cs.LG · 2026-10-21 · No. 150
Machine Learning, 2026-10-21.
58 new papers in cs.LG. Titles, authors,
abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →
01 — The papers
58 entries-
01
Planning under Noise with Neural Operators
Michael Chen, Hong Chul Nam, Xi Deng, Anima Anandkumar
cs.LG · cs.RO
Planning under uncertain control execution noise requires reliable feedback policies across diverse environments. We introduce Planning Under Noise (PUN), a neural operator that learns noise-aware value fields for planning in unseen geometries. Building on an established screened-Poisson formulation of stochastic optimal control, we address two challenges: specifying boundary conditions that support navigation across varying geometries and...
-
02
Efficient Edge-aware Attention Network for Graph Generation
Vincent Jung, Alba Carballo-Castro, Yiming Qin, Lonneke van der Plas, Pascal Frossard
cs.LG
Graph generative models commonly use graph-specific Transformer architectures to jointly update node and edge features. However, these architectures do not directly benefit from recent hardware-aware attention implementations, limiting their scalability. The adaptation is, however, non-trivial: edge and graph-level features must be injected into pairwise attention scores, graph masks must be respected, and dynamic edge updates remain a major...
-
03
Turning TS2Vec into a Foundation Model for Time Series Classification
Vasilii Feofanov, Zhenghan Tai, Andrei Ermakov, Irina Rish
cs.LG · cs.AI
TS2Vec is one of the strongest self-supervised methods for time series classification, introduced in 2022. It is designed to be trained anew on every dataset: self-supervised pre-training followed by linear probing. Since then, the field has shifted towards foundation models that reuse one encoder across datasets. Yet TS2Vec offers a compact convolutional architecture that operates directly on temporal samples and accepts variable-length...
-
04
PaPi: Parallel Picard-Duhamel Integration for Fast Neuro-Spectral Architectures
Arthur Bizzi, Márcio Marques, Kleyton Da Costa, Vitor Balestro, Tiago Novello, João M. Pereira, Daniel Yukimura
cs.LG
Neuro-spectral architectures (NeuSA) are a class of physics-informed neural networks (PINNs) that have proven effective for solving time-dependent PDEs in the frequency domain. However, training and inference often rely on costly numerical integration of the learned vector field, typically using sequential Runge-Kutta time stepping, which limits temporal parallelism. We introduce PaPi, an integration scheme inspired by weakly nonlinear flows...
-
05
Continuous-Time Policy Gradient from Discrete Data: Convergence Analysis and Model-Free Algorithms
Martin Hernandez, Xinyan Qing, Mo Zhou, Yuhua Zhu
cs.LG · math.OC
Continuous-time reinforcement learning seeks an optimal policy under unknown dynamics driven by stochastic differential equations using observations collected at discrete times. We develop a policy gradient framework that combines rigorous convergence analysis with learning from discrete data. We first establish a performance difference lemma and derive policy gradient formulas for randomized and parametrized policies. We then prove...
-
06
The Token-Energy Law of LLM Decode: Active Bytes and Realized Time Account for Joules per Token
Chenxu Niu, Batuhan Sencer, Wei Zhang, Yong Chen
cs.LG · cs.DC · cs.PF
Energy-efficient deployment of large language models requires understanding how model architecture, hardware, and serving settings shape energy per generated token. We introduce the Token-Energy Law, a compact, interpretable model of GPU-board energy per output token for dense, full-history decoding. It relates token energy to two core dimensions: active bytes and realized decode time. Active bytes describe logical byte demand per output,...
-
07
The Dynamics of Collapse and Recovery in Clustered Federated Learning
Priyanka Sinha, Snigdhansu Chatterjee
cs.LG · math.DS
Clustered federated learning (CFL) can suffer from cluster collapse. In particular, under unfavorable initialization, early client assignments concentrate heterogeneous groups on one cluster model, which is then trained toward a pooled compromise, while another model receives no updates. Since the current models determine client routing and routing determines subsequent model updates, we formulate clustered federated learning as control of an...
-
08
Asymmetric Bidirectional Predictive Dynamics in Google Search and COVID-19 Time Series: A Multi-Horizon Forecast-Error Analysis
Andrei Velichko, Imen Mahmoud
cs.LG
Online search activity is widely used as a digital trace of information seeking during epidemics, but its predictive value is often reduced to correlation, a selected lag, or a single forecast-error score. We examine eight pre-specified Google Trends-COVID-19 pairs in both directions using matched LightGBM forecasts over rolling 400-day windows and horizons of 1-14 days. Log-RMSE gain is combined with forecast-error distribution entropy...
-
09
Convergence of stochastic gradient descent for learning linear neural networks
Gabin Maxime Nguegnang, Bubacarr Bah, Holger Rauhut, Ulrich Terstiege
cs.LG · math.OC
Stochastic gradient descent (SGD) is the cornerstone of modern deep learning, yet its theoretical convergence properties remain poorly understood in non-convex settings. To address this gap, we study SGD dynamics in deep linear neural networks - simplified models that retain key non-convex characteristics while remaining analytically tractable. Building on recent results for gradient descent, we extend convergence guarantees to SGD under...
-
10
When Do LLMs Help with Feature Selection? A Systematic Empirical Study
Fateme Memar, Dongjie Wang, Arvin Agah
cs.LG · cs.AI
Large language models (LLMs) can leverage semantic feature names alongside summary statistics for tabular feature selection, yet it remains unclear when such information improves subset selection relative to established statistical methods. We investigate this question within an 18-dataset framework comprising 10 datasets for Group Relative Policy Optimization (GRPO) training, 2 for checkpoint validation, and 6 held-out evaluation datasets....
-
11
Learning Computational Structure for Sample-Efficient Prediction
Hangyu Zhou, Constantine Dovrolis, Stephen Mussmann
cs.LG · cs.AI
Statistical learning in high dimensions suffers from the curse of dimensionality. Fortunately, many functions of practical interest are compositionally sparse, meaning they admit a composition of local functions, each depending on only a few variables, allowing sample complexity to depend on local arity rather than ambient dimension. Yet realizing this benefit requires exploiting hidden compositional structure, which theory does not establish...
-
12
Discount Yourself: Self-inclusive Freshness-weighted Gossip for Decentralised Learning under Client Outages
Qiyuan Wang, Ke Xiao, Christos Anagnostopoulos
cs.LG · cs.DC
Decentralised federated learning (DFL) trains personalised models on devices that average with their neighbours rather than with a central server. Real devices, however, are often offline for hours or days, and a client returning from an outage can recover only by aggregating the models its neighbourhood holds. We observe that the stalest of these models is usually the client's own, a case that existing staleness-aware rules, which discount...
-
13
Binary Discrete Diffusion: Candidate-Set Diffusion for Logic and Planning Tasks
Rodrigo Casado Noguerales, Dimitri von Rütte, Bernhard Schölkopf, Thomas Hofmann, Aran Raoufi
cs.LG · cs.AI · cs.CL · stat.ML
Discrete diffusion models iteratively denoise a noisy sequence of categorical data such as text, enabling any-order generation with self-correction and in-filling capabilities. This makes them promising candidates for reasoning, planning, and constrained-generation tasks like coding and math. Yet masked and uniform diffusion erase a token's identity in a single corruption event, so their noisy states cannot explicitly represent partially...
-
14
GraphCast-Mamba: Learning Non-Markovian Corrections for Weather Forecasting
Ilya Vilkoviskiy, Liang-Hong Mo, Suozhi Huang, Dmitry A. Abanin
cs.LG
Unresolved atmospheric processes can introduce history dependence beyond the short input windows used by many weather forecasting models. We introduce GraphCast--Mamba, which augments a frozen, pretrained $1^\circ$ GraphCast backbone with a trainable Mamba correction branch combining graph message passing and Mamba layers. The branch retains a temporal memory state across successive six-hour forecast steps, learning corrections informed by...
-
15
ECG for laboratory values:A comprehensive physiological and technical analysis
Walentin Palanchuk
cs.LG · cs.AI
Laboratory measurements are essential for clinical decision making but typically require invasive blood sampling. Recent research suggests that ECG recordings may contain information beyond cardiac rhythm and provide information related to laboratory values. This thesis investigates NT-proBNP prediction from ECG recordings and patient metadata using a multimodal deep learning framework, while considering the physiological plausibility of...
-
16
Present but Hard to Read: Probing Latent Reasoning with Minimal Pairs
Aur Shalev Merin
cs.LG · cs.CL
Probing studies of latent reasoning often read a model's state where the answer begins and take a failed probe as evidence that a value is absent. We test this with a digit copied from the question, in the looped model Ouro-1.4B and the standard transformer Qwen2.5-1.5B. Comparing prompts that differ only in that digit shows it is present at the last prompt token, in a code shared across questions (readout about 0.9). A probe that sees one...
-
17
Model- and Layer-Level Closed-Loop Forecasting with SSM-Based Spiking Neural Networks
Erik Stolt, Sanja Karilanova, Ayca Ozcelikkale
cs.LG · cs.AI
Deep State-Space Models (Deep SSMs) have shown promising performance and computational benefits in processing of time-series data with complex temporal dependencies. Spiking neural networks (SNNs) constitute a biologically inspired class of deep SSMs, where computing blocks communicate through spikes, discrete single-bit events rather than continuous activation functions. A particularly important task in time-series processing is forecasting,...
-
18
Geometry-Agnostic Data-Driven Discovery of Extreme Scenarios
Angeline Hu, Ava K. Tse, Kaitlyn S. Yasumura, Luciana Sarabia, Malcolm Yeung, Kavin K. Prasanna, Kai Wong, William...
cs.LG · math-ph
We introduce a two-stage data-driven neural framework for discovering dynamically admissible extreme scenarios. Rather than assuming prior knowledge of the governing equations, the first stage infers the effective dynamical constraints directly from observational data. When the underlying physical system possesses suitable symmetries, the learned dynamics can be represented in a geometry-agnostic form, independent of any fixed grid,...
-
19
LocaLeJEPA: Maximum Entropy Representations under Geometric Priors
Kartik Tandon, Alessio Borgi, Pavel Snopov, Inés García-Redondo, Davide D'Ascenzo, Sebastiano Cultrera di Montesano,...
cs.LG · cs.CV
Joint-Embedding Predictive Architectures (JEPA) offer an appealing approach to self-supervised learning, but are typically prone to representation collapse. LeJEPA resolves this by introducing a theoretically grounded distribution-matching objective that drives the embedding towards an isotropic Gaussian. However, this approach is fundamentally tied to Euclidean assumptions, and its global target does not reflect domain-specific geometric...
-
20
CNN-based classifier for subject-independent MBTI Extraversion/Introversion preference classification from short-time EEG signals
Filip Postepski, Grzegorz M. Wojcik
cs.LG
This study examines whether short, minimally preprocessed EEG segments contain sufficient information to distinguish between individuals classified as extraverted or introverted according to the Myers--Briggs Type Indicator (MBTI). EEG recordings from 50 participants were divided into 4-s segments and analyzed using a subject-independent nested 5-fold cross-validation procedure to compare EEGNet, EEGProgress, and a newly proposed compact...
-
21
Do Neural Models Prefer Their Submodels?
Alvaro Ribot, Yulia Alexandr, Anna Seigal
cs.LG · math.AG · stat.ML
A neural model is the set of functions that can be expressed by a given neural network architecture. Neural models contain natural submodels, such as those obtained by setting some neurons to be inactive. We study the geometry of neural models and ask whether they prefer their submodels. By this we mean whether local minimizers of the loss function lie in a given submodel. We introduce a geometric framework to determine if this occurs. For...
-
22
Region-Specific EEG Classification of Hypnotic Susceptibility Using a Sliding-Window 2D CNN: A Subject-Independent Evaluation
Krzysztof Wrobel, Grzegorz M. Wojcik
cs.LG
We investigate whether a 2D Convolutional Neural Network (CNN) trained on minimally preprocessed EEG signals can classify hypnotic susceptibility across distinct brain regions. Sixty-two male participants (32 Control, 30 Guided Imagery / Hipno group) underwent a 20-minute EEG recording using a 256-channel EGI cap. A sliding-window approach evaluates all 10-second windows within the recording for each of seven predefined brain regions...
-
23
Rewards for Budget-Aware Multi-Turn Agents That Adapt to Arbitrary Test-Time Constraints
Tom A. Lamb, Adam Golinski, Eleonora Gualdoni, Sinead Williamson, Michael Kirchhof
cs.LG · cs.AI
Large language model agents are often given an inference budget and asked to maximise success within it. A multi-turn agent has to allocate this budget across many environment interactions. An external harness can enforce the budget, but it remains unclear whether the agent itself can learn when and how much to spend. We ask whether agents can learn this allocation across dynamically specified budgets and which rewards best teach this...
-
24
Self-Credit: Credit Assignment for Long-Horizon Agentic RL via Intra-group Rescaling
Fangcong Yin, Ming Yin, Tanvi Bhandarkar, Aashu Singh, Eryk Helenowski, Shlok Kumar Mishra, Saghar Hosseini, Karthik...
cs.LG
LLM agents are commonly trained with RLVR, but the sparse nature of typical RLVR rewards makes credit assignment challenging, especially for long-horizon tasks. We propose Self-Credit, an end-to-end agentic RL method that creates dense supervision signals from sparse outcome rewards using self-generated credit assignment via intra-group rescaling. Self-Credit builds on group-based RL with three components. First, during its rollouts, the...
-
25
Don't Skip the Baseline: A Controlled Comparison of Dynamic Expert Skipping in MoEs
Jakob Steimle, Thomas Elsken, Daniel Ehrhardt, Lukas Rinder, Daniel Cremers, Xi Wang
cs.LG · cs.AI
A growing body of work proposes dynamic expert-skipping methods for Mixture-of-Experts (MoE) models that adjust the number of experts activated per token during inference, aiming to preserve model quality while further reducing the number of active parameters. These methods have added calibrated importance scores, inter-expert similarity metrics, or learned per-token allocators in order to identify experts to skip. However, different...
-
26
Relative Representations Enable Portable White-Box Monitoring Across Language Models
Tim Bakker, Christos Louizos
cs.LG
Independently trained autoregressive language models often acquire similar capabilities, but it remains unclear whether their internal representations organize behaviorally relevant information in compatible ways. We show that models do in fact organize information similarly by using specific behavioral probes that operate on compatible representations. This unlocks portable white-box monitoring: a probe developed and validated on one model...
-
27
Beyond the Point Mass: Acceptance-Derived Weights and Distributional Losses for Speculative Decoding
Tianyu Wu, Zhuohan Wang, Jonathan Geuter, Yu Yao, Weifan Jiang, Chengxi Zhang, Ju Li, Minlan Yu
cs.LG
In speculative decoding, a lightweight draft model proposes tokens that the target model verifies in parallel, speeding up generation without changing the target distribution. D-PACE (Wu et al., 2026) trains such drafters with a loss that weights each position by its contribution to the expected accepted length. Both its weights and its cross-entropy (CE) loss use only the draft's probability of the recorded target token, treating the target...
-
28
Characterizing Cross-Relay Generalization Failure in Machine Learning-Based Power System Protection
Nithin Pradeep Nadayil Kizhakkethil, Julian Oelhaf, Christian Bergler, Andreas Maier, Siming Bayer
cs.LG
Machine learning-based fault detectors are often evaluated at the relay locations used for training, although local voltage and current measurements vary across power networks. This work investigates cross-relay generalization for binary detection of faults anywhere in the network using local measurements, rather than line-selective tripping. Using 50 ms windows from a simulated 110 kV transmission network, we evaluate a multilayer perceptron...
-
29
Identifying Dependent Components from Multi-domain Linear Mixtures
Danru Xu, Lauri Parkkonen, Sara Magliacane, Aapo Hyvärinen
cs.LG · cs.AI
We study a linear mixing model with dependent latent components, assuming multiple data domains. Most existing models assume that the components are independent or at least uncorrelated, in line with independent component analysis (ICA). Some recent work allows for dependent components, but then makes specific assumptions such as parametric forms of dependencies, multi-view settings, or interventions, or does not recover the individual...
-
30
Observation Assimilation Transformer: Probabilistic Weather Analysis from Sparse Station Observations
Christian Donner, Michele Cattaneo, Maxim Samarin, Francesco Zanetta, Icíar Lloréns Jover, Garbiela Aznar Siguan,...
cs.LG
We present the Observation Assimilation Transformer (OAT), a geometry-aware neural process that corrects and refines a prior numerical weather prediction (NWP) field into a high-resolution posterior field by assimilating sparse observations. OAT is an encoder-decoder transformer that considers a set of observed locations and their NWP prior to condition predictions at a set of arbitrary unobserved query locations. Its sparse attention...
-
31
Manifold-Informed Neural Networks for Low-Label ECG Electrolyte Inference
Juntang Wang, Hao Wu, Yihan Wang, Dongmian Zou, Shixin Xu
cs.LG
Physics-informed neural networks (PINNs) supervise on the residual of a known differential operator. Many high-stakes inference problems admit no such operator: clinical signals, behavioral measurements, materials properties. We introduce the Manifold-Informed Neural Network (MINN), a PINN-parallel paradigm that replaces the physics operator residual with a manifold operator residual: a discrete Poisson equation $\mathbf{L}_G \mathbf{H} =...
-
32
Recommending Bacterial Cultivation Conditions from 16S rRNA
Juntang Wang, Hao Wu, Yihan Wang, Runkun Guo, Shixin Xu, Dongmian Zou
cs.LG
Recommending bacterial cultivation conditions from sequence data is a challenging structured problem characterized by missing standardized benchmarks and limited use of sequence structure. To address this, we introduce DSNI, a benchmark of 65,023 bacterial strains with phylogeny-aware splits, pairing 16S rRNA sequences with four cultivation conditions (media, temperature, pH, oxygen). Exploiting the mosaic structure of 16S rRNA, where...
-
33
Learning Normal Coordinates on Riemannian Manifolds with Flow Charts
Edmond Cunningham
cs.LG · math.DG
Riemannian manifolds are powerful models for problems with non-Euclidean structure. Recent work in geometric machine learning has focused on training deep neural networks to learn Riemannian metrics from high-dimensional data. However, many geometric algorithms require repeated evaluations of the exponential and logarithmic maps, whose high computational cost restricts the utility of learned Riemannian manifolds in practice. We propose to...
-
34
TPO: Tri-level Distributionally Robust Learning for OOD Direct Preference Optimization
Chengtao Jian, Kai Yang
cs.LG · cs.AI
Direct Preference Optimization (DPO) minimizes empirical risk over observed preference comparisons, implicitly assuming that the training preference distribution matches the deployment preference distribution. In practice, this i.i.d. assumption often fails, as preference distribution shift exhibits a natural hierarchical structure across three levels: group-level shift, prompt-level shift, and preference-level shift. Such multi-level...
-
35
JAXbounce: A Continuous-Control Pong-Like Benchmark for Active Inference Beyond Low Dimensional Environments
Chaitanya Kaul, Markus Klar, Sebastian Stein, Fraser C. Paterson, John H. Williamson, Roderick Murray-Smith
cs.LG · cs.AI
We introduce JAXbounce, a continuous control pong-like benchmark designed to evaluate active inference beyond discrete, low dimensional environments. JAXbounce combines continuous states and stochastic dynamics in discrete time with flexible and composable observation models that allow representing the environment as a state vector or a pixel grid, corruptible through the rich family of noise models we provide. It is fully differentiable and...
-
36
Parameter-Efficient Adaptation of Time Series Foundation Models for Electrical Load Forecasting
Aliullah, Julian Oelhaf, Edouard Fouché, Jessica Deuschel, Christian Bergler, Andreas Maier, Siming Bayer
cs.LG · cs.AI
Accurate electrical load forecasting supports power-system operation across countries, regions, cities, and buildings, but repeated development of target-specific models is costly. Time series foundation models (TSFMs) provide reusable pretrained backbones, although multi-domain pretraining may not fully capture load-specific characteristics. This paper compares pretrained inference, full fine-tuning, and low-rank adaptation (LoRA) in terms...
-
37
Reframing Object-Centric Reinforcement Learning: From Absolute to Relative Representations
Cedric Derstroff, Jannis Blüml, Janosch Moos, Michael Girstl, Zihan Ye, Debora Clever, Kristian Kersting
cs.LG · cs.AI
Object-centric reinforcement learning agents typically receive each object's position in a fixed global frame, even though their decisions usually depend on where objects are relative to one another. These absolute views often fail to capture the local context that is most relevant to the agent's behavior. As a result, an agent must relearn the same response at every location where a configuration can occur, and its performance collapses when...
-
38
Adaptive Pullback: Reweighting Supervision at Penalty Jumps in On-Policy Distillation
Guangyu Liu, Siyuan Gan, Shilong Zhou, Linjian Meng, Boyan Wang, Jing Huo
cs.LG · cs.AI
On-policy distillation (OPD) supervises student-generated responses using sampled-token student-teacher log-ratios. A positive log-ratio penalizes a token that the student assigns higher probability to than the teacher, but its current value does not distinguish sustained high penalties from abrupt increases along a response. We examine these local changes by comparing consecutive blocks of tokens within the same response. Our diagnostic...
-
39
Coincidence-Maximizing Customer Clustering Using Smart Meter Data
Ramiro H. Gálvez, Gabriel Martos Venturini, Iván Mondrzak
cs.LG
Electricity distribution planning depends on peak-demand coincidence within customer groups, yet similarity-based clustering of smart-meter data does not optimize it directly. We formulate customer grouping to maximize the average of group coincidence factors, weighted by each group's daily share of summed individual peak demand. We prove that jointly optimizing memberships and group peak times is strongly NP-hard even for two groups. We...
-
40
Curriculum Learning for Generative Flexible Docking
Wenyin Zhou, Hossein Azizpour
cs.LG
Flexible docking is a highly practical task, aiming to predict the joint configuration of a protein-ligand complex from a protein apo structure and a ligand SMILES string. Diffusion and flow-based generative models have made this task tractable by modeling the posterior distribution over the ligand pose and the protein conformation conditioned on the apo input. The central challenge, however, remains the long-tailed apo-to-holo structure...
-
41
Causal Discovery for Linear Models with Latent Confounding via Multi-domain Identifiable Representation Learning
Danru Xu, Aapo Hyvärinen, Sara Magliacane
cs.LG · cs.AI
Causal discovery with latent confounding is particularly challenging, but a common setting in real-world data. In this work, we focus on linear causal models with latent confounding, represented by acyclic directed mixed graphs (ADMGs). Existing methods that fully identify ADMGs typically assume non-Gaussianity or graphical constraints, such as bow-freeness. In contrast, we propose Linear ADMG Learning (LinADMG), a framework for learning...
-
42
From Complex Histories to Simple Predictors: Predictive Dimension in Dynamic Graph Learning
Hantong Feng, Duxin Chen, Wenwu Yu
cs.LG · cs.AI
Existing continuous-time graph learning methods rely heavily on complex neural architectures to encode ever-expanding interaction histories. However, the future conditional distribution induced by complex histories can often be characterized by a low-dimensional predictive state manifold, implying that only a few predictive degrees of freedom are essential for future interactions. We therefore argue that continuous-time graph learning should...
-
43
PULSE: Personalized Opportunistic Learning for Wearable Devices
Mattia Giovanni Campana, Dariush Salami, Hüseyin Yiğitler, Stephan Sigg
cs.LG · cs.AI · cs.DC
Wearable devices increasingly support personalized health and well-being applications, but post-deployment data can differ substantially from those used for training, while purely local adaptation may be limited by the data available on a single device during real-world operation. Conventional federated and decentralized learning rely on recurrent communication and stable connectivity, whereas existing opportunistic approaches often exploit...
-
44
MuPACE: Learning to Pace Muon for LoRA Fine-Tuning
Jan Kobiolka, Christian Frey, Josif Grabocka
cs.LG
Learning rate schedules are typically fixed before training begins, yet when fine-tuning large language models with low-rank adapters (LoRA), the appropriate scale varies substantially with the model, task, and adapter configuration. Practitioners re-tune a schedule for every new problem, yet a tuned schedule still cannot react to the training dynamics it encounters. We instead amortize learning rate selection across LoRA fine-tuning...
-
45
Universal Online Convex Optimization With Delayed Feedback
Hao Qiu, Andrew Jacobsen, Emmanuel Esposito, Mengxiao Zhang
cs.LG
We study universal online convex optimization with delayed feedback. We develop a single algorithm that simultaneously adapts to general convex, strongly convex, and exp-concave losses without prior knowledge of the curvature class or parameters. Our regret bounds match the best known dependence on the time horizon $T$ and delays up to doubly logarithmic factors: roughly $\sqrt{T + d_{\mathrm{tot}}}$ for convex losses and $σ_{\max}\log T$...
-
46
Poisoning-Based Backdoors Are Not Cryptographic
Masahiro Kaneko
cs.LG · cs.CL
To defend against backdoor attacks on neural networks, the defender must identify the verifier function that determines whether an input is a trigger, rather than rejecting individual triggers in isolation, since an attacker can generate new triggers satisfying the same verifier faster than they can be blocked. If a backdoor verifier could be installed with the asymmetry of a cryptographic authentication scheme, in which only the attacker can...
-
47
When Does Training on Synthetic Data Provably Work?
Leroy Odunlami, Tobias Riedlinger, Hanno Gottschalk
cs.LG · stat.ML
High-quality training data are a major bottleneck in machine learning, motivating increasing use of artificially generated data. We study when a binary classifier trained on synthetic data transfers successfully to a real distribution, under a shared conditional label distribution. First, we derive finite-sample excess-error bounds when the real covariate distribution is absolutely continuous with respect to the synthetic distribution under a...
-
48
From RL Papers to Framework Code: Inclusion and Review in Seven Open-Source Training Frameworks
Taiheng Pan
cs.LG · cs.CL · cs.SE
Practitioners use reinforcement learning methods for language models through training frameworks, so a method that a framework includes is one configuration change away for its users. We measure how often this happens. We trace seven open-source frameworks for RL post-training through their commit histories and link them to 6,939 screened arXiv papers from January 2025 to June 2026. Of 1,598 papers that propose a general RL training method,...
-
49
RANGER: Retrieval Agent for Navigating Graphs with Semantic Evaluation Rewards
Artem Mikhaylov, Pavel Konoplev, Nikolay Efanov, Artyom Sorokin, Evgeny Burnaev
cs.LG
The integration of Large Language Models and Reinforcement Learning offers a promising direction for complex sequential decision-making tasks, such as multi-hop question answering over knowledge graphs. While LLMs provide strong semantic understanding and reasoning capabilities, they often struggle with precise, multi-step navigation in structured environments. Conversely, RL agents excel at learning navigation policies but are hindered by...
-
50
When Momentum Helps Polyak--Ruppert Averaging
Alexandra Suvroikova, Darina Dvinskikh, Yuriy Dorn, Alexander Gasnikov
cs.LG · math.OC
Polyak--Ruppert averaging gives SGD and momentum methods the same asymptotic covariance. We characterize how fast finite runs reach it and how far they can be parallelized. For strongly convex quadratics with additive noise, we prove an online--offline bridge: the empirical risk minimizer on the samples of a tail average is one Newton step from the tail-averaged query point of any online procedure. For SGD, heavy ball, and Nesterov's method...
-
51
On Johnson-Lindenstrauss Transforms for Byzantine-Robust Optimization
Aleksandr Shestakov, Vladimir Aletov, Aleksandr Beznosikov
cs.LG
The growing scale of data and models in machine learning calls for efficient optimization methods. Distributed optimization addresses this challenge, with compression mechanisms helping to reduce communication costs. A classical tool for dimensionality reduction is the Johnson-Lindenstrauss (JL) transform, which is simple to implement and offers strong geometric guarantees. In particular, JL transforms approximately preserve inner products...
-
52
Unsupervised Head Network Personalization for Split Computing via Federated Learning
Yutaro Horikawa, Takayuki Nishio
cs.LG · cs.DC · cs.NI
Split Computing (SC) is a distributed inference method designed for balancing computational load and reducing latency by partitioning a neural network into head and tail networks and deploying them across both a mobile device and a server, respectively. Due to the limited bandwidth and computational power of mobile devices, the head network is often replaced with a neural network that requires lower computational and communication resources....
-
53
Near-Duplicate Image Selection Transfers Preferences in Vision-Language Models
Jaeyoung Kim, Eunseok Kim, Dongsuk Jang, Seunghyun Kim
cs.LG
Subliminal learning is the phenomenon in which a model transmits its preferences to another model through data that do not explicitly express them. As models increasingly curate training data for other models, we ask whether this transfer can arise from choosing among near-duplicate images while keeping the source photographs and human-written captions fixed. For each photograph, a curator selects one of eight near-duplicate variants by...
-
54
The Shape of the DPO Hyperparameter Landscape: What to Measure, How It Moves with Scale, and When Not to Transfer It
Hannan Mahadik, David Salinas, Jacek Golebiowski, Aaron Klein
cs.LG · cs.AI
Direct preference optimisation (DPO) is widely used for language model post-training, but its performance depends strongly on the choice of its hyperparameters, such as the learning rate and the reference-policy regularisation parameter $β$. To better understand the DPO hyperparameter landscape, we conduct extensive experiments using a dense hyperparameter grid across different model scales from the same family. This allows us to analyse how...
-
55
Physical-Variable-Informed Deep Learning for Robust Diagnosis in Industrial Fluid Transport Systems
Yu Sha, Junqi Tao, Dixin Zhou, Horst Stoecker, Domagoj Vnucec, Andreas Widl, Kai Zhou
cs.LG · cs.AI
Industrial fluid transport systems are monitored through heterogeneous sensing streams, but robust fault diagnosis remains difficult because response signals vary with operating conditions and differ from physical variables in sampling rate, observation duration and data granularity. To address these challenges, we propose a physical-variable-informed deep learning framework (PhyNet) for robust fault diagnosis under real-world sampling...
-
56
Elicitation or Expansion? Demystifying Gains from Agent Harness Optimization
Younghyun Kim, Wooseok Seo, Minseo Kim, Seunghyuk Oh, Minjae Lee, Donghoon Kim, Hyung Il Koo, Jaehyung Kim, Wonjun Kang
cs.LG
Recent studies in harness optimization have shown substantial performance gains in language model agents without updating model weights. A well-designed harness more effectively leverages the underlying model's capabilities by shaping context, tool use, and execution workflows. However, naïve success rates alone do not fully characterize whether optimized harness yields more reliable success on tasks already within reach (elicitation) or a...
-
57
From Generic to Dedicated: A Novel Optimizer for Online Continual Learning
Yongyi Wu, Zheng Wang, Sen Lin
cs.LG · cs.CV
Online Continual Learning (OCL) requires models to learn from non-stationary data streams under a strict single-pass constraint, making them highly susceptible to catastrophic forgetting. While existing studies have explored various strategies from data, architecture, and optimization perspectives, the optimizer often directly uses generic approaches such as SGD and AdamW. In this work, we reveal a new link between optimization dynamics and...
-
58
SliMOO: Interpretable Multi-Objective Evolutionary Search for LLM Depth Pruning
Guanchen Li, Yixing Xu, Xuanwu Yin, Dong Li, Emad Barsoum
cs.LG
As large language models (LLMs) grow more powerful, their scale makes them costly to serve. Depth pruning offers a practical route to efficiency by directly reducing memory and latency without hardware-specific support. However, we observe that transformer layers exhibit markedly different task-dependent layer contributions, whereas existing depth-pruning methods ignore this heterogeneity and thus tend to optimize for a narrow domain or...
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.