cs.LG · 2026-06-16 · No. 25
Machine Learning, 2026-06-16.
65 new papers in cs.LG. Titles, authors,
abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →
01 — The papers
65 entries-
01
Exact Posterior Score Estimation for Solving Linear Inverse Problems
Abbas Mammadov, Ozgur Kara, Kaan Oktay, Iskander Azangulov, Adil Kaan Akan, Hyungjin Chung, James Matthew Rehg, Yee Whye Teh
cs.LG · cs.CV · stat.ML
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...
-
02
Your Privacy My Cloak: Backdoor Attacks on Differentially Private Federated Learning
Xiaolin Li, Ning Wang, Ninghui Li, Wenhai Sun
cs.LG · cs.CR
Prior research suggests that differential privacy (DP) inherently enhances 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...
-
03
HAMON: Passive Optical Sequence Mixing for Long-Horizon Forecasting
Alper Yıldırım
cs.LG · cs.AI · cs.AR
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...
-
04
ExpRL: Exploratory RL for LLM Mid-Training
Violet Xiang, Amrith Setlur, Chase Blagden, Nick Haber, Aviral Kumar
cs.LG
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 \emph{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...
-
05
Filtered Conformal Ellipsoids for Graph-Native Time Series
Yannick Limmer
cs.LG · math.ST · stat.ML
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...
-
06
From Tokens to Policy: Causal and Interpretable Heterogeneous Treatment Effects Identification
Riccardo Cadei, Frank Otchere, Nyasha Tirivayi, Gustavo Angeles Tagliaferro, Falco J. Bargagli-Stoffi, Francesco Locatello
cs.LG
Heterogeneous Treatment Effect (HTE) identification is crucial 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...
-
07
Analytic Torsion and Spectral Gap Capture Persistent-Laplacian Performance
Jernej Grlj, Aaron D. Lauda
cs.LG · math.AT
While persistent Laplacians (PL) offer a richer geometric representation of data than persistent homology, utilizing their full eigenspectrum for learning tasks is often hampered by high dimensionality and the ``varying length'' problem across different filtration scales. We propose a compact spectral representation that distills the persistent Laplacian into three mathematically grounded invariants: Betti numbers, the spectral gap, and...
-
08
Scalable Pairwise Kernel Learning with Stochastic Vec Trick
Napsu Karmitsa, Tapio Pahikkala, Antti Airola
cs.LG
Pairwise learning is a specialized form of supervised learning that focuses on predicting outcomes for pairs of objects. In this work, we introduce SPaiK, a new scalable kernel learning method tailored for pairwise settings. Our approach preserves the expressive power of kernel methods while substantially reducing computational and memory requirements. The key innovation is the stochastic generalized vec trick (sGVT), a stochastic extension...
-
09
Beyond the Smile: A Hybrid Convolutional VAE for Crypto Volatility Surfaces
Sadanand Singh, Allam Reddy, Manan Chopra
cs.LG · q-fin.CP
We present a convolutional variational autoencoder for cryptocurrency implied-volatility surfaces, together with a deployable predictor that combines it with a quadratic smile re-fit through a deterministic per-tenor routing rule. Trained on 6,034 fully-filled hourly Binance Options surfaces of BTC and ETH spanning May-October 2023 and parameterised on a common $6 \times 7$ tenor-delta grid, the model attains a hidden-cell surface-completion...
-
10
Phantoms and Disclosures: a Causal Framework for Auditing Synthetic Data
Kareem Amin, Rudrajit Das, Alessandro Epasto, Adel Javanmard, Dennis Kraft, Mónica Ribero, Sergei Vassilvitskii
cs.LG · cs.AI · stat.AP · stat.ME · stat.ML
The rapid adoption of generative AI and Large Language Models (LLMs) has spurred interest in synthetic data as a privacy-preserving alternative to sensitive real-world datasets. However, generating high-utility synthetic data often carries the risk of memorizing and regurgitating private information from the training corpus. In this work, we present a customizable empirical auditing framework designed to detect and explain such data...
-
11
Scalable Circuit Learning for Interpreting Large Language Models
Naiyu Yin, Dennis Wei, Tian Gao, Amit Dhurandhar, Karthikeyan Natesan Ramamurthy, Yue Yu
cs.LG · cs.AI
A prominent research direction in mechanistic interpretability is learning sparse circuits over LLM components to reveal how they jointly produce model behavior. However, raw neurons are polysemantic, making learned circuits hard to interpret. Sparse autoencoder (SAE) features alleviate this, but their high dimensionality makes existing intervention-based circuit learning methods computationally prohibitive. We propose CircuitLasso, a...
-
12
A Unified Causal-Origin Taxonomy of Distributional Shifts in Reinforcement Learning
Ardianto Wibowo, Paulo E Santos, Amer Baghdadi, Matthew Stephenson, Karl Sammut, Jean-Philippe Diguet
cs.LG · cs.AI
Reinforcement learning (RL) systems often degrade when operating conditions differ from those previously encountered, reflecting distributional shifts in the underlying data-generating process. Such shifts may occur between training and evaluation, as in In-Distribution (ID) and Out-of-Distribution (OOD) generalization, or within non-stationary settings where environment dynamics evolve over time. However, the formal relationship between...
-
13
Demystifying Variance in Circuit Discovery of LLMs
Frank Zhengqing Wu, Francesco Tonin, Volkan Cevher
cs.LG · cs.AI
Circuit discovery is a key technique in mechanistic interpretability to pinpoint the model components that are crucial for performing a given task. Although the current state-of-the-art method (EAP-IG) performs well on the metric of (un)faithfulness, it suffers from substantial variability. This includes resampling variance, where the circuit changes when we probe with a new batch of data from the same distribution; rephrasing variance, where...
-
14
Factorized Neural Operators Decompose Dynamic and Persistent Responses
Hao Tang, Yuechen Duan, Jiongyu Zhu, Zimeng Feng, Hao Li, Chao Li
cs.LG
Physical systems often exhibit heterogeneous mechanisms, where rapidly evolving dynamics coexist with persistent structures. Capturing such multiscale physical behavior remains challenging for existing neural operators, which typically rely on single dominant inductive bias and therefore couple distinct physical responses into a shared representation. We introduce the Unified Green's Function Framework across domains and propose the...
-
15
Fantastic Pretraining Optimizers and Where to Find Them II: Hyperball Optimization
Kaiyue Wen, Xingyu Dang, Kaifeng Lyu, Tengyu Ma, Percy Liang
cs.LG
Matrix based optimizers such as Muon can substantially speed up language model pretraining, but their gains over AdamW are observed to shrink as model size and data scale grow when using standard constant decoupled weight decay. We propose Hyperball, a simple optimizer wrapper that addresses this issue. Given a base optimizer such as Adam or Muon, Hyperball sets the Frobenius norms of weight matrices and their corresponding optimizer updates...
-
16
Beyond Weights and Gradients: A Taxonomy of Federated Learning Messages
Alvaro Javier Vargas Guerrero, Xinguang Wang, Quang Manh Doan, Guy Nagels
cs.LG · cs.AI
Federated Learning is rapidly evolving beyond the exchange of traditional model weights and gradients, yet existing definitions fail to capture the full scope of modern payloads like synthetic data and federated analytics. This paper addresses the gap by proposing a formal mathematical definition of a federated message that accounts for both utility and privacy. We introduce a taxonomy that organizes these exchanges into three categories:...
-
17
Upper Bounds on the Generalization Error of Deep Learning Models via Local Robustness and Stability
Abdul-Rauf Nuhu, Parham M. Kebria, Vahid Hemmati, Mahmoud N. Mahmoud, Edward Tunstel, Abdollah Homaifar
cs.LG · cs.AI
Generalization is a critical property of data-driven models, particularly deep learning models deployed in safety-critical applications. Robustness-based generalization bounds have gained attention as a principled way to link robustness properties to generalization performance, often in a data-dependent manner. However, most existing bounds suffer from vacuousness in practical settings, yielding loose upper bounds that greatly exceed the...
-
18
Integrated Marketing Attribution: A Bayesian Framework for Privacy-Safe Granular Measurement Anchored in MMM
Meghana R. Bhat, Ankit Umare, Utsav Aggarwal, Richard Vecsler, Arunkumar Mani, Karthik Nair, Chandhu Nair
cs.LG
Retail marketing measurement increasingly requires granular campaign-level insights without relying on user-level tracking. However, the two dominant approaches, Marketing Mix Modeling (MMM) and Multi-Touch Attribution (MTA), often produce fragmented insights. MMM is privacy-safe and robust for channel-level planning but is too coarse for campaign optimization, while MTA provides granular attribution but has become less reliable under...
-
19
HawkesNest: A Multi-Axis Synthetic Benchmark for Spatiotemporal Pattern Complexity
Yahya Aalaila, Sumantrak Mukherjee, Gerrit Großmann, Sebastian Vollmer
cs.LG
Evaluation of spatiotemporal point process (STPP) models relies heavily on opaque real-world datasets, where latent generative structure is unknown and model failures are difficult to attribute. We introduce HawkesNest, a generator-aligned benchmark for controlled spatiotemporal pattern complexity built on a multivariate Hawkes backbone. HawkesNest defines four complexity axes: space--time entanglement, background heterogeneity, cross-type...
-
20
Deep Q-Learning on Hölder Spaces
Qian Qi
cs.LG · cs.AI
We study the operator-theoretic core of Q-learning in continuous-time stochastic control with continuous states and actions. In value-based reinforcement learning, each Q-learning or DQN update is built from a Bellman optimality target; our analysis isolates this target in a diffusion setting and studies its regularity and approximation complexity. Under uniform ellipticity and Hölder-regular coefficients, we show that a Bellman update maps...
-
21
Decision-Weighted Flow Matching for Contextual Stochastic Optimization
Jize Xie, Haomiao Wu, Qiang Chen, Xiu Su, Yi Chen
cs.LG · cs.AI
Conditional generative models are increasingly used as scenario generators for stochastic optimization, but standard training objectives emphasize uniform distributional fit rather than the downstream decisions induced by generated scenarios. This creates an objective mismatch: errors in statistically common regions may have little effect on decision regret, whereas errors in decision-sensitive regions can substantially change the optimal...
-
22
We Need Explanation Cards to Connect Explanation Algorithms to the Real World
Eric Günther, Balázs Szabados, Kristof Meding, Gunnar König, Sebastian Bordt, Ulrike von Luxburg
cs.LG
Algorithmic explanations are intended to help stakeholders understand opaque algorithmic decisions, but in practice, they often fall short. First, the meaning of algorithmic explanations is often not what one might intuitively expect, so expert knowledge is required to interpret them correctly. Second, recent work has shown that popular explanation algorithms are uninformative about the behavior of complex decision functions. Together, these...
-
23
GD$^2$PO: Mitigating Multi-Reward Conflicts via Group-Dynamic reward-Decoupled Policy Optimization
Haotian Liu, Yihao Liu, Jingwei Ni, Siyuan Huang, Xinpeng Liu, Pengyu Cheng, Jiajun Song, Ruijin Ding, Junfeng Li,...
cs.LG
As LLMs advance, post-training reinforcement learning (RL) increasingly relies on multi-dimensional rewards to cultivate comprehensive capabilities. This shift demands new algorithms capable of optimizing diverse and potentially competing objectives simultaneously. To address this, existing methods such as Group reward-Decoupled Policy Optimization (GDPO) decompose the overall score into independent reward groups, then compute the RL loss...
-
24
Taming Curvature: Architecture Warm-Up for Stable Transformer Training
Sameera Ramasinghe, Ajanthan Thalaiyasingam, Hadi Mohaghegh Dolatabadi, Chamin Hewa Koneputugodage, Gil Avraham,...
cs.LG
Training billion-parameter Transformers is often brittle, with transient loss spikes and divergence that waste compute. Even though the recently developed Edge of Stability (EoS) theory provides a powerful tool to understand and control the stability of optimization methods via the (preconditioned) curvature, these curvature-controlling methods are not popular in large-scale Transformer training due to the complexity of curvature estimation....
-
25
A Validated LBM Dataset and Pipeline for Surrogate Modeling of Turbulent 3D Obstructed Channel Flows
Lukas Schröder, Shubham Kavane, Harald Köstler
cs.LG · physics.flu-dyn
Evaluating neural operators for 3D turbulent flow requires validated datasets with physical benchmarks. We present a reproducible pipeline generating training data for 3D channel flows around generated geometries at Re=1,000-10,000. Our lattice Boltzmann solver with cumulant collision operators is rigorously verified against experimental measurements (Strouhal number, drag coefficients, turbulent fluctuations) with comprehensive grid...
-
26
Maximum Entropy Inverse Reinforcement Learning for Mean-Field Games with Average Reward
Şevket Kaan Alkır, Naci Saldı, Berkay Anahtarcı, Can Deha Karıksız
cs.LG
We study inverse reinforcement learning for discrete-time, infinite-horizon mean-field games (MFGs) under an average-reward criterion. Expert demonstrations are assumed to arise from a stationary mean-field equilibrium under an unknown reward, and the goal is to recover a policy explaining the observed behaviour via the maximum causal entropy principle. We formulate the inverse problem by enforcing consistency with the expert mean-field term...
-
27
MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents
Lawrence Keunho Jang, Andrew Keunwoo Jang, Jing Yu Koh, Ruslan Salakhutdinov
cs.LG · cs.CL
Current benchmarks for computer-use agents evaluate models in impersonal environments. This leaves a gap between evaluation and deployment where personal assistants are expected to work across a user's whole digital life, including their context, historical data, and logged-in accounts. This gap is widest on web tasks, where live web evaluations cannot exercise sites that require logging in or personal information, the kind of site a real...
-
28
Learning Policy from a Single Trajectory in Average-Reward Markov Decision Process
Jongmin Lee, Ernest K. Ryu, Vaneet Aggarwal
cs.LG · math.OC
While there is an extensive body of work characterizing the sample complexity of discounted cumulative-reward MDPs, finite sample analyses for average-reward MDPs have been limited, and most existing works rely on restrictive assumptions such as ergodicity or access to a generative model. In this work, we establish the first finite sample complexity guarantees from a single trajectory for weakly communicating average-reward MDPs. To this end,...
-
29
Adaptive inference and function vectors in deep transformers
Ravin Raj, Gautam Reddy
cs.LG · cs.AI · physics.app-ph · q-bio.NC
Transformers are widely used as a general-purpose substrate for learning complex correlations between a large collection of coupled variables, but their internal mechanisms have remained mysterious. We introduce a theory of a deep transformer as a mean-field interacting system that implements distributed inference, subject to constraints on communication, locality and depth. We show that such a system can exploit internal state...
-
30
Multimodal Evaluator Preference Collapse: Cross-Modal Contagion in Self-Evolving Agents
Zewen Liu
cs.LG · cs.CL
When AI agents use language models to evaluate their own outputs in a feedback loop, systematic biases emerge. We show that Evaluator Preference Collapse (EPC) is dramatically amplified in multimodal settings. Using GPT-4o to evaluate DeepSeek-chat across text and visual tasks, we find that a single strategy (step_by_step) absorbs 48.4% of all weight -- 3.2x the collapse observed in text-only self-evaluation -- while three visual-domain...
-
31
Beyond Defensive Reporting: Machine Learning for Active Anti-Money Laundering Control in Insurance
Dara Goldar, Geir Kjetil Ferkingstad Sandve, Martin Jullum
cs.LG
Money laundering through insurance claims poses a threat to insurers both through fraudulent payouts and reputational and regulatory risk. Despite this, little research has examined how such laundering can be prevented. This paper examines whether machine learning can help insurers flag suspicious claims before payout, shifting the focus from passive reporting to active prevention. Using production data from a major Norwegian insurer, we...
-
32
Near-Optimal Stochastic Linear Bandits with Delay
Ofir Schlisselberg, Mengxiao Zhang, Yishay Mansour
cs.LG
We study stochastic linear bandits with delayed feedback under several delay models and establish near-optimal regret guarantees. Our results identify when delayed linear bandits exhibit the same qualitative behavior as multi-armed bandits (MAB), and when the linear structure creates fundamentally new challenges. Specifically, (1) for \emph{loss-independent delays}, where the delay does not depend on the realized loss (but potentially depends...
-
33
Distribution Alignment for One-Shot Federated Learning via Optimal Transport
Daniele Berardini, Vito Paolo Pastore, Vittorio Murino
cs.LG
One-Shot Federated Learning (OSFL) addresses extreme communication regimes in which clients interact with the server only once, amplifying the impact of heterogeneous client data distributions. In particular, the interaction of domain shift and label shift across clients induces misaligned feature representations that cannot be corrected through iterative optimization. Existing OSFL methods rely on distillation, server-side generation or...
-
34
SPICE: Synergy and Partial Information Based Curriculum Evolution
Ankush Pratap Singh, Houwei Cao, Yong Liu
cs.LG
Multimodal learning exploits complementary information across heterogeneous modalities. The informativeness of each modality can vary widely across samples and training stages. Existing multimodal curriculum learning strategies often assume that the relative complexity of samples remains unchanged throughout training and therefore cannot adapt to model evolution. We propose SPICE (Synergy and Partial Information based Curriculum Evolution), a...
-
35
Entropy-Gated Latent Recursion
Soham Bhattacharjee, Dushyant Singh Chauhan, Salem Lahlou, Martin Takac, Nils Lukas
cs.LG · cs.AI
Inference-time scaling has become the dominant lever for improving language-model reasoning, but existing methods derive rollout diversity from a single source: stochastic token-level sampling. We argue that this single-axis sampling space is fundamentally limiting, and identify a second, fully deterministic and complementary axis: the layer span $L$ at which a frozen model's top decoder layers are recursively re-applied at high-uncertainty...
-
36
TCHG: Tri-Trust Conditioned Heterogeneous Graph Learning for Reliable Dynamic Trust Prediction
Bohao Liao, Boyu Deng, Qipeng Song, Jieling Wang, Jingchao Wang
cs.LG
Trust prediction infers latent user-user trust relations and provides important support for social recommendation, fake-review and manipulation detection, and risk identification. Graph neural networks have become a prominent approach to trust prediction because of their ability to learn network structures and complex trust dependencies. However, existing methods often rely on a unified representation of trust signals and do not disentangle...
-
37
PhysGuard: Fisher-Guided Gradient Projection for Sim-to-Real Neural PDE Surrogates
Changjian Zhou, Junfeng Fang, Negin Yousefpour, Peng Wu, Bin Yan, Guillermo A Narsilio
cs.LG · math.NA · physics.comp-ph
Neural operator models trained on simulation data often lose accuracy when applied to experimental measurements due to the sim-to-real gap. Standard fine-tuning with limited real data can reduce this gap, but it may also damage the core physics-relevant representations learned during pretraining. Although knowledge-preserving adaptation has been widely investigated in vision or language tasks, it remains unclear whether these methods are...
-
38
Infant Spontaneous Movement Noise Improves Exploration in Deep RL
Francisco M. López, Markus R. Ernst, Francisco Cruz, Matej Hoffmann, and Jochen Triesch
cs.LG · cs.AI · q-bio.NC
Exploration in deep reinforcement learning (RL) is commonly implemented as temporally uncorrelated white noise. However, recent works show that temporally correlated colored noise can improve exploration efficiency by producing smooth trajectories with better coverage of the state space. We inquire whether action noise inspired by infant spontaneous movements can also improve exploration in deep RL. We find that the power spectral densities...
-
39
Multi-Modal Spatio-Temporal Graph Neural Network with Mixture of Experts for Soil Organic Carbon Prediction
Daniele Mos, Felipe Drummond, Anton Bossenbroek, Soufiane el Khinifri
cs.LG · cs.CV
Top-soil organic carbon (SOC) prediction is fundamental to agricultural sustainability, land use policy and fertilization planning. Existing approaches face two limitations: they pair hand-crafted covariates with classical ML or single-modal deep models that miss rich spectral and temporal information, and grid-based architectures ignore the irregular spatial structure of field measurements. We introduce SpTGNN, a multi-modal spatio-temporal...
-
40
On the Entropy Formula for Real, Complex, and Quaternionic Deep Linear Networks
Luis Contreras, Marco Nahas, Tejas Kotwal
cs.LG · math-ph · math.DG
We extend the entropy formula of Menon and Yu for the real Deep Linear Network (DLN) to its complex and quaternionic analogues, obtaining a unified formula for DLNs over $\mathbb{R}$, $\mathbb{C}$, and $\mathbb{H}$.
-
41
RepNet: Tackling spectral bias in deep neural networks via parameter reparameterization
Yong Wang, Tao Zhou, Xuhui Meng
cs.LG · math-ph
Deep neural networks (DNNs) have achieved remarkable success in scientific computing, yet they often suffer from spectral bias in capturing oscillatory and multiscale behaviors. In this study, we investigate this limitation by examining the failure of shallow ReLU neural networks in fitting high-frequency functions. This observation identifies two important factors in resolving rapid oscillations: the initial slope scale and the distribution...
-
42
MIRAGE: Auditing Anti-Muslim Bias in Frontier LLMs Across Reasoning, Agentic, and Time-Coupled Conditions
Noor Islam S. Mohammad, Tamim Sheikh
cs.LG
Five years after the discovery of persistent anti-Muslim bias in large language models, most evaluations remain confined to single-turn prompt completion, a setting that no longer reflects how frontier LLMs are deployed. We introduce \textbf{MIRAGE} (Muslim-Identity Reasoning and Agentic Generation Evaluation), a benchmark of 1{,}200 prompts spanning three deployment-realistic conditions: direct completion, chain-of-thought reasoning, and...
-
43
Assessing Reliability of Symbol Detection in Concept Bottleneck Models
Javier Fumanal-Idocin, Javier Andreu-Perez
cs.LG · cs.CV · cs.SC
Concept Bottleneck Models (CBMs) are a relevant tool for explainable Artificial Intelligence because they make their predictions through human-interpretable symbols. However, high task accuracy does not guarantee that these symbols are detected faithfully: jointly trained CBMs may encode task-specific shortcuts in the bottleneck, making their explanations unreliable. In this paper, we study concept-detection reliability by swapping...
-
44
Neural Bayesian Anomaly Mitigation: A Robust Loss that Doubles as an Unsupervised Contamination Classifier
S. A. K. Leeney, W. J. Handley, H. T. J. Bevins, E. de Lera Acedo
cs.LG · astro-ph.CO · stat.ML
Engineered robust losses such as Huber, Student-$t$, and generalised cross-entropy make supervised models tolerant of contamination but cannot answer which observations are corrupted. We introduce Neural Bayesian Anomaly Mitigation (NBAM), a general-purpose drop-in loss derived from a Bayesian latent-switch mixture model: the marginal likelihood defines a robust supervised loss, and the associated posterior defines an unsupervised...
-
45
How Post-Training Shapes Biological Reasoning Models
Lukas Fesser, Hanlin Zhang, Michelle M. Li, Eric Wang, Bryan Perozzi, Shekoofeh Azizi, Sham M. Kakade, Marinka Zitnik
cs.LG · q-bio.QM
Scientific reasoning models for biology combine language models with foundation models trained on multimodal biological data, including DNA, RNA, and proteins. These models are built through post-training, yet how each stage shapes reasoning and generalization remains poorly understood. We study when post-training improves performance and when it induces over-specialization. Across genomics, transcriptomics, and proteins, we train and...
-
46
Direction-Conditioned Policies via Compositional Subgoal Scoring for Online Goal-Conditioned Reinforcement Learning
Swaminathan S K, Damiya Gondha, Theyanesh Eswaramoorthy Rajahkrishnan, Aritra Hazra
cs.LG · cs.AI · cs.RO
Hamilton-Jacobi-Bellman theory implies that the optimal goal-conditioned action depends on the goal only through the gradient of the goal-reaching distance at the current state, yet standard online GCRL still conditions the actor on the raw goal -- a signal that is geometrically uninformative when the goal is far from the data distribution. We propose Direction-Conditioned Policies (DCP), a fully online method that decomposes goal-reaching...
-
47
Tail-Shape Estimation in LLM Evaluation Is Fragile: A Protocol for Diagnosing False Positives
Luca Zhou
cs.LG
Recent work motivates moving large language model (LLM) evaluation from mean-based to tail-aware metrics, including conditional value-at-risk and tail-index estimates of reward-model error. We ask whether the canonical extreme-value-theory tail-index parameter, which isolates how heavy a tail is from how large the tail mass is, adds discriminative information beyond the mean and a standard tail-magnitude statistic in LLM evaluation. We...
-
48
daVinci-kernel: Co-Evolving Skill Selection, Summarization, and Utilization via RL for GPU Kernel Optimization
Dayuan Fu, Mohan Jiang, Tongyu Wang, Dian Yang, Jiarui Hu, Liming Liu, Jinlong Hou, Pengfei Li
cs.LG · cs.AI · cs.CL
GPU kernel optimization represents a paradigm where functional correctness is assumed and execution efficiency is the objective. We present daVinci-kernel, a reinforcement learning framework that couples skill discovery with skill exploitation through a dynamically evolving skill library. daVinci-kernel jointly trains three agents sharing one LLM backbone: a Skill Selection Agent that retrieves relevant techniques via BM25 and LLM reranking,...
-
49
BRICKS-WM: Building Reusability via Interface Composition Kinetics for Structured World Models
Shaowei Zhang, Jiahan Cao, Xunlan Zhou, Shenghua Wan, De-Chuan Zhan
cs.LG
Model-based Reinforcement Learning (MBRL) has achieved remarkable success in continuous control by leveraging latent world models. However, prevailing approaches typically rely on monolithic latent dynamics, entangling environment dynamics into a coupled process. This coupling severely limits reusability: altering the agent necessitates retraining the entire world from scratch, even if the environment remains constant. To address this, we...
-
50
Learning aligned EEG representations with subject-specific encoders
Bruna J. Lopes, Gabriel Schwartz, Sylvain Chevallier, Raphael Y. de Camargo, Bruno Aristimunha
cs.LG · cs.AI
Cross-subject EEG decoding promises more training data, but it also exposes neural networks to strong inter-subject distribution shifts. We study whether task supervision and architecture alone can learn subject-aligned representations. We replace a shared EEG encoder with subject-specific encoders followed by a common classifier, and compare this hybrid model with standard EEGNet, AttentionBaseNet, and CTNet baselines with Euclidean...
-
51
Privacy from Symmetry: Orthogonally Equivariant Transformers for LLM Inference
Alexander Yukhimchuk, Andrey Shulga, Mladen Kolar, Martin Takáč
cs.LG
Running large language models locally is often impractical, pushing inference on sensitive text to third-party providers. Split inference partially mitigates this by keeping tokens on the client and sending only hidden representations, but these representations can still be recovered via nearest-neighbor search against the public embedding table. We propose an orthogonal obfuscation procedure in which the client multiplies embeddings by a...
-
52
SPRI: SVD-Partitioned Residual Initialization for Data-Constrained MoE Upcycling
Weiqiao Shan, Ruixiang Mao, Yuang Li, Yuhao Zhang, Yingfeng Luo, Tong Zheng, Chen Xu, Yucheng Qiao, Chunxiang Jin,...
cs.LG · cs.AI
Mixture-of-Experts (MoE) models enable efficient scaling, but training them from scratch remains prohibitively expensive. MoE upcycling mitigates this cost by converting pretrained dense models into sparse MoE models. However, existing upcycling methods typically rely on large-scale continued training and often perform poorly under data-constrained supervised adaptation, due to either homogeneous experts or overly disruptive perturbations to...
-
53
SDS-LoRA: Overcoming Anisotropic Gradient Scaling in Low-Rank Adaptation
Junghun Oh, Sungyong Baik, Kyoung Mu Lee
cs.LG · cs.AI
Low-Rank Adaptation (LoRA) enables efficient adaptation of large pre-trained models to downstream tasks by parameterizing weight updates with low-rank matrices. In this paper, we investigate the limitations of the LoRA parameterization from a geometric perspective. Specifically, we show that when a full fine-tuning gradient is backpropagated to the low-rank matrices, it undergoes anisotropic scaling driven by their singular values. We argue...
-
54
Autonomous End-to-End SOH Prediction Services for Battery Systems via Temporal-Contrastive Representation Learning
Junting Wen, Dan Li, Qihao Quan, Xiwen Wang, Hang Yang, Zhaohong Meng, Zigui Jiang, Changlin Yang, Tianle Liu, Diego...
cs.LG · cs.AI
Accurate state of health (SOH) estimation is a critical diagnostic service for lithium-ion battery management. However, reliance on labor-intensive manual feature engineering and opaque black-box models hinders scalable industrial deployment. To address this, we introduce TC-SOH: a modular, plug-and-play service architecture for autonomous, end-to-end SOH prediction. TC-SOH employs a temporal-contrastive mechanism and a cross-window...
-
55
Taylor-Calibrate: Principled Initialization for Hybrid Linear Attention Distillation
Zhongzhu Zhou, Qingyang Wu, Junxiong Wang, Mayank Mishra, Shuaiwen Leon Song, Ben Athiwaratkun, Chenfeng Xu
cs.LG · cs.CL
Hybrid linear attention models offer an appealing path to faster long-context inference: they reduce the quadratic cost and KV-cache burden of full softmax attention while retaining much of the quality of Transformer models. A practical way to obtain such models is to convert a pretrained Transformer instead of pretraining a new architecture from scratch, but this conversion is still brittle. Simply copying the teacher attention projections...
-
56
Not all Jensen-Shannon Divergence Estimators are Equal
Alba Garrido, Alejandro Almodóvar, Mar Elizo, Patricia A. Apellániz, Santiago Zazo, Juan Parras
cs.LG
The Jensen-Shannon divergence is widely reported as a scalar measure of fidelity for synthetic tabular data. Yet, in practice, it is estimated from finite samples using protocols that are often underspecified. This creates a measurement problem. Although the population divergence is well defined, the empirical value depends on the estimator family, sampling protocol, calibration, dimensionality, and class balance. We show that different...
-
57
MUNI: Multimodal Unified Latent Diffusion for Coherent Any-to-Any Generation
Kyeongmin Yeo, Yunhong Min, Minhyuk Sung
cs.LG
We introduce MUNI, an end-to-end multimodal latent diffusion framework for any-to-any generation that unifies subset-conditioned cross-modal generation and unconditional joint sampling through a shared stochastic latent. Existing multimodal generative models are largely LLM-based, which limits leveraging modality-specific generators and requires text-paired data for training. Recent diffusion- and flow-based any-to-any extensions take a...
-
58
Robust Neural Tucker Factorization with Bias Correction and Adaptive Initialization
Yuchao Su, Yixin Ran
cs.LG
High-dimensional incomplete (HDI) tensors are widely used in traffic and climate applications, but sparse observations make accurate completion difficult. The intrinsic non-linear dynamics and non-stationary variations across distinct multi-modal fields severely hinder the efficacy of conventional linear reconstruction frameworks. Neural Tucker factorization provides an effective framework for modeling high-order interactions among tensor...
-
59
Mixtures of Subspaces for Bandwidth Efficient Context Parallel Training
Sameera Ramasinghe, Ajanthan Thalaiyasingam, Hadi Mohaghegh Dolatabadi, Gil Avraham, Violetta Shevchenko, Yan Zuo,...
cs.LG
Pretraining language models with extended context windows enhances their ability to leverage rich information during generation. Existing methods split input sequences into chunks, broadcast them across multiple devices, and compute attention block by block which incurs significant communication overhead. While feasible in high-speed clusters, these methods are impractical for decentralized training over low-bandwidth connections. We propose...
-
60
Scalable and Interpretable Representation Alignment with Ordinal Similarity
Diogo Soares, Pankhil Gawade, Andrea Dittadi, Ewa Szczurek
cs.LG · stat.ML
Evaluating representation similarity is fundamental to representation learning. However, existing metrics suffer from significant limitations: they lack interpretability due to shifting baselines, lack robustness to outliers, and are computationally intractable for large datasets, forcing reliance on heuristic approximations. To address this, we develop an ordinal-similarity framework, instantiated by the Triplet (TSI) and Quadruplet (QSI)...
-
61
CacheMuon: Using Temporal Preconditioning To Approximate Polar Factor
Bishnu Dev, Sushil Bohara, Martin Takáč, Samuel Horváth
cs.LG
Muon is an optimizer that computes updates using the polar factor of the momentum matrix and has shown strong empirical performance across a range of training settings. A key component of Muon is the Newton-Schulz iteration used to compute this polar factor. Although this avoids the cost of an exact singular value decomposition, it remains expensive in practice because it is applied at every optimization step. At the same time, the momentum...
-
62
Simulation-Augmented Multi-Step Split Conformal Prediction for Aggregated Forecasts
Andro Sabashvili
cs.LG
We study uncertainty quantification for aggregated forecasting tasks such as annual totals and year-over-year growth rates. We propose SA-MSCP, a simulation-augmented multi-step split conformal method that generates future paths from cross-validated residuals using a block bootstrap and constructs prediction intervals from empirical quantiles. Experiments show that SA-MSCP improves empirical coverage over a simulated-path baseline for...
-
63
Communication-Efficient Verifiable Attention for LLM Inference
Ziqun Chen, Ming Wu, Michael Heinrich, Jason Zeng, Huiying Lan, Tianwei Zhang, Rui Tan
cs.LG · cs.AI
Computation integrity of remote large language model (LLM) serving can be questionable. For conventional deep neural networks (DNNs), the existing TEE-shielded DNN partitioning (TSDP) approach uses Trusted Execution Environment (TEE) to compute non-linear components and verify the integrity of linear components offloaded to an untrusted GPU. However, directly applying TSDP to Transformer-based LLMs incurs significant TEE computation and...
-
64
Filtered ANN as a Phase Transition: When Selectivity-Estimation Error Causes Plan Regret
Madhulatha Mandarapu, Sandeep Kunkunuru
cs.LG · cs.DB
A filtered approximate-nearest-neighbor (ANN) query returns the k nearest vectors among those satisfying an attribute predicate P of selectivity s. The best execution strategy -- pre-filter, post-filter, or in-filter -- changes with s, so a system must estimate s and choose. We model this as an argmax over a landscape with phases (regions where each strategy wins) separated by boundaries, and show that selectivity-estimation error produces...
-
65
Diffusion Offline Reinforcement Learning for Fair and Energy-Efficient UAV-Assisted Wireless Networks
Eslam Eldeeb, Hirley Alves
cs.LG
The integration of generative artificial intelligence with wireless communication and signal processing systems has opened new avenues for intelligent, data-driven decision-making in future 6G networks. This work proposes a diffusion soft actor-critic (Diffusion-SAC) approach that leverages offline reinforcement learning (RL) enhanced by denoising diffusion probabilistic models (DDPMs) to optimize trajectory and scheduling control in unmanned...
This edition is part of The Daily Abstract — cs.LG archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.
#D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.