Sarah F. McGough, Tapabrata Chakraborti
Machine Learning
The Critical Role of Model Selection in Causal Inference: A Comparative Analysis of Classification Models within the InferBERT Framework for Pharmacovigilance
Csaba Kiss, Roland Molontay and Gabriele Pergola
Distinguishing causal adverse drug events (ADEs) from spurious correlations remains a central challenge in pharmacovigilance. The InferBERT framework integrates
transformer models with Do-calculus, but its success hinges on the underlying classification model. This study evaluates the impact of model choice in InferBERT, assessing whether simpler models suffice, if domain-specific pre-training helps, whether scaling to LLMs improves causal detection, and the effect of post-hoc calibration. We performed a comparative study on two
benchmarks: Analgesics-induced Acute Liver Failure (AILF) and Tramadol-related Mortalities (TRAM). Four models were evaluated-XGBoost (baseline), ALBERT (original InferBERT), BioBERT (biomedical
transformer), and Med-LLaMA (medical LLM)-using 5-fold cross-validation repeated over 20 runs. We measured accuracy, Expected Calibration Error (ECE) pre- and post-isotonic regression, and Jaccard concordance of causal terms with PRR, ROR, and EBGM; significance was tested with paired t-tests. BioBERT achieved the highest accuracy on both
datasets, while Med-LLaMA underperformed despite its size and parameter-efficient fine-tuning. Domain-specific pre-training was decisive. Calibration improved ECE but had mixed effects on accuracy and causal discovery. BioBERT's superiority also yielded the strongest concordance with traditional pharmacovigilance signals. These results show that domain-specific pre-training provides a clear advantage over simpler baselines and larger LLMs. Investing in manageable, domain-aware models is more effective for computational pharmacovigilance than simply scaling model size.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CR
Loss Landscape Poisoning: Targeted Extraction of Unseen Training Data from LLMs
Md Abdullah Al Mamun, Ngoc Phu Doan, Pedram Zaree et al. (+2 more)
Large
Language Models are increasingly trained on proprietary or sensitive data, from private healthcare and financial records to user conversations containing secrets. Ensuring the
privacy of such data against extraction attacks has become a central concern. In this paper, we ask whether an attacker who can poison a portion of the training data can facilitate the leakage of a separate target record they have no access to. We answer in the affirmative and show that such leakage can be induced by a poisoning mechanism that reshapes the model's local loss landscape around the target completion. Our key insight is that poisoning to create a sharp loss minimum at the target, surrounded by elevated loss on nearby alternatives, forces the model to memorize the target as the unique low-loss solution in its neighborhood. The attack requires no architectural changes, and generalizes across centralized and
federated learning settings. We demonstrate that the attack amplifies
privacy leakage across
language (up to 100% successful extraction), and
vision-
language models (up 90% successful extraction). We show that the attack is thwarted when the model is trained to be differentially private. However, we introduce a new attack that directly probes the loss landscape bypassing even differential
privacy defenses.
Read the full abstract ▾
All authors Md Abdullah Al Mamun, Ngoc Phu Doan, Pedram Zaree, Nael Abu-Ghazaleh, Ihsen Alouani
Machine Learning · Cross-listed from cs.CR
Timestamp-Aware Spatio-Temporal Graph Contrastive Learning for Network Intrusion Detection
Jianli Dai, Guangwei Wu, Jiacheng Li et al. (+3 more)
Given their effectiveness in modeling the relational structure among network traffic flows, graph neural networks (GNNs) have been widely adopted in network intrusion detection systems (NIDSs). However, most existing GNN-based NIDS approaches focus on the relational structure of traffic flows, and treat them as temporally independent, which limits their ability to cope with evolving attack behaviors. Moreover, their reliance on supervised or semi-supervised learning often restricts generalization to unseen attacks. To address these limitations, we propose a novel self-supervised GNN-based framework. To the best of our knowledge, the proposed model is among the first self-supervised GNN-based NIDS models to explicitly leverage real timestamps, which provides faithful temporal dependencies for representation learning. We first construct a series of temporal graphs from network traffic flows according to their timestamps, and then employ an E-GraphSAGE and LSTM based encoder to fully extract temporal information and spatial dependencies of network traffic, without introducing time-costly
attention mechanisms. A multi-view graph contrastive learning (GCL) scheme is introduced, where temporal, spatial, and feature contrasts are jointly performed to capture temporal continuity, preserve structural consistency, and improve the generalization and robustness of the learned representations, respectively. In addition, a gradient-norm-based adaptive weighting strategy is designed to optimize the contrastive loss weights. Experimental results on four representative NIDS
datasets with real timestamps demonstrate that our method significantly outperforms existing self-supervised approaches and achieves performance comparable to the supervised state-of-the-art GNN method, while maintaining high computational efficiency.
Read the full abstract ▾
All authors Jianli Dai, Guangwei Wu, Jiacheng Li, Weiping Wang, An He, Xinjun Xiao
Machine Learning · Cross-listed from cs.RO
Human Universal Grasping
Kevin Yuanbo Wu, Tianxing Zhou, Isaac Tu et al. (+5 more)
Humans can grasp objects effortlessly, whereas multi-fingered
robots are far from this level of generality. We argue that the most natural source of
robot grasping data is from humans, who pick up thousands of objects every day. We present HUG, a flow-matching model that generates diverse human grasps for any user-specified object in a single RGB-D image captured from a stereo camera. Using smart glasses, we first collect 1M-HUGs, an
egocentric dataset of human grasps spanning 1M frames (27.8 hrs) and 6,707 object instances across 41 buildings. Next, to model the distribution of natural human grasps, our novel flow-matching model fuses RGB and depth observations to output a grasp parameterized by wrist translation, wrist rotation, and MANO hand pose. Predicted grasps can be retargeted to various
robot hands, enabling zero-shot grasping in everyday scenes. To standardize evaluation, we build a new simulated
benchmark, HUG-Bench, of 90 unseen objects from five
geometric categories and various sizes, with metric-scale 3D meshes. We evaluate HUG in the real world on the 30-object test set of HUG-Bench across multiple stereo cameras,
robot embodiments, and household environments. HUG outperforms the state-of-the-art grasping baselines by +23% and +34% on our challenging object set. Code, data,
benchmark, checkpoints, and an interactive demo are released on our website: https://grasping.io/
Read the full abstract ▾
All authors Kevin Yuanbo Wu, Tianxing Zhou, Isaac Tu, Billy Yan, Irmak Guzey, David Fouhey, Dandan Shan, Lerrel Pinto
Machine Learning
Exact Posterior Score Estimation for Solving Linear Inverse Problems
Abbas Mammadov, Ozgur Kara, Kaan Oktay et al. (+5 more)
Diffusion and flow-based models learn powerful data priors by training a denoiser to reverse Gaussian corruption. To use this prior to solve a linear inverse problem, one needs to sample from the posterior, but the score that the prior provides is the unconditional score, not the posterior score. Existing methods either steer a fixed pretrained denoiser with approximate measurement-matching corrections, or train a conditional restoration model that abandons the denoising structure of the prior. We derive the exact posterior score in closed form for linear Gaussian inverse problems under general Gaussian interpolants, and show that posterior sampling reduces to a denoising problem at an operator-dependent shifted pivot under an anisotropic noise covariance. We turn this identity into Exact Posterior Score (EPS), a denoising training objective that preserves the input/output structure of standard pretraining and can therefore be trained from scratch or fine-tuned from a pretrained denoiser. At inference, EPS uses the same sampler as the underlying backbone, with no likelihood gradients or projections. We evaluate EPS on five linear inverse problems across FFHQ and ImageNet, where it outperforms training-free and training-based baselines on fidelity, perceptual, and distributional metrics, while using roughly an order of magnitude fewer denoiser evaluations than gradient-based posterior samplers.
Read the full abstract ▾
All authors Abbas Mammadov, Ozgur Kara, Kaan Oktay, Iskander Azangulov, Adil Kaan Akan, Hyungjin Chung, James Matthew Rehg, Yee Whye Teh
Machine Learning · Cross-listed from cs.RO
Geometric Action Model for Robot Policy Learning
Jisang Han, Seonghu Jeon, Jaewoo Jung et al. (+7 more)
Generalist
robot policies must follow user instructions while
reasoning about how objects, cameras, and
robot actions interact in the 3D physical world. Recent
vision-
language-action models (VLAs) and
video world-action models (WAMs) inherit strong semantic or temporal priors from large-scale foundation models, but they still operate primarily on 2D image frames or 2D-derived latent spaces, leaving implicit the 3D geometry required for contact-rich manipulation. We propose the
Geometric Action Model (GAM), a
language-conditioned manipulation policy that directly repurposes a pretrained
geometric foundation model (GFM) as a shared substrate for perception, temporal prediction, and action decoding. GAM splits the GFM at an intermediate layer: the shallow layers serve as an observation encoder, and a causal future predictor inserted at the split layer forecasts future latent
tokens conditioned on
language, proprioception, and action history. The predicted future
tokens are then routed through the remaining GFM blocks for feature propagation and decoding, allowing a single backbone to produce both future geometry and actions. This design equips the GFM with
language-conditioned temporal world modeling through minimal architectural modification while preserving its rich
geometric priors. Across a broad suite of simulation and real-
robot manipulation
benchmarks, GAM is more accurate, more robust, faster, and lighter than current foundation-model-scale baselines.
Read the full abstract ▾
All authors Jisang Han, Seonghu Jeon, Jaewoo Jung, René Zurbrügg, Honggyu An, Tifanny Portela, Marco Hutter, Marc Pollefeys, Seungryong Kim, Sunghwan Hong
Machine Learning · Cross-listed from cs.RO
Hierarchical Advantage Weighting for Online RL Fine-Tuning of VLAs from Sparse Episode Outcomes
Tongyan Fang, Siyuan Huang, Naiyu Fang et al. (+6 more)
When pretrained VLA policies are fine-tuned through online RL, each rollout episode produces only a single binary outcome (success or failure), yet the actor update requires per-transition supervision. Existing approaches commonly reduce this sparse outcome to a single scalar reward or advantage signal, which conflates distinct forms of transition-level feedback and provides limited guidance once basic task success becomes achievable. First, a single scalar signal conflates the two objectives of viability and efficiency; once basic success is achieved, the binary label provides no gradient to distinguish efficient completions from slow ones. Second, real-world rollouts mix autonomous and intervention segments; naively assigning episode outcomes across these boundaries introduces incorrect credit assignment. To address these issues, we propose Hierarchical Advantage-Weighted Behavior Cloning (HABC), which trains separate critic heads for these two objectives on different data subsets and combines their outputs with a state-adaptive balance. A state-adaptive gate $g_t$ merges their one-step advantages, prioritizing viability when success is uncertain and shifting to efficiency only when viability is high, and converts the result into per-transition weights on the actor loss. Intervention-aware credit assignment further restricts outcome labels to segments executed by the current policy, preventing supervision from leaking across intervention boundaries. In real-
robot experiments on three contact-rich bimanual tasks, HABC raises success from supervised fine-tuning (SFT) baselines of 36%, 44%, and 12% to 92%, 88%, and 38%.
Read the full abstract ▾
All authors Tongyan Fang, Siyuan Huang, Naiyu Fang, Ganlong Zhao, Zhongjin Luo, Jianbo Liu, Xiaogang Wang, Ying Dong, Hongsheng Li
Machine Learning · Cross-listed from cs.CV
The Importance of Phase in Neural Representations: An Internal Oppenheim-Lim Test of Image Classifiers
Alper Y{\i}ld{\i}r{\i}m
Oppenheim and Lim (1981) showed that natural images stay recognizable when reconstructed from their
Fourier phase alone, while the magnitude carries little of their identity. We ask whether trained image classifiers reproduce this asymmetry inside their hidden layers, and we test it causally: given two images, we transplant the phase of one onto the magnitude of the other at a chosen layer and record which image the prediction follows. In PRISM2D, GFNet, and ViT-B/16 the prediction follows the phase or sign donor, and deleting all image-specific magnitude barely moves accuracy, so identity rides on phase while image-specific magnitude is largely dispensable to the readout. ResNet-50 at first seems to break the pattern, because transplanting sign after its ReLUs does nothing; a fair intervention before the ReLU reveals a strong latent sign code in the late blocks, and a DC-only control shows the readout consumes a channel-wise spatial average. Controls rule out the trivial case in which magnitude simply stops depending on the image. The architectures therefore share a phase/sign identity code but expose it in different bases, set by rectification and readout geometry, which gives a mechanistic account of the texture–shape gap between CNNs and
attention models.
Read the full abstract ▾
Machine Learning
Your Privacy My Cloak: Backdoor Attacks on Differentially Private Federated Learning
Xiaolin Li, Ning Wang, Ninghui Li et al. (+1 more)
Prior research suggests that differential
privacy (DP) can enhance the robustness of
federated learning (FL) against
backdoor attacks. In this paper, we challenge this assumption. Through an empirical analysis of two baseline attack strategies, we uncover a fundamental tension in DP-FL: while bypassing DP allows state-of-the-art defenses to detect and filter malicious updates, complying with DP inadvertently masks their distinguishing statistical characteristics. Consequently, existing defenses become ineffective as DP reduces the raw
backdoor signal. Building on this masking effect, we propose RING, a novel attack that explicitly exploits DP to conceal malicious contributions while maximizing attack impact. By collaboratively crafting
adversarial perturbations, compromised clients reconstruct a strong
backdoor signal during aggregation without triggering anomaly detection. RING operates as a perturbation layer that is agnostic to the underlying
backdoor technique, making it broadly applicable and composable with existing attacks – a property that significantly amplifies the threat it poses to DP-FL. Extensive evaluations across four image and text
datasets under non-iid distributions show that RING achieves an average attack success rate of 90.3% against six state-of-the-art defenses under a moderate
privacy budget, an improvement of up to 38.4% over baseline strategies. Finally, we evaluate potential countermeasures and find that mitigating this threat incurs significant utility trade-offs, exposing a fundamental security gap in the deployment of differentially private FL.
Read the full abstract ▾
All authors Xiaolin Li, Ning Wang, Ninghui Li, Wenhai Sun
Machine Learning · Cross-listed from cs.CL
KVEraser: Learning to Steer KV Cache for Efficient Localized Context Erasing
Mufei Li, Shikun Liu, Dongqi Fu et al. (+5 more)
Post-hoc context erasing over the KV cache is challenging because a local edit has a global consequence: once a span has been processed, its influence propagates into the cached states of all subsequent
tokens. This issue arises naturally in long-context LLM applications, where stale, incorrect, or harmful context may be identified only after prefill. Exact erasing must then recompute all
tokens after the deleted span, making its computational cost depend on suffix length rather than erased-span length. We introduce KVEraser, a learned KV-cache editing method for efficient localized context erasing. KVEraser replaces the KV states of the erased interval with learned steering states while reusing the remaining cache unchanged. To learn a transferable erasing mechanism, we use a two-stage pipeline: generic span-neighbor pre-training followed by task-specific fine-tuning. Experiments show that KVEraser nearly matches full recomputation in post-erasure performance on in-domain tasks across 1K-32K contexts, while its latency increases by only 29.6% compared with a 17.6x increase for full recomputation. KVEraser also generalizes to unseen long-document QA with harmful factual distractors and tool-selection with malicious skill-file injections, achieving the best performance among approximate baselines with a 2.9-10.7x speedup over full recomputation. Notably, full-parameter eraser training is unnecessary: a rank-16 LoRA eraser, which trains only approximately 0.6% as many parameters as the generator, performs comparably to or better than its full-parameter counterpart.
Read the full abstract ▾
All authors Mufei Li, Shikun Liu, Dongqi Fu, Haoyu Wang, Yinglong Xia, Hong Li, Hong Yan, Pan Li
Machine Learning
HAMON: Passive Optical Sequence Mixing for Long-Horizon Forecasting
Alper Y{\i}ld{\i}r{\i}m
Simple linear and frequency-domain models remain surprisingly competitive in long-horizon time-series
forecasting, and recent mechanistic evidence suggests that standard
forecasting benchmarks may not require the dense superposed representations that make
transformers powerful in other domains. This raises a substrate-level question: if the core
forecasting operator is often low-complexity and approximately linear, does it need to be implemented as learned digital temporal mixing? We introduce HAMON, a passive diffractive optical
forecasting core in which historical values are encoded onto an optical aperture, future positions are left dark, and cascaded trainable phase masks with free-space diffraction shape the forecast directly in the output field. At inference, prediction is performed by a single passive optical propagation pass with no trainable digital sequence-mixing layer. Across standard
benchmarks, HAMON outperforms the strongest digital baselines considered on ETTm2 at all horizons and on ETTh2 at all but the longest horizon, improving MSE by up to 14% and doing so consistently across horizons rather than at isolated points. It is competitive on Weather and trails the strongest baselines on the remaining ETT settings and on the high-channel-count Traffic and Electricity
datasets. Phase encoding, intensity-compatible readout, and phase-scrambling ablations, together with a TorchOptics cross-simulator check, indicate that the forecasts arise from the data-bearing optical field rather than from a digital
forecasting head. Because the passive core uses standard
Fourier optics, HAMON defines a concrete target for optical hardware and for passive physical sequence mixing.
Read the full abstract ▾
Machine Learning
ExpRL: Exploratory RL for LLM Mid-Training
Violet Xiang, Amrith Setlur, Chase Blagden et al. (+2 more)
Sparse reward reinforcement learning (RL) has become a standard tool for improving LLM
reasoning, but its success depends critically on the coverage present in the base model. In practice, models are often primed for RL through mid-training on curated
reasoning traces that teach useful primitive skills such as decomposition, verification, or self-correction. Although effective, this strategy requires manually specifying what the model should learn, and it remains unclear whether such primitive coverage is enough for much harder problems, which require combining these skills into broader solution strategies. We study a more automated approach: RL-based mid-training using large corpora of human-written question-answer data. Rather than treating reference solutions as targets to imitate, our method, ExpRL, uses them as reward scaffolds: references are hidden from the policy and used only to construct problem-specific grading rubrics for judging on-policy
reasoning traces. The policy samples from the original problem
prompt, while an LLM judge compares the sampled
reasoning trace against the reference solution and assigns outcome-level or process-level dense rewards. This lets ExpRL reinforce partial progress, useful intermediate reductions, and productive
reasoning behaviors that sparse final-answer rewards often fail to upweight. On challenging math
reasoning tasks, ExpRL yields stronger RL priming than SFT, sparse-reward GRPO, and self-
distillation, and provides a better initialization for subsequent sparse-reward RL. Additional mixed-domain experiments further suggest that ExpRL can extend beyond the original math-only setting.
Read the full abstract ▾
All authors Violet Xiang, Amrith Setlur, Chase Blagden, Nick Haber, Aviral Kumar
Machine Learning · Cross-listed from math.ST
Learning the Geometry of Data: A Mathematical Review of Shape Space Analysis
Gary P. T. Choi, Khanh Dao Duc, Shira Faigenbaum-Golovin et al. (+6 more)
A central objective of machine learning is to identify structure and patterns in data. Advances in data acquisition have increasingly produced
datasets whose observations possess rich
geometric form, giving rise to shape spaces that encode variability in object geometry. Such
datasets arise across a wide range of disciplines, including biology, medicine, anthropology, and computer
vision, where subtle
geometric differences often carry important scientific information. Traditional machine learning methods, however, are frequently ill-equipped to account for the nonlinear
geometric structure underlying these data. This survey synthesizes a rapidly growing body of work on shape space analysis, which provides a mathematical and computational framework for the study of
geometric data. Drawing on ideas from differential geometry, statistics, and machine learning, we organize the literature around a common analytical pipeline: shape representation and parameterization, the rigorous construction of robust geodesic metrics, statistical analysis on shape spaces, and geometry-aware learning methods. We discuss how these tools enable the characterization of shape variability, the comparison of
geometric objects, and the analysis of structural trajectories across populations and time. To illustrate the breadth of the field, we highlight applications spanning multiple scales of biological organization, including studies of subcellular morphology and primate tooth
evolution. Across these and many other domains, researchers face common challenges arising from complex, nonlinear, and often unaligned
geometric variation. The review concludes by identifying key theoretical and computational challenges, as well as emerging opportunities driven by increasingly large and diverse
geometric datasets.
Read the full abstract ▾
All authors Gary P. T. Choi, Khanh Dao Duc, Shira Faigenbaum-Golovin, Karen Habermann, Emmanuel Hartman, Christoph von Tycowicz, Chi Zhang, Wenjun Zhao, Felix Zhou
Machine Learning · Cross-listed from cs.CL
TokenPilot: Cache-Efficient Context Management for LLM Agents
Buqiang Xu, Zirui Xue, Dianmou Chen et al. (+12 more)
As LLM
agents are deployed in long-horizon sessions, context accumulation drives up inference costs. Existing approaches utilize text
pruning or dynamic
memory eviction to minimize
token footprints; however, their unconstrained sequence mutations alter layouts, introducing prefix mismatches and cache invalidation. This reveals a critical trade-off between text sparsity and
prompt cache continuity. To address this, we present TokenPilot, a dual-granularity context management framework. Globally, Ingestion-Aware Compaction acts as a framework harness to stabilize
prompt prefixes and eliminate open-world environmental noise at the ingestion gate. Locally, Lifecycle-Aware Eviction monitors the ongoing residual utility of context segments, enforcing a conservative batch-turn schedule to offload content segments only when task relevance expires. Experiments on PinchBench and Claw-Eval under both isolated and continuous modes demonstrate that TokenPilot reduces costs by 61% and 56% in isolated mode, and 61% and 87% in continuous mode, while maintaining competitive performance compared to prior systems. TokenPilot has been integrated into LightRSI at https://github.com/zjunlp/RSI.
Read the full abstract ▾
All authors Buqiang Xu, Zirui Xue, Dianmou Chen, Chenyang Fu, Chiyu Wu, Caiying Huang, Chen Jiang, Jizhan Fang, Xinle Deng, Yijun Chen, Yunzhi Yao, Xuehai Wang, Jin Shang, Gong Yu, Ningyu Zhang
Machine Learning
Filtered Conformal Ellipsoids for Graph-Native Time Series
Yannick Limmer
Joint prediction sets for multivariate time series should control a single event while adapting to cross-coordinate dependence. We study filtered conformal ellipsoids: a frozen state-space filter emits a one-step predictive mean and covariance, and split-conformal calibration is applied to the resulting Mahalanobis scores. The filter is used to choose the ellipsoid shape; conformal calibration chooses the scalar radius, so the construction benefits from a learned predictive covariance without relying on Gaussian tail probabilities for coverage. The main difficulty is that filtered scores are dependent and learned recurrent filters need not contract in their raw hidden state; we therefore analyse contraction in an observable predictive-law quotient that identifies hidden states producing the same future sequence of emitted Gaussian laws. Under a stable Bayes Gaussian-projection filter, covariance bounds, and a finite-horizon observability Fisher condition, small excess Gaussian negative log-likelihood implies contraction of the learned emitted laws. Combined with a threshold-autocovariance envelope this yields a Chebyshev-type approximate coverage bound for filtered split-conformal prediction under dependence; a sharper Bernstein-type bound requires an additional
geometric-mixing concentration assumption. Under Gaussian oracle realisability we also obtain a near-oracle log-volume comparison within the class of conditionally valid Gaussian ellipsoid rules. We instantiate the framework with a GCN-GRU filter with diagonal-plus-low-rank covariance. On moderate-size graph-native traffic
benchmarks (METRLA-$20$ and PEMSBAY-$50$), the learned filter gives sharper at-target ellipsoids than static-covariance and non-filter baselines; at full-graph scale and on non-graph-native
datasets, factor and copula baselines can be stronger.
Read the full abstract ▾
Machine Learning · Cross-listed from math.OC
Exploding and vanishing gradients in deep neural networks: the effect of residual connections
Vivek S Borkar
The well known phenomenon of exploding and vanishing gradients in deep neural networks is analyzed using multiplicative ergodic theory. The effect of adding a residual connection is explained in this context. Specifically, a characterization of Liapunov exponents due to Furstenberg and Kifer is exploited in order to make a precise statement about the Liapunov spectrum and the effect of residual connections on it.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.RO
ROVE: Unlocking Human Interventions for Humanoid Manipulation via Reinforcement Learning
Wei Xiao, Weiliang Tang, Yuying Ge et al. (+4 more)
Human interventions provide crucial corrective signals for post-training
Vision-
Language-Action (VLA) models. However, enabling seamless
humanoid interventions is a formidable systems challenge due to complex whole-body kinematics and dexterous-hand control. Consequently, the collected intervention trajectories are often suboptimal, and methods that rely on human interventions as expert supervision can absorb hesitant, inefficient, or even erroneous behaviors. To address both the system and algorithmic challenges, we propose ROVE, a reinforcement learning framework for
humanoid VLA post-training with imperfect human interventions. First, ROVE introduces a human-in-the-loop pipeline capable of collecting deployment and intervention data for
humanoid manipulation. Second, it utilizes Optimistic Value Estimation (OVE) to prioritize high-value behaviors from mixed-quality trajectories. To further robustify value estimation, we incorporate cross-embodiment human experience
videos to provide rich supervision for long-tailed failure and recovery modes. The resulting critic yields informative advantage signals, steering the VLA actor to focus on high-value behaviors rather than indiscriminately imitating all actions. On challenging real-world contact-rich and fine-grained
humanoid manipulation tasks, ROVE outperforms experience-learning baselines and consistently improves across multiple rollout-intervention iterations.
Read the full abstract ▾
All authors Wei Xiao, Weiliang Tang, Yuying Ge, Hui Zhou, Yao Mu, Li Zhang, Yixiao Ge
Machine Learning
From Tokens to Policy: Causal and Interpretable Heterogeneous Treatment Effects Identification
Riccardo Cadei, Frank Otchere, Nyasha Tirivayi et al. (+3 more)
Heterogeneous Treatment Effect (HTE) identification is essential to explain the impact of an intervention and optimize our policies accordingly. Existing approaches trade expressivity for interpretability, but, if some active heterogeneity drivers are unmeasured, methods at both ends of this spectrum allow for spurious HTE characterization with no causal reading. In this work, we focus on controlled experiments and argue that an oracle HTE causal characterization via the latent interactors is now within reach, thanks to (i) more extensive pre-treatment measurements, i.e., multi-modal and multi-view, and (ii) scalable representations with minimal human supervision. We then re-frame HTE identification as a Markov-blanket discovery problem on a sufficient and aligned pre-treatment representation, and introduce Neural EXposure Interaction Search (NEXIS), an iterative procedure with a provable selection guarantee. We deploy NEXIS on two anti-poverty programs in Africa, augmenting each with satellite imagery capturing previously unmeasured environmental effect modifiers, leading to new candidate prescriptive guidelines to optimize the programs' next iterations.
Read the full abstract ▾
All authors Riccardo Cadei, Frank Otchere, Nyasha Tirivayi, Gustavo Angeles Tagliaferro, Falco J. Bargagli-Stoffi, Francesco Locatello
Machine Learning · Cross-listed from cs.SD
TuneJury: An Open Metric for Improving Music Generation Preference Alignment
Yonghyun Kim, Junwon Lee, Haiwen Xia et al. (+5 more)
We introduce TuneJury, an open, instance-level pairwise reward model for text-to-music that predicts a music preference score from a text
prompt and an audio clip. The released checkpoint is trained on publicly available human-preference labels covering arena-style (A vs. B) votes, metric-
alignment preference pairs, crowdsourced pairwise comparisons, and expert aesthetic ratings. The predicted score margin between two clips is well calibrated on our held-out test split, supporting data filtering via a simple score threshold. TuneJury generalizes to both held-out test pairs and out-of-distribution
benchmarks, remaining competitive with prior baselines on the latter. For generators released after training, we introduce anchor calibration, a post-hoc, per-system Bradley-Terry calibration that recovers agreement at substantially better data efficiency than from-scratch retraining. The same frozen reward drives consistent reward-axis gains across three downstream applications: inference-time best-of-N selection, DITTO-style latent optimization, and expert-iteration post-training. TuneJury is available at https://github.com/yonghyunk1m/TuneJury.
Read the full abstract ▾
All authors Yonghyun Kim, Junwon Lee, Haiwen Xia, Yinghao Ma, Junghyun Koo, Koichi Saito, Yuki Mitsufuji, Chris Donahue
Machine Learning · Cross-listed from cs.CC
The Complexity of Min-Max Optimization for Quadratic Polynomials
Martino Bernasconi, Matteo Castiglioni, Andrea Celli et al. (+1 more)
We prove that computing approximate stationary points of min-max optimization over the hypercube is PPAD-hard for quadratic
polynomials. This holds even when the
polynomials are multilinear, each variable appears in at most three monomials, and the approximation factor is inverse
polynomial. As a direct consequence, we obtain the first PPAD-hardness results for two-team zero-sum polymatrix games.
Read the full abstract ▾
All authors Martino Bernasconi, Matteo Castiglioni, Andrea Celli, Alexandros Hollender
Machine Learning · Cross-listed from cs.SE
Selection Without Signal, Recovery Through Expression: A Measurement Study of Post-Hoc Falsification Operators for Frozen Small Code Models
Mehmet Iscan
Frozen small code models (<=1.5B parameters, run locally without fine-tuning) suit offline and
privacy-constrained use, but often emit plausible-but-wrong programs. A natural remedy is a post-hoc operator that selects, verifies, repairs, or re-processes the model's samples without retraining; in principled form it is Popperian: attack each candidate with a severe test, keep what survives. We measure whether such operators help. Under one deterministic execution oracle and a leakage-free, matched-compute protocol, 26 semantic post-hoc operators (selection, verification, repair, elimination, portfolios, sound vetoes, generation conditioning) are evaluated against Best-of-N (BoN); on the
cells and
benchmarks tested, none improves held-out accuracy over BoN. The negative is mechanistic: a coverage wall (systematic hard-task failures deeper sampling does not rescue), a capability scissors (a competent generator leaves almost no discriminable error among visible-test passers), and a near-empty consensus trap (the visible-pass-but-hidden-wrong majority a leakage-free selector needs rarely co-occurs with a correct alternative). A distribution-free do-no-harm bound cannot certify a harm rate <=alpha at zero observed harm unless n>=45. Two operators help on a different axis, outside the semantic output space. An expression-layer recovery (M1), the only accuracy gain here, recovers correct programs the standard extractor discards (robust extraction and public-test signature
alignment); it does no harm (b10=0), is leakage-free, and lifts DeepSeek-Coder-1.3B by +12 tasks on HumanEval+ (p=2.4e-4). An adaptive consensus early-stop (ACE) is a calibrated compute-saving control (~19% saving, zero harm). M1 and the selection negative replicate on HumanEval+ and MBPP+ across three model
cells. The lesson: fix the harness and measure coverage before blaming semantic post-hoc
reasoning.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CV
ActiveSAM: Fast and Accurate Open-Vocabulary Semantic Segmentation with Frozen SAM 3
Tran Dinh Tien and Zhiqiang Shen
Segment Anything Model 3 (SAM 3) provides a strong frozen backbone for concept-prompted
segmentation, but applying it directly to open-vocabulary semantic
segmentation (OVSS) is inefficient: full-resolution decoding is typically run over the entire
dataset vocabulary, whereas each image contains only a small active subset of classes. We introduce ActiveSAM, a training-free inference framework that turns SAM 3 into an active-vocabulary segmenter. ActiveSAM first canonicalizes and expands class
prompts, then uses evidence-proportional grounding to estimate an image-conditioned active set from a low-resolution presence preview. Only retained
prompts receive full-resolution mask prediction, using bucketed
prompt multiplexing with the frozen SAM 3 decoder. The preview stage uses only class-presence evidence and skips unnecessary
segmentation-head computation. To resolve overlapping concept responses, exclusive concept decoding compares each pixel's joint score vector with class signatures estimated once per vocabulary from unlabeled images. ActiveSAM requires no weight updates, no oracle class-presence labels and no per-
dataset hyperparameter tuning. Across eight OVSS
benchmarks, ActiveSAM improves the speed-accuracy tradeoff of training-free open-vocabulary semantic
segmentation, outperforming the current state-of-the-art SegEarth-OV3 by +2.1 mIoU on average while running much faster, with 7.3-12.2x speedups on large-vocabulary
datasets. ActiveSAM also achieves the highest accuracy under image corruptions that simulate real-world distribution shift, making it well-suited for deployment in noisy-input domains such as autonomous driving and
embodied AI. Code is available at https://github.com/VILA-Lab/ActiveSAM
Read the full abstract ▾
Machine Learning · Cross-listed from cs.AI
When in Doubt, Plan It Out: Committed Small Language Model Deliberation for Reactive Reinforcement Learning
Nathan Gavenski, Juarez Monteiro, Francisco Galuppo et al. (+2 more)
Reinforcement Learning (RL) policies often degrade in unfamiliar environments because they lack explicit deliberation. We propose Plan, Align, Commit, Think (PACT), a hybrid architecture that combines a fast, reactive RL policy with a slow, deliberative Small
Language Model (SLM) planner. PACT invokes the SLM asynchronously to generate and validate candidate action plans. Once a plan is verified through simulation as safe, feasible, and complete, it is executed directly, bypassing the RL policy without retraining or modifying it. Evaluated on three FrozenLake configurations of increasing difficulty, PACT outperforms all baselines while relying on a 2B-parameter SLM backbone, suggesting that deliberative planning and reactive execution are more powerful in concert than either is alone in these settings.
Read the full abstract ▾
All authors Nathan Gavenski, Juarez Monteiro, Francisco Galuppo, Adriano Veloso, Odinaldo Rodrigues