Machine Learning
Compute Where it Counts: Self Optimizing Language Models
Yash Akhauri and Mohamed S. Abdelfattah
Efficient LLM inference research has largely focused on reducing the cost of each decoding step (e.g., using quantization,
pruning, or sparse
attention), typically applying a uniform computation budget to every generated
token. In practice,
token difficulty varies widely, so static compression can over-compute on easy steps and under-compute on hard ones. We study dynamic budget allocation for autoregressive decoding: learning how much computation to spend per
token from within a single model. Self-Optimizing
Language Models (SOL) pair a frozen LLM with a lightweight policy network that reads the LLM hidden state and selects a discrete efficiency action at each decode step. Actions can jointly control (i)
token-level
attention sparsity, (ii) structured activation
pruning in the MLP, and (iii) activation quantization bit-width, while leaving the base model weights unchanged. We train the policy with group-relative policy optimization on teacher-forced episodes: the
token sequence is fixed, while we sample multiple compute schedules (i.e., "counterfactual" schedules that vary only the efficiency actions for the same
token path) and compare their likelihoods under the same supervision. Our reward trades off
language-model quality against soft penalties that encourage episode-average budget usage to match a requested target. Across model variants and compute regimes, SOL improves quality at matched budget over static allocation and strong random schedule search, offering a complementary axis for inference-efficiency optimization. SOL discovers a better quality-efficiency pareto-front across all our experiments and improves MMLU accuracy by up to 7.3% over uniform budget allocation strategies.
Read the full abstract ▾
Machine Learning · Cross-listed from physics.med-ph
Attractor-Vascular Coupling Theory: Formal Grounding and Empirical Validation for AAMI-Standard Cuffless Blood Pressure Estimation from Smartphone Photoplethysmography
Timothy Oladunni and Farouk Ganiyu Adewumi
We present Attractor-Vascular Coupling Theory (AVCT), a theory-guided framework for deriving interpretable blood pressure (BP) predictors from photoplethysmography (PPG) attractor geometry. AVCT connects vascular dynamics, PPG observation, delay-coordinate reconstruction, and attractor descriptors within a unified framework for formulating testable hypotheses, without presupposing the physiological sufficiency of a particular low-dimensional feature set. We evaluate a LightGBM-based
estimator using leave-one-subject-out cross-validation (LOSO-CV) across two complementary experimental settings. Experiment 1 comprises 46 ICU/
surgical subjects from BIDMC and VitalDB, with ECG, PPG, and repeated reference BP measurements. The calibrated full model achieves systolic/diastolic BP (SBP/DBP) mean absolute errors (MAEs) of 2.05/1.67 mmHg, compared with 2.02/1.63 mmHg for the PPG-attractor-only model. Experiment 2 includes 217 ambulatory subjects from the PPG-BP
dataset (651 PPG windows; one cuff-based BP reference per subject) and assesses cross-subject calibrated BP reconstruction rather than within-subject BP tracking. The PPG-attractor model achieves SBP/DBP MAEs of 1.98/0.96 mmHg. Across both cohorts, incorporating ECG-derived features provides no practically meaningful improvement in MAE. Calibration substantially reduces between-subject offset error, while morphological and dynamical descriptors contribute complementary predictive information. Although the reported numerical errors fall within commonly cited thresholds associated with AAMI/ISO and BHS criteria, these findings do not establish compliance with device-validation standards. Collectively, the results support AVCT as a mechanistically motivated representation framework for interpretable PPG-based BP estimation and provide a foundation for prospective validation using repeated reference BP measurements and smartphone-acquired PPG.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CR
BEACON: A Multimodal Dataset for Learning Behavioral Fingerprints from Gameplay Data
Ishpuneet Singh, Gursmeep Kaur, Uday Pratap Singh Atwal et al. (+3 more)
Continuous authentication in high-stakes digital environments requires
datasets with fine-grained behavioral signals under realistic cognitive and motor demands. But current
benchmarks are often limited by small scale, unimodal sensing or lack of synchronised environmental context. To address this gap, this paper introduces BEACON (Behavioral Engine for Authentication & Continuous Monitoring), a large-scale
multimodal dataset that captures diverse skill tiers in competitive Valorant gameplay. BEACON contains approximately 430 GB of synchronised modality data (461 GB total on-disk including auxiliary Valorant configuration captures) from 79 sessions across 28 distinct players, estimated at 102.51 hours of active gameplay, including high-frequency mouse dynamics, keystroke events, network packet captures, screen recordings, hardware metadata, and in-game configuration context. BEACON leverages the high precision motor skills and high cognitive load that are inherent to tactical shooters, making it a rigorous stress test for the robustness of behavioral biometrics. The
dataset allows for the study of continuous authentication, behavioral profiling, user drift and
multimodal representation learning in a high-fidelity esports setting. The authors release the
dataset and code on Hugging Face and GitHub to create a reproducible
benchmark for evaluating next-generation behavioral fingerprinting and security models.
Read the full abstract ▾
All authors Ishpuneet Singh, Gursmeep Kaur, Uday Pratap Singh Atwal, Guramrit Singh, Gurjot Singh, Maninder Singh
Machine Learning · Cross-listed from cs.CV
Masked Generative Transformer Is What You Need for Image Editing
Wei Chow, Linfeng Li, Xian Sun et al. (+14 more)
Diffusion models dominate image editing, yet their global denoising mechanism entangles edited regions with surrounding context, causing modifications to propagate into areas that should remain intact. We propose a fundamentally different approach by leveraging Masked Generative
Transformers (MGTs), whose localized
token-prediction paradigm naturally confines changes to intended regions. We present EditMGT, an MGT-based editing framework that is the first of its kind. Our approach employs multi-layer
attention consolidation to aggregate cross-
attention maps into precise edit localization signals, and region-hold sampling to explicitly prevent
token flipping in non-target areas. To support training, we construct CrispEdit-2M, a 2M-sample high-resolution (>1024) editing
dataset spanning seven categories. With only 960M parameters, EditMGT achieves state-of-the-art image similarity on multiple
benchmarks while delivering 6x faster editing, demonstrating that MGTs offer a compelling alternative to
diffusion-based editing.
Read the full abstract ▾
All authors Wei Chow, Linfeng Li, Xian Sun, Lingdong Kong, Zefeng Li, Qi Xu, Hang Song, Tian Ye, Xian Wang, Jinbin Bai, Shilin Xu, Xiangtai Li, Junting Pan, Shaoteng Liu, Ran Zhou, Tianshu Yang, Songhua Liu
Machine Learning · Cross-listed from cs.AI
Generalizing the Turing Test to Interactive Agents
Daniel Mitropolsky, Riccardo Neumarker, Emanuele Rimoldi et al. (+2 more)
We initiate the study of the Generalized Turing Test (GTT), a formal generalization of Turing's imitation game from humans to arbitrary interactive
agents. For
agents $A$ and $B$, $A$ passes the GTT against $B$ if an instance of $B$, acting as a distinguisher, cannot reliably distinguish an $A$ instructed to imitate $B$ from another instance of $B$; if so, we write $A \geq B$. We study the theoretical and empirical consequences of this idea. On the theory side, we prove sufficient conditions under which this "Turing Comparator" is transitive. We introduce natural variants with querying (the imitator can first interact with a specimen of the target), a Universal Turing Test with arbitrary distinguishers and targets, and complexity-theoretic variants that control interaction length. As a proof of concept, we evaluate the GTT and its variants across nine large
language models. Remarkably, Turing Scores recover a clear model stratification consistent with standard external
benchmarks despite being derived entirely from pairwise imitation games. Transcript analysis reveals that models use both stylistic signatures and substantive STEM and logic-based probes. Together, these results suggest indistinguishability could provide a meaningful signal for comparing
agents, yielding an inherently adaptive form of evaluation that does not rely on fixed
benchmarks.
Read the full abstract ▾
All authors Daniel Mitropolsky, Riccardo Neumarker, Emanuele Rimoldi, Susan S. Hong, Tomaso Poggio
Machine Learning
Conditional anomaly detection methods for patient-management alert systems
Michal Valko, Gregory Cooper, Amy Seybert et al. (+3 more)
Anomaly detection methods can be very useful in identifying unusual or interesting patterns in data. A recently proposed conditional anomaly detection framework extends anomaly detection to the problem of identifying anomalous patterns on a subset of attributes in the data. The anomaly always depends (is conditioned) on the value of remaining attributes. The work presented in this paper focuses on instance-based methods for detecting conditional anomalies. The methods rely on the distance metric to identify examples in the
dataset that are most critical for detecting the anomaly. We investigate various metrics and metric learning methods to optimize the performance of the instance-based anomaly detection methods. We show the benefits of the instance-based methods on two real-world detection problems: detection of unusual admission decisions for patients with the community-acquired pneumonia and detection of unusual orders of an HPF4 test that is used to confirm Heparin induced thrombocytopenia - a life-threatening condition caused by the Heparin therapy.
Read the full abstract ▾
All authors Michal Valko, Gregory Cooper, Amy Seybert, Shyam Visweswaran, Melissa Saul, Miloš Hauskrecht
Machine Learning
Clin-JEPA: A Multi-Phase Co-Training Framework for Joint-Embedding Predictive Pretraining on EHR Patient Trajectories
Yixuan Yang, Mehak Arora, Ryan Zhang et al. (+10 more)
Joint-
embedding predictive architectures (JEPA) learn representations by predicting in latent space, as in computer
vision; retaining the action-conditioned predictor at inference turns them into latent world models, enabling planning in robotics (V-JEPA 2-AC). Bringing this design to EHR patient trajectories—a predictor that simulates a patient's trajectory in latent space—has not been explored. We use an LLM as the encoder, reading the hourly record as text, avoiding feature engineering and vocabulary harmonisation. But an LLM adapted by supervised fine-tuning does not organise its latent space around physiological dynamics, and freezing it to train the predictor, as in V-JEPA 2-AC, leaves the encoder unaware of the rollout signal: the predictor degrades under rollout. We instead co-train encoder and predictor under one latent-prediction objective, grounding the encoder in the dynamics its predictor must follow. Naïve co-training, however, is unstable: the untrained predictor drags the encoder toward collapse, and the predictor's rollout diverges as its target space moves. We present Clin-JEPA, a five-phase curriculum that stably co-trains an LLM encoder with a latent trajectory predictor on MIMIC-IV. Three evaluations support the design: (1) under 48-hour autoregressive rollout the co-trained predictor degrades least (predictor degradation $\times$1.06, against $\times$1.23–1.36 for two-stage designs and $\times$6.3–66 for curriculum ablations) while the co-trained encoder resolves the progression of patient state most sharply (largest state displacement); (2) the co-trained encoder separates deteriorating from stable patients in its latent space with Cohen's $d{=}1.59$, against $\leq$0.50 for two-stage encoders; (3) one set of
embeddings serves 34 downstream tasks across three
benchmarks, outperforming strong per-task tuned baselines and a pretrained EHR foundation model.
Read the full abstract ▾
All authors Yixuan Yang, Mehak Arora, Ryan Zhang, Baraa Abed, Junseob Kim, Tilendra Choudhary, Md Hassanuzzaman, Kevin Zhu, Ayman Ali, Chengkun Yang, Alasdair Edward Gent, Victor Moas, Rishikesan Kamaleswaran
Machine Learning · Cross-listed from cs.CV
Transcoda: End-to-End Zero-Shot Optical Music Recognition via Data-Centric Synthetic Training
Daniel Dratschuk and Paul Swoboda
Optical Music Recognition (OMR), the task of transcribing sheet music into a structured textual representation, is currently bottlenecked by a lack of large-scale, annotated
datasets of real scans. This forces models to rely on either few-shot transfer or synthetic training pipelines that remain overly simplistic. A secondary challenge is encoding non-uniqueness: in the popular Humdrum **kern format for transcribing music, multiple different text encodings can render into the same visual sheet music. This one-to-many mapping creates a harder learning task and introduces high uncertainty during decoding. We propose Transcoda, an OMR system built on (i) an advanced synthetic data generation pipeline, (ii) a normalization of the **kern encoding to enforce a unique normal form and (iii) grammar-based decoding to ensure the syntactic correctness of the output. This approach allows us to train a compact 59M-parameter model in just 6 hours on a single GPU that outperforms billion-parameter baselines. Transcoda achieves the best score among state of the art baselines on a newly curated
benchmark of synthetically rendered scores at 18.46% OMR-NED (compared to 43.91% for the next-best system, Legato) and reduces the error rate on historical Polish scans to 63.97% OMR-NED (down from 80.16% for SMT++).
Read the full abstract ▾
Machine Learning
SLIM: Sparse Latent Steering for Interpretable and Property-Directed LLM-Based Molecular Editing
Mingxu Zhang, Yuhan Li, Lujundong Li et al. (+3 more)
Large
language models possess strong chemical
reasoning capabilities, making them effective molecular editors. However, property-relevant information is implicitly entangled across their dense hidden states, providing no explicit handle for property control: a substantial fraction of edits fail to improve or even degrade target properties. To address these issues, we propose SLIM (Sparse Latent Interpretable Molecular editing), a plug-and-play framework that decomposes the editor's hidden states into sparse, property-aligned features via a Sparse Autoencoder with learnable importance gates. Steering in this sparse feature space precisely activates property-relevant dimensions, improving editing success rate without modifying model parameters. The same sparse basis further supports interpretable analysis of editing behavior. Experiments on the MolEditRL
benchmark across four model architectures and eight molecular properties show consistent gains over baselines, with improvements of up to 42.4 points.
Read the full abstract ▾
All authors Mingxu Zhang, Yuhan Li, Lujundong Li, Dazhong Shen, Hui Xiong, Ying Sun
Machine Learning · Cross-listed from cs.CV
Predicting 3D structure by latent posterior sampling
Azmi Haider and Dan Rosenbaum
The remarkable achievements of both generative models of 2D images and neural field representations for 3D scenes present a compelling opportunity to integrate the strengths of both approaches. In this work, we propose a methodology that combines a NeRF-based representation of 3D scenes with probabilistic modeling and
reasoning using
diffusion models. We view 3D reconstruction as a perception problem with inherent uncertainty that can thereby benefit from probabilistic inference methods. The core idea is to represent the 3D scene as a
stochastic latent variable for which we can learn a prior and use it to perform posterior inference given a set of observations. We formulate posterior sampling using the score-based inference method of
diffusion models in conjunction with a likelihood term computed from a reconstruction model that includes volumetric rendering. We train the model using a two-stage process: first we train the reconstruction model while auto-decoding the latent representations for a
dataset of 3D scenes, and then we train the prior over the latents using a
diffusion model. By using the model to generate samples from the posterior we demonstrate that various 3D reconstruction tasks can be performed, differing by the type of observation used as inputs. We showcase reconstruction from single-view, multi-view, noisy images, sparse pixels, and sparse depth data. These observations vary in the amount of information they provide for the scene and we show that our method can model the varying levels of inherent uncertainty associated with each task. Our experiments illustrate that this approach yields a comprehensive method capable of accurately predicting 3D structure from diverse types of observations.
Read the full abstract ▾
Machine Learning
NoRIN: Backbone-Adaptive Reversible Normalization for Time-Series Forecasting
Shun Zhang and Yuyang Xiao
Reversible instance normalization (RevIN) and its successors (Dish-TS, SAN, FAN) have become the de facto plug-in for time-series
forecasting, yet the map they apply to each data point is strictly affine, $x \mapsto ax+b$, so they cannot reshape the underlying distribution – heavy tails remain heavy and skewness remains uncorrected. We propose NoRIN, a non-linear reversible normalization based on the arcsinh-form Johnson $S_U$ transform with two shape parameters $(\delta,\varepsilon)$ that control tailedness and skewness; the linear $Z$-score used by RevIN is recovered only in the limit $\delta \to \infty$. Training $(\delta,\varepsilon)$ jointly with the backbone via gradient descent reliably pushes them toward this linear limit within a few epochs – a phenomenon we name the degeneration problem: the
forecasting loss is locally indifferent to shape, and the high-capacity backbone compensates for any monotone reparameterization of its input. NoRIN escapes the degeneration by decoupling shape selection from gradient training: $(\delta,\varepsilon)$ are initialized by a closed-form Slifker-Shapiro
quantile fit and refined by
Bayesian optimization on the validation objective, while the inner training loop is identical to standard RevIN-style training. Across six representative backbones x five real-world
datasets x three prediction horizons (90 configurations), decoupled shape optimization recovers $(\delta^\star,\varepsilon^\star)$ that sit systematically far from the linear limit, with values that vary in a backbone-dependent way. This empirically supports the central thesis: different backbones genuinely require different normalization parameters to reach their best performance.
Read the full abstract ▾
Machine Learning
Benchmarking Sensor-Fault Robustness in Forecasting
Alexander Windmann, Philipp Wittenberg, Gianluca Manca et al. (+3 more)
Cyber-physical system (CPS)
forecasting models depend on sensor streams with noisy, biased, missing, or temporally misaligned readings, yet standard
forecasting evaluation often selects models by nominal error without showing whether they remain robust under such faults. We introduce SensorFault-Bench, a CPS-grounded protocol that evaluates
forecasting architectures and robustness-improvement methods through controlled synthetic stress tests of isolated input faults, together with an operational taxonomy of the methods. Across four real-world
datasets and eight scored scenarios governed by a standardized severity model, it reports worst-scenario degradation, clean mean squared error (MSE), and worst-scenario fault-time MSE, separating relative robustness from absolute error. Explicit fault-training methods train on adjacent fault families disjoint from the scored scenarios. Empirically,
forecasting architectures favored by clean MSE can degrade sharply under faults, and clean-MSE rankings can disagree with worst-scenario fault-time error rankings. Chronos-2, the evaluated zero-shot foundation model, matches or trails the last-value naive forecaster in clean MSE on the two single-target
datasets and has the largest worst-scenario degradation on ETTh1 and Traffic, where all channels are forecast targets. For the evaluated robustness-improvement method set, paired deltas show selective degradation reductions: projected gradient descent
adversarial training and randomized training lead where value faults dominate observed degradation, and fault augmentation leads on Traffic, where availability faults dominate. High degradation reveals fault sensitivity that clean MSE misses, while low degradation supports robustness only within the evaluated scenarios. The
benchmark therefore screens forecasters before use in a specific deployment, which requires validation against its recorded faults.
Read the full abstract ▾
All authors Alexander Windmann, Philipp Wittenberg, Gianluca Manca, Marcel Dix, Jens U. Brandt, Oliver Niggemann
Machine Learning · Cross-listed from cs.AI
MaD Physics: Evaluating information seeking under constraints in physical environments
Moksh Jain, Mehdi Bennani, Johannes Bausch et al. (+4 more)
Scientific discovery is fundamentally a resource-constrained process that requires navigating complex trade-offs between the quality and quantity of measurements due to physical and cost constraints. Measurements drive the scientific process by revealing novel phenomena to improve our understanding. Existing
benchmarks for evaluating
agents for scientific discovery focus on either static knowledge-based
reasoning or unconstrained experimental design tasks, and do not capture the ability to make measurements and plan under constraints. To bridge this gap, we propose Measuring and Discovering Physics (MaD Physics), a
benchmark to evaluate the ability of
agents to make informative measurements and conclusions subject to constraints on the quality and quantity of measurements. The
benchmark consists of three environments, each based on a distinct physical law. To mitigate contamination from existing knowledge, MaD Physics includes altered physical laws. In each trial, the
agent makes measurements of the system until it exhausts an allotted budget and then the
agent has to infer the underlying physical law to make predictions about the state of the system in the future. MaD Physics evaluates two fundamental capabilities of scientific
agents: inferring models from data and planning under constraints. We also demonstrate how MaD Physics can be used to evaluate other capabilities such as multimodality and in-context learning. We
benchmark agents on MaD Physics using four Gemini models (2.5 Flash Lite, 2.5 Flash, 2.5 Pro, and 3 Flash), identifying shortcomings in their structured exploration and data collection capabilities and highlighting directions to improve their scientific
reasoning.
Read the full abstract ▾
All authors Moksh Jain, Mehdi Bennani, Johannes Bausch, Yuri Chervonyi, Bogdan Georgiev, Simon Osindero, Nenad Tomašev
Machine Learning
On periodic distributed representations using Fourier embeddings
Jakeb Chouinard
Periodic signals are critical for representing physical and perceptual phenomena. Scalar, real angular measures, e.g., radians and degrees, result in difficulty processing and distinguishing nearby angles, especially when their absolute difference exceeds pi. We can avoid this problem by using real-valued, periodic
embeddings in high-dimensional space. These representations also allow us to control the nature of their dot product similarities, allowing us to construct a variety of different kernel shapes. In this work, we aim of highlight how these representations can be constructed and focus on the formalization of Dirichlet and periodic Gaussian kernels using the neurally-plausible representation scheme of Spatial Semantic Pointers.
Read the full abstract ▾
Machine Learning
Policy Gradient Methods for Non-Markovian Reinforcement Learning
Avik Kar, Siddharth Chandak, Rahul Singh et al. (+4 more)
We study policy gradient methods for reinforcement learning in non-Markovian decision processes (NMDPs), where observations and rewards depend on the entire interaction history. To handle this dependence, the
agent maintains an internal state that is recursively updated to provide a compact summary of past observations and actions. In contrast to approaches that treat the
agent state dynamics as fixed or learn it via predictive objectives, we propose a reward-centric formulation that jointly optimizes the
agent state dynamics and the control policy to maximize the expected cumulative reward. To this end, we consider a class of
Agent State-Markov (ASM) policies, comprising an
agent state dynamics and a control policy that maps the
agent state to actions. We establish a novel policy gradient
theorem for ASM policies, extending the classical policy gradient results from the Markovian setting to episodic and infinite-horizon discounted NMDPs. Building on this gradient expression, we propose the
Agent State-Markov Policy Gradient (ASMPG)
algorithm, which leverages the recursive structure of the
agent state dynamics for efficient optimization. We establish finite-time and almost sure convergence guarantees, and empirically demonstrate that, on a range of non-Markovian tasks, ASMPG outperforms baselines that learn state representations via predictive objectives.
Read the full abstract ▾
All authors Avik Kar, Siddharth Chandak, Rahul Singh, Soumitra Sinhahajari, Eric Moulines, Shalabh Bhatnagar, Nicholas Bambos
Machine Learning
Likelihood scoring for continuations of mathematical text: a self-supervised benchmark with tests for shortcut vulnerabilities
Daniel Ranard
We introduce an automatically generated
benchmark for predicting hidden text in technical papers. A paper supplies visible context $X$ and a hidden continuation $Y$; the evaluated model writes an auxiliary forecast string $Z$, and a separate scorer assigns next-
token probability to $Y$ both with and without conditioning on $Z$. This gives a label-free test of whether $Z$ transmits information about the continuation, compared against controls where $Z$ is recent context rather than a forecast. Our main testbed is equation-suffix prediction: the predictor sees context and the first part of a displayed equation, then forecasts the rest. The task mixes surface-level arXiv/TeX text modeling with
reasoning-sensitive inference; the suffix is one of many roughly equivalent continuations, so the
benchmark is read statistically rather than item-by-item. On 1363 equation continuations from 138 recent physics and mathematics papers, forecasts from GPT-5.5, Opus 4.7, and GPT-5.4 nano all improve clipped likelihood over the context control under both Qwen3-8B and Kimi K2.6 scorers, distinguishing model families and
reasoning-effort settings without human labels. To emulate shortcuts where $Z$ further
primes the scorer rather than making a useful forecast, we also fine-tune the scorer on context-only
prompts and apply it to held-out papers as a stronger control. GPT-5.5 forecasts still beat this fine-tuned control; GPT-5.4 nano forecasts do not. Longer prose/TeX continuations show positive but noisier lift over controls, concentrated near the beginning of the target. These results support cross-model likelihood scoring as a static
benchmark and as a setup for probing shortcut vulnerabilities before reinforcement learning or model-selection optimization is applied.
Read the full abstract ▾
Machine Learning
Mistake-Bounded Language Generation
Jon Kleinberg, Charlotte Peale and Omer Reingold
We investigate the learning task of
language generation in the limit, but shift focus from the traditional time-of-last-mistake metric of a generator's success to a new notion of "mistake-bounded generation." While existing results for
language generation in the limit focus on guaranteeing eventual consistency, they are blind to the cumulative error incurred during the learning process. We address this by shifting the goal to minimizing the total number of invalid elements output by a generation
algorithm. We establish a formal reduction to the Learning from Correct Demonstrations framework of Joshi et al. (2025), enabling a general recipe for deriving mistake bounds via weighted update rules. For finite classes, we provide an
algorithm that simultaneously achieves an optimal last-mistake time of $\mathsf{Cdim}(L)$ and a mistake bound of $\lfloor \log_2 |L| \rfloor$, whereas for the non-uniform setting of countably infinite streams of
languages, we prove a fundamental trade-off: achieving logarithmic mistakes $O(\log i)$ necessarily precludes convergence guarantees established in prior work. Finally, we show that our framework can be extended to accommodate noisy adversaries and guarantee mistake bounds that scale with the adversary's suboptimality.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CR
LLMs for Secure Hardware Design and Related Problems: Opportunities and Challenges
Johann Knechtel, Ozgur Sinanoglu and Ramesh Karri
The integration of Large
Language Models (LLMs) into Electronic Design Automation (EDA) and hardware security is rapidly reshaping the
semiconductor industry. While LLMs offer unprecedented capabilities in generating Register Transfer Level (RTL) code, automating testbenches, and bridging the semantic gap between high-level specifications and silicon, they simultaneously introduce severe vulnerabilities. This comprehensive review provides an in-depth analysis of the state-of-the-art in LLM-driven hardware design, organized around key advancements in EDA synthesis, hardware trust, design for security, and education. We systematically expand on the methodologies of recent breakthroughs – from
reasoning-driven synthesis and multi-
agent vulnerability extraction to data contamination and
adversarial machine learning (ML) evasion. We integrate general discussions on critical countermeasures, such as dynamic benchmarking to combat data memorization and aggressive red-teaming for robust security assessment. Finally, we synthesize cross-cutting lessons learned to guide future research toward secure, trustworthy, and autonomous design ecosystems.
Read the full abstract ▾
Machine Learning · Cross-listed from cs.CV
PhyGround: Benchmarking Physical Reasoning in Generative World Models
Juyi Lin, Arash Akbari, Yumei He et al. (+13 more)
Generative world models are increasingly used for
video generation, where learned simulators are expected to capture the physical rules that govern real-world dynamics. However, evaluating whether generated
videos actually follow these rules remains challenging. Existing physics-focused
video benchmarks have made important progress, but they still face three key challenges, including the coarse evaluation frameworks that hide law-specific failures, response biases and fatigue that undermine the validity of annotation judgments, and automated evaluators that are insufficiently physics-aware or difficult to audit. To address those challenges, we introduce PhyGround, a criteria-grounded
benchmark for evaluating physical
reasoning in
video generation. The
benchmark contains 250 curated
prompts, each augmented with an expected physical outcome, and a taxonomy of 13 physical laws across solid-body mechanics, fluid dynamics, and optics. Each law is operationalized through observable sub-questions to enable per-law diagnostics. We evaluate eight modern
video generation models through a large-scale, quality-controlled human study, grounded on social science lab experiment design. A total of 459 annotators provided 5,796 complete annotations and over 37.4K fine-grained labels; after quality control, the retained annotations exhibited high split-half model-ranking correlations (Spearman's rho > 0.90). To support reproducible automated evaluation, we release PhyJudge-9B, an open physics-specialized VLM judge. PhyJudge-9B achieves substantially lower aggregate relative bias than Gemini-3.1-Pro (3.3% vs. 16.6%). We release
prompts, human annotations, model checkpoints, and evaluation code on the project page https://phyground.github.io/.
Read the full abstract ▾
All authors Juyi Lin, Arash Akbari, Yumei He, Lin Zhao, Haichao Zhang, Arman Akbari, Xingchen Xu, Zoe Y. Lu, Enfu Nan, Hokin Deng, Edmund Yeh, Sarah Ostadabbas, Yun Fu, Jennifer Dy, Pu Zhao, Yanzhi Wang
Machine Learning
The Last Word Often Wins: A Format Confound in Chain-of-Thought Corruption Studies
Gabriel Garcia
Corruption studies, the standard tool for evaluating chain-of-thought (CoT) faithfulness, infer which steps are “computationally important” from accuracy loss when steps are corrupted. We show that when
benchmark chains end with an explicit terminal answer line, as in GSM8K and MATH, these tests largely measure answer placement rather than where intermediate computation is carried out. Using matched GSM8K examples, removing only the final answer statement while preserving all
reasoning collapses suffix sensitivity by about $19\times$ for Qwen~2.5-3B ($N{=}300$, $p{=}0.022$). Conflicting-answer
prompts, which contain correct
reasoning but a wrong explicit final answer, drive accuracy to zero or near-zero at 7B across five open-weight model families; wrong-answer following is strong at 3B–7B and attenuates sharply at larger scales. Replications on MATH, within-stable comparisons at 7B, and suffix-free chains show the same pattern in different guises: corruption sensitivity tracks the location of explicit answer text, not a fixed computational depth in the
reasoning. Generation-time probes indicate that final answers are rarely early-determined during generation (${<}5\%$ early commitment), yet consumption-time behavior systematically follows explicit answer text. The confound is therefore largely a readout effect when the chain is consumed. We propose a three-prerequisite protocol (question-only control, format characterization, and an all-position sweep) as a practical minimum for future corruption-based faithfulness studies.
Read the full abstract ▾
Machine Learning
Muown: Row-Norm Control for Muon Optimization
Kai Lion, Florian Hübler, Bingcong Li et al. (+2 more)
Muon has emerged as a strong competitor to AdamW for
language model pre-training, yet its behavior at scale is sensitive to weight decay. Recent work has observed that, for
Muon without decoupled weight decay, the spectral norm of weight
matrices drifts upward over training. Through a decomposition of the spectral norm into a row-magnitude factor and a row-coherence factor, we identify the former as the empirical driver of this drift under
Muon, while the latter remains well-behaved along the trajectory. Motivated by this diagnosis, we introduce Muown, a drop-in replacement for
Muon that treats the row-magnitude vector as an explicit optimizer variable, updating it under the $\ell_\infty$ geometry induced by the decomposition, while applying
Muon unchanged to the remaining direction component. We prove that Muown attains the optimal non-convex rates in both deterministic and
stochastic regimes under a dual norm aligned with the underlying geometries and with a
stochastic noise coefficient that empirically remains below that of
Muon throughout training. Across GPT-style pre-training on FineWeb-Edu with model sizes from 124M up to 2.7B parameters, Muown improves perplexity over
Muon, SOAP, AdamW, and Lion. It also widens the plateau of near-optimal learning rates across model scales, reduces sensitivity to weight decay, and avoids the spectral norm drift at negligible step-time overhead when appropriately sharded.
Read the full abstract ▾
All authors Kai Lion, Florian Hübler, Bingcong Li, Antonio Orvieto, Niao He
Machine Learning · Cross-listed from stat.ML
Factual recall in linear associative memories: sharp asymptotics and mechanistic insights
Alessio Giorlandino, Sebastian Goldt and Antoine Maillard
Large
language models excel at factual recall, yet the fundamental limits of storing input–output associations with neural networks remain unclear. We study these limits in a minimal setting: a linear associative
memory that maps $p$ input
embeddings in $\mathbb{R}^d$ to their $d$-dimensional targets via a linear layer, requiring each mapped input to score its own target above all other outputs. Since all inputs compete for the same outputs, these constraints are strongly correlated, which makes the storage capacity $p_c$ — the largest number of associations that can be stored — hard to characterise. We address this in three steps. First, we introduce a decoupled model, in which each input competes with its own independent set of outputs, and provide numerical and analytical evidence that it is equivalent to the original model, both in capacity and in the structure of the learnt solution. Second, using tools from statistical physics, we show that the decoupled model has capacity $p_c \log p_c / d^2 = 1/2$, extend this computation to linear two-layer networks, and go beyond the capacity by predicting the spectrum of the learnt weights and the storage mechanism. The latter explains how the optimal solution outperforms a naïve Hebbian rule: rather than boosting input–output
alignments with broad fluctuations, it raises the correct scores just above the extreme-value threshold set by the competing outputs. Third, we prove the decoupled model's capacity thresholds rigorously, including in the non-convex two-layer case; combined with a recent proof for the original model, this shows that both models have the same capacity for full-rank weights. These findings give a sharp characterisation of factual storage in linear networks, and suggest decoupling as a general tool for memorisation and learning problems in which each input must be matched to one of an extensive number of outputs.
Read the full abstract ▾
Machine Learning
ConQuR: Corner Aligned Activation Quantization via Optimized Rotations for LLMs
Chayne Thrash, Ali Abbasi and Soheil Kolouri
Large
language models (LLMs) are costly to deploy due to their large
memory footprint and high inference cost. Weight-activation quantization can reduce these costs, but low-bit activation quantization remains difficult because activation outliers induce large quantization error. Recent rotation-based methods address this by applying orthogonal transformations that redistribute activation magnitude across dimensions, but existing approaches either require expensive end-to-end rotation training or rely on stored activation corpora, introducing significant compute or storage overhead. We propose a lightweight post-training rotation calibration method for LLM activation quantization. Our method learns orthogonal rotations that align normalized activations with the corners of an inscribed hypercube, encouraging activation energy to be distributed more evenly across dimensions. This objective admits an efficient closed-form update via the orthogonal Procrustes problem, avoiding gradient-based optimization over the orthogonal group. We further introduce an online calibration procedure that updates rotations as calibration samples are processed, eliminating the need to store activations on disk and allowing rotations to adapt to quantized activation distributions during calibration. Experiments on Llama-2 and Llama-3 models from 3B to 70B parameters show that our method achieves competitive or improved performance across perplexity
benchmarks and common sense
reasoning tasks while avoiding both costly end-to-end training and large offline activation storage.
Read the full abstract ▾
Machine Learning · Cross-listed from math.OC
Implicit Neural Optimal Transport via Fixed-Point Optimization
Yesom Park, Eric Gelphman, Stanley Osher et al. (+1 more)
We propose an implicit neural formulation of optimal transport that eliminates
adversarial min–max optimization and multi-network architectures commonly used in existing approaches. Our key idea is to parameterize a single potential in the Kantorovich dual and reformulate the associated c-transform as a proximal fixed-point problem. This yields a stable single-network framework in which dual feasibility is enforced exactly through proximal optimality conditions rather than
adversarial training. Despite the inner fixed-point computation, gradients can be computed without differentiating through the fixed-point iterations, enabling efficient training without requiring implicit differentiation. We further establish convergence of
stochastic gradient descent. The resulting framework is efficient, scalable, and broadly applicable: it simultaneously recovers forward and backward transport maps and naturally extends to class-conditional settings. Experiments on high-dimensional Gaussian
benchmarks, physical
datasets, and image translation tasks demonstrate strong transport accuracy together with improved training stability and favorable computational and
memory efficiency.
Read the full abstract ▾
All authors Yesom Park, Eric Gelphman, Stanley Osher, Samy Wu Fung
Machine Learning
Elucidating Representation Degradation Problem in Diffusion Model Training
Zhipeng Yao, Dazhou Li, Zitong Zhang et al. (+6 more)
Diffusion models have achieved remarkable success, yet their training remains inefficient due to a severe optimization bottleneck, which we term Representation Degradation. As noise levels increase, the outputs of the trained model exhibit progressive structural distortion, which can destabilize training and impair generation quality. Our analysis suggests that this instability is driven by mismatched target recoverability, which is associated with Neural Tangent Kernel (NTK) spectral weakening and effective low-rank behavior. To address this, we propose Elucidated Representation
Diffusion (ERD), a plug-and-play framework that dynamically reallocates optimization effort according to effective recoverability. By stabilizing representation learning without external supervision, ERD accelerates convergence and achieves strong empirical performance across
diffusion backbones.
Read the full abstract ▾
All authors Zhipeng Yao, Dazhou Li, Zitong Zhang, Durude Mahee, Fan Zhu, Wenbin Zhang, Xinwei He, Yeying Jin, Rui Yu
Machine Learning
MASS-DPO: Multi-negative Active Sample Selection for Direct Policy Optimization
Rohan Surana, Xintong Li, Sheldon Yu et al. (+7 more)
Multi-negative preference optimization under the Plackett–Luce (PL) model extends Direct Preference Optimization (DPO) by leveraging comparative signals across one preferred and multiple rejected responses. However, optimizing over large negative pools is costly, and many candidates contribute redundant gradients due to their similar effects on policy updates. We introduce MASS-DPO, a multi-negative active sample selection method that derives a PL-specific Fisher-information objective for selecting compact, informative negative subsets within each
prompt. The resulting log-determinant objective selects negatives that contribute complementary information for policy updates, yielding compact subsets that retain the full pool's information while reducing redundancy. In practice, this favors negatives whose gradients cover different update directions, reducing redundant signal from near-duplicate candidates while preserving the most useful training information. Across four
benchmarks spanning recommendation and multiple-choice QA and three model families, MASS-DPO consistently exceeds or matches existing methods in accuracy, improves Recall/NDCG and margin-based optimization dynamics, and delivers stronger
alignment with substantially fewer negatives.
Read the full abstract ▾
All authors Rohan Surana, Xintong Li, Sheldon Yu, Yiran Jenny Shen, Chuhan Wang, Tong Yu, Prithviraj Ammanabrolu, Jingbo Shang, Julian McAuley, Junda Wu
Machine Learning
Rebellious Student: Reversing Teacher Signals for Reasoning Exploration with Self-Distilled RLVR
Jeonghye Kim, Jiwon Jeon, Dongsheng Li et al. (+1 more)
Self-
distillation has emerged as a powerful framework for post-training LLMs, where a teacher conditioned on extra information guides a student without it, both from the same model. While this guidance is useful when the student has failed, on successful rollouts, the same mechanism instead overwrites the student's choices and suppresses it's own
reasoning. Therefore, we propose reading the original self-
distillation signal in reverse: when the student succeeds along a path the teacher would not have predicted, these
tokens reflect its self-driven
reasoning. Building on this, we propose RLRT (RLVR with Reversed Teacher), which augments GRPO by reinforcing these
tokens on correct rollouts. We interpret this as a new form of exploration in RLVR: not uniform diversity, but valuable exploration grounded in the student's own success. Across base, instruction-tuned, and thinking-tuned Qwen3 checkpoints, RLRT substantially outperforms self-
distillation and exploration-based baselines, establishing information asymmetry as a new, principled design axis for RLVR.
Read the full abstract ▾
All authors Jeonghye Kim, Jiwon Jeon, Dongsheng Li, Yuqing Yang
Machine Learning
Locking Pretrained Weights via Deep Low-Rank Residual Distillation
Keitaro Sakamoto, Pierre Ablin, Federico Danieli et al. (+1 more)
The quality of open-weight
language models has dramatically improved in recent years. Sharing weights greatly facilitates model adoption by enabling their use across diverse hardware and software platforms. They also allow for more open research and testing, to the extent that users can use them as checkpoints, fine-tune them according to their needs, and potentially redistribute them. In some cases, however, concerns on modifying these weights towards unauthorized uses may outweigh the pros of giving users such freedom. Defending against such adaptation is non-trivial: since an adaptive attacker can observe all weights and architectures by definition, they can reverse simple structural defenses, and use optimization to defeat the simplest locking mechanisms. In this work, we exploit the inference-training asymmetry of automatic differentiation as a novel defense axis. We propose DLR-Lock, a method where the purveyor of the model purposely replaces each pretrained MLP in their model with a deep low-rank residual network (DLR-Net) of comparable parameter count, forcing activation
memory that grows linearly with depth during backpropagation. DLR-Nets are efficiently trained via module-wise regression. We show that, beyond this
memory overhead, DLR-Lock results in architectural mismatches that complicate the optimization landscape of standard fine-tuning, and a backward pass that incurs disproportionately more overhead than the forward pass. Our defense succeeds in withstanding adaptive attackers with full knowledge of the defense strategy while preserving the original model's capabilities. Experiments on LLMs validate these claims.
Read the full abstract ▾
All authors Keitaro Sakamoto, Pierre Ablin, Federico Danieli, Marco Cuturi
Machine Learning · Cross-listed from math.OC
On the global convergence of gradient flow for wide shallow models beyond homogeneous nonlinearities
Romain Petit, Clarice Poon and Gabriel Peyré
A surprising phenomenon in the training of neural networks is the ability of gradient descent to find global minimizers of the training loss despite its non-convexity. Following earlier work, we investigate this behavior for wide shallow models. Existing global convergence results primarily concern models with positively one-homogeneous nonlinearities, such as ReLU activations, and models with scalar output weights and bounded nonlinearities, such as sigmoid activations. We study a broader class of models, including multi-head
attention layers and two-layer networks with bounded or asymptotically positively one-homogeneous activations and vector output weights. Building upon [Chizat and Bach, 2018], we prove that, in the limit of many hidden
neurons or
attention heads, non-global minimizers of the training loss are unstable under mean-field gradient flow dynamics by constructing "escape regions" in the parameter space. Our global convergence statements are conditional in the following sense: if the mean-field gradient flow converges in W2, then its limit must be a global minimizer. We revisit the bounded nonlinearity, scalar-output setting of [CB18], giving an escape region construction adapted to unbounded nonlinear parameter domains. We also propose new constructions for nonlinearities with at most linear growth under a non-degeneracy assumption and for asymptotically positively one-homogeneous nonlinearities. Finally, we show the well-posedness and stability estimates for the mean-field training dynamics under sub-Gaussian initializations.
Read the full abstract ▾
Machine Learning
DynaMiCS: Fine-tuning LLMs with Performance Constraints using Dynamic Mixtures
Eleonora Gualdoni, Sonia Laguna, Louis Bethune et al. (+3 more)
Multi-domain fine-tuning of large
language models requires improving performance on target domains while preserving previous capabilities, such as general knowledge, instruction following, or safety evaluations. Existing data mixing strategies rely on fixed heuristics or adaptive rules that cannot explicitly enforce preservation of such capabilities. We propose DynaMiCS, a dynamic mixture optimizer that casts multi-domain fine-tuning as a constrained optimization problem. At each update, DynaMiCS performs short domain-specific probing runs to estimate a slope
matrix of local cross-domain effects, capturing how training on each fine-tuning
dataset affects each evaluation domain. These estimates are then used to compute mixture weights through optimization over the probability simplex, with the objective of improving target-domain performance while keeping constrained-domain metrics within a specified tolerance of reference levels. Because these effects are measured by finite differences rather than gradients, targets and constraints need not be differentiable, or present in the fine-tuning data, and can be specified directly as
benchmark accuracies. Across scenarios with varying numbers of target and constrained domains, and with loss- or accuracy-based objectives, DynaMiCS achieves stronger target-domain improvements and higher constraint satisfaction than static, dynamic, similarity-based and probing-based alternatives, without a reference model, per-example scoring, or manually tuned weights.
Read the full abstract ▾
All authors Eleonora Gualdoni, Sonia Laguna, Louis Bethune, Joao Monteiro, Pierre Ablin, Marco Cuturi