cs.LG · 2026-11-14 · No. 171
Machine Learning, 2026-11-14.
54 new papers in cs.LG. Titles, authors,
abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →
01 — The papers
54 entries-
01
Internalizing Investment Mandates: A Neurosymbolic RL Approach to Portfolio Allocation
Zeynep Yıldız, Satchit Chatterji, Ercan Atam, Erman Acar
cs.LG
Institutional portfolios operate under investment policy statements that specify asset allocation constraints, often conditioned on market regime information. Deep reinforcement learning has been shown to optimize financial performance but typically does not enforce rule compliance, making it unsuitable for real-world deployment. Existing neurosymbolic methods such as Probabilistic Logic Policy Gradient (PLPG) offer constraint-satisfying...
-
02
Linking Rapid Learning to Compositional Structure in Neural Networks: Evidence from Additive Analogies in Meta-Learned Parameter Spaces
Abigail L. Tenenbaum, R. Thomas McCoy
cs.LG · cs.CL
Humans can learn rapidly from limited experience, while neural networks typically require extensive training. In this work, we investigate what properties of a neural network can enable it to learn rapidly. We study neural networks that have been optimized via meta-learning, a technique that exposes a network to many related functions to facilitate its subsequent learning of similar functions. Prior work has shown that these meta-learned...
-
03
Towards a Theory of Learnability via Debate
Tosca Lechner, Vinayak Pathak, Sivan Sabato, Ruth Urner
cs.LG
Debate has been recently proposed as a training protocol that is hoped to produce more aligned models even as model capabilities scale. The idea is that the human, instead of providing supervision in the form of labels, judges a debate between two models that compete to prove the correctness of opposite labels. It can be shown that functions of a much higher computational complexity can be correctly judged in this way, compared to the...
-
04
Metal Sandbox: Isolated Execution and Command Scheduling for GPU Coding Agents on Apple Silicon
Haozhe Wang, Lina Chen, Fanqi Wan, Qi Han, Yinmin Zhang, Zheng Ge, Daxin Jiang, Junchi Yan
cs.LG · cs.AI · cs.CL · cs.DC
On-device inference requires kernels that make effective use of local accelerators. Efficient training and evaluation of models that generate Apple Metal kernels benefit from native macOS tools and concurrent agent sessions. These sessions alternate between model requests, development commands, and GPU execution, leaving GPU capacity idle for much of a rollout. We present Metal Sandbox, infrastructure that combines restricted per-instance...
-
05
Negative-Only Policy Optimization for One-Sided Verifiable Rewards
Jiacheng Xu, Shuo He, Fuxiang Zhang, Chaojie Wang, Bo An
cs.LG · cs.CL
Reinforcement learning with verifiable rewards (RLVR) typically relies on a verifier that can reliably judge whether sampled solutions are correct. However, in many realistic settings, complete verification is generally unavailable, while cheap checks can only falsify failures. We formalize this regime as one-sided verifiability (OSV): verified negatives are reliable counterexamples, but verified positives are merely non-falsified candidates...
-
06
Two mathematical models of knowledge distillation
Audrey Xie, Ludwig Schmidt, John Duchi
cs.LG
Many hypotheses compete to explain the successes of knowledge distillation. To help address this, we propose and analyze a mathematical model of distillation, which suggests that distillation's performance comes not from obtaining better models but from easier to optimize landscapes. For generalized linear models trained with stochastic gradient descent, we prove that distillation fits performant student models asymptotically more quickly...
-
07
Introducing Feedback Delays to Non-Delayed Adversarial MDPs: Optimal $\sqrt S$ and $\sqrt K$ Dependencies in Tabular and Linear AMDPs
Aarush Kulkarni, Khang Nguyen, William Chang, Yan Dai
cs.LG · cs.AI · stat.ML
We study regret minimization in episodic Adversarial Markov Decision Processes (AMDPs) with unknown transitions and bandit feedback. In tabular AMDPs with $S$ states, $A$ actions, and horizon $H$, the best known $K$-episode regret is $\tilde O(HS\sqrt{AK})$, whose $\sqrt S$ gap to the $Ω(H\sqrt{SAK})$ lower bound has long remained open. In linear AMDPs with $d$-dimensional features, after a five-year line of improvements starting from...
-
08
Physically Interpretable Basin Selection in Discrete-Variable Origami Design
Lucy Liu, Leon M. Kamp, Katia Bertoldi, L. Mahadevan
cs.LG · cond-mat.mtrl-sci · cond-mat.soft
Automatic differentiation has enabled rapid progress on continuous-variable mechanical design problems, but gradient-based optimization still struggles with discrete variables. The standard approach is to optimize using a continuous relaxation of the choices, then commit to the nearest discrete design at the end. Why this procedure selects unfavorable basins is poorly understood. Here, we introduce a practical, interpretable design problem...
-
09
Maximizing Capacity Utilization of LoRA Adapters for Mitigating the Learning-Forgetting Trade-off
Tomohiro Sawada, Kartik Goyal
cs.LG · cs.CL
Low-Rank Adaptation (LoRA) often underperforms full fine-tuning, especially in heterogeneous and continual-learning settings in which substantial forgetting occurs. In general, it suffers from an undesirable learning-forgetting trade-off across diverse fine-tuning settings. We identify a potential contributor: LoRA adapters, especially high-rank ones, frequently fail to realize the available capacity afforded by the configured rank....
-
10
Symmetry in Time: A Representation Theory of Linear Recurrent Memory
Anif N. Shikder, Ramit Dey, Jan Minac, Lyle E. Muller
cs.LG · cs.NE · math.RT
State space models (SSMs) are an emerging alternative to Transformers. SSMs offer linear-time inference and parallel training, but most implementations make the parallel scan efficient by making the recurrence diagonal: each channel of the hidden state is multiplied by an input-dependent real or complex number. These numbers commute, so the state records which inputs occurred but not their order. We formalize this with representation theory,...
-
11
AeroZoo: A Benchmark of Well-Posed Flow Problems for Foundation Models of Aerodynamics
Peter Sharpe, Carmelo Gonzales, Corey Adams, Mehdi Ataei, Sanjay Choudhry
cs.LG · cs.CE · physics.flu-dyn
Neural surrogates for computational fluid dynamics (CFD) are trained and ranked on datasets that each vary one design family in one flow domain. Such datasets cannot test the least that an aerodynamic foundation model must do: predict CFD-quality incompressible flow in seconds for new shapes in new enclosures. On DrivAerML, larger models and longer training barely lower Transolver's error; doubling the training cases cuts it by about 12% up...
-
12
Enhancing Preference Detection using Eye Movement Cascade and Multilayer Combinatorial Fusion
Yuanhong Wu, Mary Tsaryk, Sam Watson, Tahsin Ahmed, Christina Schweikert, D. Frank Hsu
cs.LG
Understanding how visual attention influences human preference formation is a fundamental problem in consumer behavior and cognitive science. In this work, we investigate how eye movement patterns can be used to predict preference decisions when subjects compare two alternatives. To address this problem, we analyze an eye-tracking dataset, where subjects view pairs of face images and their gaze behavior is recorded throughout the decision...
-
13
A Convex-Analytic Reformulation for Poisoning Attacks on Logistic Regression
Carina Moreira Costa, Maria Eduarda Pinheiro, Martin Schmidt
cs.LG · math.OC
We study data poisoning attacks on logistic regression with $\ell_2$-regularization, where an attacker chooses the features of a fixed number of injected observations with prescribed labels to maximize the logistic loss on the clean training data. Strong convexity of the defender's problem yields an exact reformulation through its first-order optimality conditions. Using convex conjugacy of the softplus function, we replace the sigmoid...
-
14
ISLA: A Coordinate-Free, Measure-Weighted Transformer for Boundary-Driven PDEs
Peter Sharpe
cs.LG · cs.CE · math-ph · physics.flu-dyn
A foundational principle of physics is that all coordinate systems are equally valid: the behavior of a PDE-governed system does not depend on the origin or orientation of the frame, nor on the units or mesh used to write it down, and neither should a surrogate model. Yet the neural surrogates with the lowest errors on today's aerodynamics benchmarks lack these invariances and fail under simple transformations of the same problem: for models...
-
15
Support-Constrained Offline Policy Improvement for Pediatric ECMO from Limited Clinical Data
Fateme Golivand Darvishvand, Saurabh Mathur, Nikhilesh Prabhakar, Michael A. Skinner, Ameet Soni, Neel Shah, Ethan L...
cs.LG
Managing children on ECMO (Extracorporeal Membrane Oxygenation) requires balancing short-term physiological stability with the long-term goal of preventing neurological injury. Standard reinforcement learning can propose risky, out-of-distribution actions, while imitation learning is limited to reproducing clinician behavior. We introduce Support-Constrained Fitted Q-Iteration (FQI-SC), an offline reinforcement learning framework that...
-
16
Paper-Faithful Federated Learning: Scientific Checks for Practitioner and Agent Development
Ali Dadras, Alp Yurtsever
cs.LG · cs.SE
Federated learning experiments can produce reasonable-looking results even when the experiment implementation differs from the intended experimental protocol. Small errors in configuration, data partitioning, aggregation, randomness, evaluation, or checkpoint recovery may therefore affect a comparison without causing the program to fail. We present FedBrew, a research codebase for controlled and reproducible federated learning experiments....
-
17
Using an energy pixels approach to estimate the size a 2nd Generation DEL must be to train a high-resolution binding model
Boris P. Belotserkovskii, R. Edward Watts
cs.LG · q-bio.BM
DNA-encoded libraries (DELs) are commonly used to find small-molecule binders to a target protein. But the hard part is no longer finding a hit. It is building a machine-learning model good enough to guide the medicinal chemistry that follows, one that can resolve the shift in affinity when a chemist changes a single atom. Such a model needs two things from its training data: dense sampling of the small patch of chemical space around the hit,...
-
18
Bellman Contraction under MMD: A Unified Framework
Baptiste Debes, Tinne Tuytelaars
cs.LG
We introduce a mathematical framework for studying contraction of the distributional Bellman operator under Maximum Mean Discrepancy (MMD). We show that contraction reduces to conditional positive definiteness of an associated kernel constructed from the base kernel, which we call the Bellman difference kernel. Focusing on radial kernels, we use Schoenberg's theorem to connect this condition to Bernstein functions. This gives contraction...
-
19
Near-Optimal Regret for Oracle-Efficient Adversarial Contextual Bandits
Kiarash Banihashem, MohammadTaghi Hajiaghayi
cs.LG · cs.DS
We study contextual bandits with $K$ actions, a class of $N$ policies, and i.i.d.\ contexts from a distribution that the learner can sample. For any sequence of loss functions fixed before play, we give an oracle-efficient algorithm with expected regret $O(\sqrt{KT\log(2N)}\,\log(2KT))$ against the best policy in hindsight on the realized contexts. This improves the $T^{2/3}$ dependence on the horizon in prior oracle-efficient bounds and...
-
20
LazyWeights: Decode-on-Demand Edge LLMs
Arnab Sanyal, Gourav Datta, Sandeep P. Chinchali
cs.LG · cs.AR · cs.CL · cs.IT
Memory decides which LLMs an edge device runs, and how fast: capacity on a $4$ GB Jetson Nano, bandwidth on an $8$ GB Jetson Orin Nano Super. Deployed formats (GPTQ, AWQ, $\texttt{llama.cpp}$'s K- and I-quants) spend equal bits on every integer code of a matrix, although the codes are far from uniform. $\texttt{LazyWeights}$ keeps these formats but tunes GPTQ so the codes compress, Huffman-codes them losslessly in interleaved lanes, and...
-
21
Capacity-Constrained Delayed Feedback for Online Conformal Risk Control
Bowen Wang, Osvaldo Simeone
cs.LG · eess.SP
Online conformal risk control (OCRC) enables reliable uncertainty quantification in machine learning by converting model outputs into prediction sets with long-term risk guarantees. These guarantees, however, hinge on the assumption that the loss of each round is revealed immediately and losslessly. In practice, feedback may be delayed, and limited resources may prevent tracking all pending predictions, causing some feedback to be lost. To...
-
22
Temporal Regularized Learning: Self-Supervised Learning Local in Space and Time
Davide Wiest
cs.LG · stat.ML
Layer-local self-supervised objectives remove the backward pass between layers, but almost all still compute their loss over a batch of samples and are therefore local in space but not in time. We introduce Temporal Regularized Learning (TeReL), a spatially and temporally local learning rule for self-supervised representation learning from sequential input streams. The key idea is to instantiate a VICReg-style objective with temporally...
-
23
Time Flies When You Parallelise: Pareto-efficient Time Series Foundation Models with Linear Recurrences
Torben Berndt, Lingyi Yang, Federico Nardi, Yi Heng Lim, AJ Shears, Benjamin Walker
cs.LG
Time series foundation models have demonstrated strong zero-shot forecasting performance, but computational efficiency remains a central architectural challenge. Temporal self-attention incurs quadratic cost in the number of temporal tokens, while nonlinear recurrent state transitions restrict parallel computation along the temporal dimension. We introduce ci-proto, a family of time series foundation models based on Mamba-3, whose linear...
-
24
Factored One-sided Adam-Moment for Practical and Scalable SOAP
Jaemyung Yu, Byeongho Heo, Sangdoo Yun, Dongyoon Han
cs.LG
Second-order optimizers (e.g., Shampoo, SOAP) have emerged as compelling alternatives to Adam/AdamW and promise faster per-step convergence than AdamW in transformer training. However, they still remain costly in practice, and even SOAP still pays substantial optimizer-state memory and eigendecomposition overhead. We ask whether the benefits of SOAP can be retained at nearly the efficiency of AdamW. In this paper, we answer yes with a...
-
25
DPA: Decentralized Primal Averaging with Quasi-Global Momentum for Highly Heterogeneous Data
Diyako Ghaderyan, Stefan Werner
cs.LG · cs.AI
Decentralized training of deep learning models keeps training data local and supports on-device learning over networks. In realistic scenarios, heterogeneity across agents' local data poses an optimization challenge and can severely degrade test accuracy, especially on sparse communication graphs. Building on recent advances in centralized primal averaging, we propose Decentralized Primal Averaging (DPA), a decentralized learning algorithm...
-
26
Sustainable Federated Fine-Tuning with LoRA: Environmental Impact and Performance Footprint
Pedro Augusto Londe Ribeiro, Rodrigo Moreira, Larissa Ferreira Rodrigues Moreira, Joberto S. B. Martins, Tereza C....
cs.LG · cs.NI
Federated Learning (FL) with Parameter-Efficient Fine-Tuning (PEFT) enables efficient adaptation of Large Language Models (LLMs), yet its energy and carbon costs across hardware remain poorly understood. We evaluated federated Low-Rank Adaptation (LoRA) fine-tuning on NVIDIA A40, RTX 6000 Ada, and A100 GPUs using a $2^2$ full factorial design with 135M/360M models and 10/100 local steps. CodeCarbon tracked energy and estimated...
-
27
Calibration and Conformal Prediction for Medical Question Answering in Arabic and English: Evidence from Two Non-Parallel Benchmarks
Mariya
cs.LG
Large language models (LLMs) are increasingly tested on medical exam questions, but accuracy alone does not show when an answer can be trusted. I study the calibration and uncertainty of two open-weight models, Qwen2.5-7B-Instruct and Jais-adapted-7B-chat (both run in 4-bit precision), on 2,816 English multiple-choice questions from MedMCQA and 4,927 Arabic multiple-choice questions from MedAraBench. Accuracy was 0.561 (English) and 0.430...
-
28
Detection Is Not Early Warning: Evaluating Token-Level Alarms on Language Model Failures
Marco De Negri, Luca Sartori, Lorenzo Baggi, Moritz Reihs, Eduard Frank Ďurech, Mennatallah El-Assady, Valentina...
cs.LG · cs.CL
Many failures of language models begin at a specific token, the onset. A monitor that reads the model during generation can classify the finished response, detect the failure once it is under way, or warn before it begins, and only a warning leaves time to stop or steer the generation. Monitors are usually evaluated with response-level metrics such as AUROC, which ignore when the monitor fires. We introduce PREWARN, an evaluation protocol for...
-
29
ScrewSense: Mapping the Invisible - Anomaly Detection and Error Classification in Screwdriving Processes
Tobias Lettner, Tatyana Ivanovska, Anwar Al Assadi, Christian Bergler
cs.LG · cs.RO · eess.SP
The growing volume of process data generated by screwdriving systems, combined with advancements in computing power, data storage, and algorithms, is facilitating the application of artificial intelligence in this field. Accurately detecting and classifying screwdriving errors is crucial for ensuring product quality, avoiding costly rework in manufacturing, and supporting sustainability by facilitating the maintenance, reuse, and recycling of...
-
30
Class-Competence Aggregation: Output-Layer Routing for Federated Multiclass DDoS Detection under Class-Concentrated Non-IID Data
Tapu Datta, Hafiz Imtiaz
cs.LG · stat.ML
Federated learning over distributed intrusion-detection sensors naturally produces \emph{class-concentrated} non-IID data, where an entire attack class is held by a single client and absent elsewhere. We show this collapses federated averaging in a specific, localized way: the output layer holds one decision boundary per class, and when a class's aggregation weight comes overwhelmingly from clients that have never seen it, its neuron receives...
-
31
Beyond Binary Classification: Multiclass Machine Learning for Diabetes and Prediabetes Risk Prediction Using NHANES
Aman Mujahid, Anita Penkova
cs.LG
Detecting diabetes risk early is important for preventing disease progression. Most existing machine learning approaches focus on binary classification, which limits their ability to capture intermediate risk states such as prediabetes. We develop a supervised learning framework to classify adults as normal, prediabetic, or diabetic using National Health and Nutrition Examination Survey data. The framework uses routinely collected...
-
32
Interpretable machine learning diagnosis of degradation mechanisms in lead-free $RbSn_{0.5}Ge_{0.5}I_{3}$ solar cells from temperature- and illumination-dependent electrical measurements
Nabanita Sarker, Hafiz Imtiaz
cs.LG · math-ph
Machine learning coupled with device simulation is widely used to optimize thin-film solar cells, but almost exclusively in the forward direction, from device parameters to predicted performance. The inverse problem encountered after fabrication, namely identifying the mechanism that limits a given device, is still addressed by manual inspection of characterization data. In this work, both directions are examined for lead-free...
-
33
On the Difficulty of Sequence-Level Supervised Learning for Autoregressive Language Models
Mathieu Blondel, Michaël E. Sander, Germain Vivier-Ardisson, Tianlin Liu, Vincent Roulet
cs.LG · stat.ML
During the supervised learning phase, autoregressive language models are typically trained by minimizing the negative log-likelihood of the training data, which decomposes as a sum of token-level cross-entropies via teacher forcing. This objective is simple and efficient but potentially suffers from so-called exposure bias: a mismatch between the way the model is trained and the way it is used at inference time. Can we instead design...
-
34
From Data to Wisdom: Rethinking AI Through an Epistemic Lens
Afonso Eduardo
cs.LG · cs.AI
This paper proposes the Data-to-Wisdom Epistemic Architecture for AI (EAAI), a conceptual framework treating Data, Information, Knowledge, Understanding and Wisdom as five distinct but recursively interacting epistemic functions: observation, contextual interpretation, modeling and prediction, critical evaluation and informed judgment. Two observations motivate it: sequential readings of the Knowledge Pyramid obscure how these functions...
-
35
Edge-Gradient Augmentation under Reinforcement Learning for Temporal Graph
Yongqing Zhou, Jiawei Chen, Wenjun Wang, Xinyu Wang, Wei Yu, Qiyao Peng
cs.LG · cs.AI
Temporal link prediction is critical for understand- ing dynamic interactions in evolving networks, informing appli- cations ranging from social network analysis to recommendation systems. Existing methods typically rely on raw temporal graph data, often neglecting explicit edge-change information that can significantly improve predictive performance. To address this gap, we propose edge-gradient reinforced learning (EG-RL), a framework that...
-
36
Manifold Attracting Flow Matching: Physics-Constrained Sampling via Continuous Geometric Dynamics
Guanyu Chen, Jinchen Wang, Shengze Xu, Xingyu Ren, Tieyong Zeng, Antonios Armaou, Dong Ni
cs.LG
Generating scientific data under previously unseen equality constraints requires satisfying the prescribed constraints without destroying the physical structure learned from data. Existing inference-time approaches often enforce physical constraints by repeatedly correcting predicted endpoints, but such discrete corrections can distort learned generative structure and introduce unstable feedback. In this work, we propose Manifold Attracting...
-
37
GNN-Enhanced TCN Algorithms for ECG Anomaly Detection
Angelica Simonetti, Ferdinando Zanchetta
cs.LG
To a time series can be also given the structure of directed graph, opening the doors to the usage of graph neural networks (GNNs) in this context. In this paper we develop a new pipeline involving novel geometric deep learning models that merge the capabilities of ordinary TCNs with the ones of GNNs to solve an anomaly detection problem on electrocardiogram signals.
-
38
Dataset Distillation via Drifting
George Cazenavette, Angelina Quan, Giannis Daras, Antonio Torralba, Vincent Sitzmann
cs.LG · cs.AI · cs.CV
The task of dataset distillation concerns the learning of a small set of synthetic samples that can serve as an effective replacement for a large real dataset when used for downstream tasks. While many solutions have been proposed since the problem's inception, most of them primarily seek to optimize the distilled data with respect to a classifier. In this work, we propose a new distillation method that learns a set of samples such that they...
-
39
Asymmetric Multimodal Learning with Decomposed JEPAs
Yusuf Shihata
cs.LG · cs.CV
Cross-modal alignment is the standard objective for building multimodal representations. We show that alignment as a \emph{sole} objective is lossy wherever modalities are informationally asymmetric---where neither can express what the other carries. Visual question answering is an obvious case: a question is a selective relational operator probing a dense visual field whose fine-grained features no query recovers, so what alignment preserves...
-
40
DySR-GNN: Dynamic Sensor-Relationship Graph Neural Network for Root Cause Analysis in Automotive HiL Multivariate Time Series
Hamza Ouarrad, Mohammad Abboush, Andreas Rausch
cs.LG
Root-cause analysis in automotive Hardware-in-the-Loop (HiL) systems is challenging because faults can propagate across multiple interacting signals, making the most abnormal sensor different from the actual fault origin. This work proposes DySR-GNN, a Dynamic Sensor-Relationship Graph Neural Network for supervised multi-label root-cause identification in multivariate HiL time series. The approach jointly extracts temporal patterns within...
-
41
HALight: Heterogeneous Multi-Agent Reinforcement Learning for Traffic Signal Control under Lane-Level Missing Observations
Chenkai Hu, Yingqian Zhang, Ming Xie, Yi Fang, Chengxiang Jin, Chao Li
cs.LG
Multi-agent reinforcement learning has shown great promise for traffic signal control. However, most existing methods rely on two key assumptions: (i) agents are assumed to be homogeneous with unified observation/action spaces; (ii) full and reliable lane-level traffic observations are available to all agents. In practice, urban road networks exhibit substantial structural heterogeneity (e.g., lane configurations and phase designs), while...
-
42
RESYNC: Residual Latent-State Synchronization for Deadline-Bound Streaming Inference
Eljas Linna, Hao Dong, Juho Kanniainen
cs.LG · cs.AI
Predicting every observation in a high-rate stream requires a model to finish before the next arrival. A large model may offer better predictions but miss that deadline. We propose RESYNC: the large model runs asynchronously, while a fast path updates its latest exact but stale latent state using the observations received since that state was computed and applies its unchanged prediction head. The updater learns from the frozen model's...
-
43
Learning Temporal Dynamics with Derivative Representations for Time Series Forecasting
Shujie Duan, Jia Hong Puah, Haocheng Wu, Yifei Mao, Hongting Zhou, Chenxuan Li, Zitao Fang, Yang Lu, Sim Kuan Goh
cs.LG
Time series forecasting aims to predict the future evolution of complex systems by learning their underlying dynamics from past observations. Despite recent advances in deep learning, Transformers in particular still struggle to capture the temporal evolution of system dynamics. Existing approaches have been largely driven by architectural innovation, with comparatively little attention paid to training strategies that explicitly exploit the...
-
44
Helgason-Fourier Random Features for Scalable Hyperbolic Latent Variables
Koshi Watanabe, Daichi Mochihashi, Hideitsu Hino, Takahiro Ogawa, Miki Haseyama
cs.LG
Hyperbolic geometry is well suited to embedding hierarchical data, capturing tree-like structure faithfully even in the low dimensions where Euclidean dimensionality reduction is severely constrained. We introduce Random Helgason--Fourier Features (RHFF), an explicit finite-dimensional feature map for kernels on hyperbolic space. We characterize the spectral density from which RHFF frequencies are drawn, allowing kernels such as the Matérn...
-
45
A Recipe for Physical Action Model Training
Shaojiang Wang, Geer Yang
cs.LG
Choosing robot model capacity and demonstration budgets also requires decisions about data composition, representation, optimization, and deployment. We connect these decisions through a common conditional-distribution target. Nested learning oracles give exact risk accounting; local information geometry, conditional mutual information, and empirical convergence provide conditional training prescriptions. Known-target numerical checks expose...
-
46
SoScale: Second-Order Scale Search for FP4 Quantization
Pengxiang Zhao, Xing Li, Hui-Ling Zhen, Weizhe Lin, Xianzhi Yu, Zhenhua Dong
cs.LG · cs.AI
Low-bit quantization reduces the memory footprint and inference cost of large language models (LLMs), yet accurate FP4 quantization remains challenging. Practical formats such as MXFP4, NVFP4, and HiF4 pair a 4-bit floating-point payload with block-wise or hierarchical scale codes, making accuracy highly sensitive to how each local distribution is scaled. Default maximum-absolute-value scaling preserves dynamic range, but can spend scarce FP4...
-
47
Experience-Handoff Reinforcement Learning for Recursive Self-Improvement
Doyoung Kim, Byeongguk Jeon, Kailin Zhang, Sherry Yang
cs.LG · cs.CL
It is common to train Large Language Models (LLMs) for agentic tasks with Reinforcement Learning (RL) from scalar rewards. However, this throws away the fine-grained information about the environment that the agent experienced during the trial (e.g., error logs). To utilize this rich information, recent works wrap the agent in a harness that refines its own inputs through reflection or compaction and hands past experience to future rollouts,...
-
48
Recursive State-Augmented Language Models: Persistent Latent-State Adaptation of Pretrained Language Models
Omer Haq
cs.LG · cs.AI · cs.CL · stat.ML
Language models rely on finite token contexts, limiting access to earlier interactions. We introduce Recursive State-Augmented Language Models (Recursive SALM), a framework in which a persistent latent state evolves with each observation and generates adaptations of a pretrained language model. A modular read-transition-write architecture separates observation encoding, latent dynamics, and predictor adaptation. We develop deterministic and...
-
49
Bait: Mitigating Harmful Agentic Reinforcement Learning via Probing Reward Function
Tiansheng Huang, Ananth Balashankar, Virat Shejwalkar, Santosh Pande, Ling Liu
cs.LG · cs.CR
We expose a safety vulnerability on nascent agentic reinforcement-learning-as-a-service. Specifically, with a cost as low as \$20, we show that safety alignment of a large language model (e.g., Inkling) can be completely subverted through rounds of harmful agentic reinforcement learning (agentic RL) with a toxic environment from its users. We also demonstrate that upon a successful harmful agentic RL attack via the service API, the attackers...
-
50
MutAtlas: A PDB-Wide Energy-Guided Atlas of Protein Mutation Effects
Ruihan Guo, Chaoran Cheng, Zhanghan Ni, Neil He, Bangji Yang, Ge Liu
cs.LG · q-bio.GN
Protein mutation effect prediction is fundamental to protein engineering and disease variant interpretation, yet experimentally measured mutation data remain accurate but extremely sparse. To provide scalable supplementary mutation signals, we construct a PDB-wide mutation augmentation dataset that exhaustively enumerates single-site substitutions on experimentally resolved protein structures and aligns mutation signals from physics-based...
-
51
DRAMA: Dissecting Attention Redundancy for Accelerating Multimodal Diffusion Large Language Models
Yi-Hsin Hung, Fangfu Liu, Shao-Yuan Lo, Chu-Song Chen
cs.LG · cs.AI
Multimodal diffusion large language models (dLLMs) generate text through iterative masked denoising with bidirectional attention, offering a compelling alternative to autoregressive vision-language modeling. However, each denoising step requires a full forward pass over long multimodal sequences in which visual tokens dominate the sequence length, and a fixed step budget further compounds this quadratic per-step cost. Existing inference...
-
52
Does Deep Knowledge Tracing Learn Semantic Relations Between Knowledge Components?
Kia Karbasi, Mahdi Saiedi, Kevin Hong, Gregory Pottie
cs.LG · cs.CL
Deep Knowledge Tracing (DKT) is a knowledge tracing (KT) model that updates a shared latent state from sequences of interactions, each pairing a knowledge component (KC) with a response. An interaction with one KC can therefore affect predictions for other KCs. We ask: Are the KC relations learned by DKT semantic? We construct text-derived semantic KC representations using nine pipelines to test whether the conclusions hold across alternative...
-
53
Climate Temporal Temperature Prediction via an Adaptive Stacked Bidirectional Residual Mamba Network
Yongxiang Lei
cs.LG
Accurate temperatures forecasting relies on traditional meteorological parameters that are essential for monitoring weather informatics and guiding forecasting efforts. This study presents Deep DualMamba, a novel deep learning framework for temporal temperature prediction, built upon an enhanced variant of the recently proposed Mamba network. Unlike conventional Transformer-based architectures, which dominate foundation models but suffer from...
-
54
Wanda-Guided Sparse Perturbations for Efficient Zeroth-Order Optimization
Ziqing Wen, Xiaohui Song, Jiahuan Wang, Ping Luo, Junlin Zeng, Dongsheng Li, Tao Sun
cs.LG · cs.AI
Zeroth-order (ZO) optimization enables memory-efficient fine-tuning of large language models through forward-only evaluations, but high-dimensional random perturbations can limit optimization efficiency. Sparse perturbations reduce the search dimension, making the selection of informative coordinates critical. We propose Wanda-Importance Sparsity-MeZO (WIS-MeZO), which uses Wanda weight--activation saliency to construct task-conditioned...
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.