Tootfinder

This https://arxiv.org/abs/2212.02658 has been replaced.
link: https://scholar.google.com/scholar?q=a

Pooling information in likelihood-free inference
Likelihood-free inference (LFI) methods, such as approximate Bayesian computation, have become commonplace for conducting inference in complex models. Many approaches are based on summary statistics or discrepancies derived from synthetic data. However, determining which summary statistics or discrepancies to use for constructing the posterior remains a challenging question, both practically and theoretically. Instead of relying on a single vector of summaries for inference, we propose a new po…

This https://arxiv.org/abs/2505.18570 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

VISTA: Vision-Language Inference for Training-Free Stock Time-Series Analysis
Stock price prediction remains a complex and high-stakes task in financial analysis, traditionally addressed using statistical models or, more recently, language models. In this work, we introduce VISTA (Vision-Language Inference for Stock Time-series Analysis), a novel, training-free framework that leverages Vision-Language Models (VLMs) for multi-modal stock forecasting. VISTA prompts a VLM with both textual representations of historical stock prices and their corresponding line charts to pre…

AMD says it has acquired the team behind AI inference chip developer Untether AI, a day after announcing it acquired AI software optimization startup Brium (Dylan Martin/CRN)
https://www.crn.com/news/components-periph

Exclusive: AMD Acquires Team Behind AI Chip Startup Untether AI
AMD confirmed on Thursday to CRN that it has acquired the employees behind AI chip startup Untether AI.

Membership Inference Attacks on Sequence Models
Lorenzo Rossi, Michael Aerni, Jie Zhang, Florian Tram\`er
https://arxiv.org/abs/2506.05126 https://

Membership Inference Attacks on Sequence Models
Sequence models, such as Large Language Models (LLMs) and autoregressive image generators, have a tendency to memorize and inadvertently leak sensitive information. While this tendency has critical legal implications, existing tools are insufficient to audit the resulting risks. We hypothesize that those tools' shortcomings are due to mismatched assumptions. Thus, we argue that effectively measuring privacy leakage in sequence models requires leveraging the correlations inherent in sequential g…

FlowSpec: Continuous Pipelined Speculative Decoding for Efficient Distributed LLM Inference
Xing Liu, Lizhuo Luo, Ming Tang, Chao Huang
https://arxiv.org/abs/2507.02620

FlowSpec: Continuous Pipelined Speculative Decoding for Efficient Distributed LLM Inference
Distributed inference serves as a promising approach to enabling the inference of large language models (LLMs) at the network edge. It distributes the inference process to multiple devices to ensure that the LLMs can fit into the device memory. Recent pipeline-based approaches have the potential to parallelize communication and computation, which helps reduce inference latency. However, the benefit diminishes when the inference request at the network edge is sparse, where pipeline is typically …

Deep learning inference with the #EventHorizonTelescope I. Calibration improvements and a comprehensive synthetic data library / II. The ZINGULARITY framework for Bayesian artificial neural networks / III. ZINGULARITY results from the 2017 observations and predictions for future array expansions: https://www.aanda.org/articles/aa/full_html/2025/06/aa53784-25/aa53784-25.html / https://www.aanda.org/articles/aa/full_html/2025/06/aa53785-25/aa53785-25.html / https://www.aanda.org/articles/aa/full_html/2025/06/aa53786-25/aa53786-25.html -> Self-learning neural network cracks iconic black holes: https://www.astronomie.nl/nieuws/en/self-learning-neural-network-cracks-iconic-black-holes-4528

Interesting, letting AI models cooperate.
#ai #sakana

Sakana AI
Inference-Time Scaling and Collective Intelligence for Frontier AI

VeFIA: An Efficient Inference Auditing Framework for Vertical Federated Collaborative Software
Chung-ju Huang, Ziqi Zhang, Yinggui Wang, Binghui Wang, Tao Wei, Leye Wang
https://arxiv.org/abs/2507.02376

VeFIA: An Efficient Inference Auditing Framework for Vertical Federated Collaborative Software
Vertical Federated Learning (VFL) is a distributed AI software deployment mechanism for cross-silo collaboration without accessing participants' data. However, existing VFL work lacks a mechanism to audit the execution correctness of the inference software of the data party. To address this problem, we design a Vertical Federated Inference Auditing (VeFIA) framework. VeFIA helps the task party to audit whether the data party's inference software is executed as expected during large-scale infere…

Lipschitz stability for Bayesian inference in porous medium tissue growth models
Tomasz D\k{e}biec, Piotr Gwiazda, B{\l}a\.zej Miasojedow, Katarzyna Ryszewska, Zuzanna Szyma\'nska, Aneta Wr\'oblewska-Kami\'nska
https://arxiv.org/abs/2506.04769

Lipschitz stability for Bayesian inference in porous medium tissue growth models
We consider a macroscopic model for the dynamics of living tissues incorporating pressure-driven dispersal and pressure-modulated proliferation. Given a power-law constitutive relation between the pressure and cell density, the model can be written as a porous medium equation with a growth term. We prove Lipschitz continuity of the mild solutions of the model with respect to the diffusion parameter (the exponent $γ$ in the pressure-density law) in the $L_1$ norm. While of independent analytica…

Classification of Extremal Dependence in Financial Markets via Bootstrap Inference
Qian Hui, Sidney I. Resnick, Tiandong Wang
https://arxiv.org/abs/2506.04656

Classification of Extremal Dependence in Financial Markets via Bootstrap Inference
Accurately identifying the extremal dependence structure in multivariate heavy-tailed data is a fundamental yet challenging task, particularly in financial applications. Following a recently proposed bootstrap-based testing procedure, we apply the methodology to absolute log returns of U.S. S&P 500 and Chinese A-share stocks over a time period well before the U.S. election in 2024. The procedure reveals more isolated clustering of dependent assets in the U.S. economy compared with China which e…

Differentiable Fuzzy Cosmic-Web for Field Level Inference
P. Rossell\'o, F. -S. Kitaura, D. Forero-S\'anchez, F. Sinigaglia, G. Favole
https://arxiv.org/abs/2506.03969

Differentiable Fuzzy Cosmic-Web for Field Level Inference
A comprehensive analysis of the cosmological large-scale structure derived from galaxy surveys involves field-level inference, which requires a forward modelling framework that simultaneously accounts for structure formation and tracer bias. While structure formation models are well-understood, the development of an effective field-level bias model remains imprecise, particularly in the context of tracer perturbation theory within Bayesian reconstruction methods, which we address in this work. …

Tokyo-based Sakana AI details a new Monte Carlo tree search-based technique that lets multiple LLMs cooperate on a single task, outperforming individual models (Ben Dickson/VentureBeat)
https://venturebeat.com/ai/sakana-ais-

Sakana AI’s TreeQuest: Deploy multi-model teams that outperform individual LLMs by 30%
Sakana AI's new inference-time scaling technique uses Monte-Carlo Tree Search to orchestrate multiple LLMs to collaborate on complex tasks.

Active inference as a unified model of collision avoidance behavior in human drivers
Julian F. Schumann, Johan Engstroem, Leif Johnson, Matthew O'Kelly, Joao Messias, Jens Kober, Arkady Zgonnikov
https://arxiv.org/abs/2506.02215

Active inference as a unified model of collision avoidance behavior in human drivers
Collision avoidance -- involving a rapid threat detection and quick execution of the appropriate evasive maneuver -- is a critical aspect of driving. However, existing models of human collision avoidance behavior are fragmented, focusing on specific scenarios or only describing certain aspects of the avoidance behavior, such as response times. This paper addresses these gaps by proposing a novel computational cognitive model of human collision avoidance behavior based on active inference. Activ…

This https://arxiv.org/abs/2505.24502 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_qu…

Predictability of quantum observables: Applications in QKD and top quarks
We quantify and optimize the predictability of local observables in bipartite quantum systems by employing the Bayes risk and the inference variance, two measures rooted in statistical learning theory. Specifically, we minimized these quantities when the prediction is improved by an additional quantum system, providing analytical expressions for arbitrary two-qubit states, and showcasing a connection with Einstein-Podolsky-Rosen steering criteria. Then, we embed our Bayes risk minimization into…

Amortized variational transdimensional inference
Laurence Davies, Dan Mackinlay, Rafael Oliveira, Scott A. Sisson
https://arxiv.org/abs/2506.04749 https://…

Amortized variational transdimensional inference
The expressiveness of flow-based models combined with stochastic variational inference (SVI) has, in recent years, expanded the application of optimization-based Bayesian inference to include problems with complex data relationships. However, until now, SVI using flow-based models has been limited to problems of fixed dimension. We introduce CoSMIC, normalizing flows (COntextually-Specified Masking for Identity-mapped Components), an extension to neural autoregressive conditional normalizing fl…

Reconstructing North Korea's Plutonium Production History with Bayesian Inference-Based Reprocessing Waste Analysis
Benjamin Jung, Johannes Bosse, Malte G\"ottsche
https://arxiv.org/abs/2506.03865

Reconstructing North Korea's Plutonium Production History with Bayesian Inference-Based Reprocessing Waste Analysis
Although North Korea's nuclear program has been the subject of extensive scrutiny, estimates of its fissile material stockpiles remain fraught with uncertainty. In potential future disarmament agreements, inspectors may need to use nuclear archaeology methods to verify or gain confidence in a North Korean fissile material declaration. This study explores the potential utility of a Bayesian inference-based analysis of the isotopic composition of reprocessing waste to reconstruct the operating hi…

Inverse design for robust inference in integrated computational spectrometry
Wenchao Ma, Rapha\"el Pestourie, Zin Lin, Steven G. Johnson
https://arxiv.org/abs/2506.02194

Inverse design for robust inference in integrated computational spectrometry
For computational spectrometers, we propose an inverse-design approach in which the scattering media are topology-optimized to achieve better performance in inference, without the need of a training set of spectra and a distribution of detector noise. Our approach also allows the selection of the inference algorithm to be decoupled from that of the scatterer. For smooth spectra, we additionally devise a regularized reconstruction algorithm based on Chebyshev interpolation, which yields higher a…

This https://arxiv.org/abs/2506.02814 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDC_…

Adaptive Configuration Selection for Multi-Model Inference Pipelines in Edge Computing
The growing demand for real-time processing tasks is driving the need for multi-model inference pipelines on edge devices. However, cost-effectively deploying these pipelines while optimizing Quality of Service (QoS) and costs poses significant challenges. Existing solutions often neglect device resource constraints, focusing mainly on inference accuracy and cost efficiency. To address this, we develop a framework for configuring multi-model inference pipelines. Specifically: 1) We model the de…

Bayesian Doubly Robust Causal Inference via Posterior Coupling
Shunichiro Orihara, Tomotaka Momozaki, Shonosuke Sugasawa
https://arxiv.org/abs/2506.04868 h…

Bayesian Doubly Robust Causal Inference via Posterior Coupling
In observational studies, propensity score methods are central for estimating causal effects while adjusting for confounders. Among them, the doubly robust (DR) estimator has gained considerable attention because it provides consistent estimates when either the propensity score model or the outcome model is correctly specified. Like other propensity score approaches, the DR estimator typically involves two-step estimation: first, estimating the propensity score and outcome models, and then esti…

Generator Based Inference (GBI)
Chi Lung Cheng, Ranit Das, Runze Li, Radha Mastandrea, Vinicius Mikuni, Benjamin Nachman, David Shih, Gup Singh
https://arxiv.org/abs/2506.00119

Generator Based Inference (GBI)
Statistical inference in physics is often based on samples from a generator (sometimes referred to as a ``forward model") that emulate experimental data and depend on parameters of the underlying theory. Modern machine learning has supercharged this workflow to enable high-dimensional and unbinned analyses to utilize much more information than ever before. We propose a general framework for describing the integration of machine learning with generators called Generator Based Inference (GBI). A …

This https://arxiv.org/abs/2505.15380 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSD_…

Accelerating Autoregressive Speech Synthesis Inference With Speech Speculative Decoding
Modern autoregressive speech synthesis models leveraging language models have demonstrated remarkable performance. However, the sequential nature of next token prediction in these models leads to significant latency, hindering their deployment in scenarios where inference speed is critical. In this work, we propose Speech Speculative Decoding (SSD), a novel framework for autoregressive speech synthesis acceleration. Specifically, our method employs a lightweight draft model to generate candidat…

This https://arxiv.org/abs/2504.01759 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_ees…

Alpha-Beta HMM: Hidden Markov Model Filtering with Equal Exit Probabilities and a Step-Size Parameter
The hidden Markov model (HMM) provides a powerful framework for inference in time-varying environments, where the underlying state evolves according to a Markov chain. To address the optimal filtering problem in general dynamic settings, we propose the $αβ$-HMM algorithm, which simplifies the state transition model to a Markov chain with equal exit probabilities and introduces a step-size parameter to balance the influence of observational data and the model. By analyzing the algorithm's dyna…

This https://arxiv.org/abs/2505.23655 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Keyed Chaotic Tensor Transformations for Secure And Attributable Neural Inference
This work introduces a novel framework for secure and privacy-preserving neural network inference based on keyed chaotic dynamical transformations. The proposed method applies a deterministic, cryptographically seeded chaotic system to tensors, producing non-invertible, user-specific transformations that enable authenticated inference, tensor-level watermarking, and data attribution. This framework offers a scalable and lightweight alternative to conventional cryptographic techniques, and estab…

Exact operator inference with minimal data
Henrik Rosenberger, Benjamin Sanderse, Giovanni Stabile
https://arxiv.org/abs/2506.01244 https://

Exact operator inference with minimal data
This work introduces a novel method to generate snapshot data for operator inference that guarantees the exact reconstruction of intrusive projection-based reduced-order models (ROMs). To ensure exact reconstruction, the operator inference least squares matrix must have full rank, without regularization. Existing works have achieved this full rank using heuristic strategies to generate snapshot data and a-posteriori checks on full rank, but without a guarantee of success. Our novel snapshot dat…

This https://arxiv.org/abs/2502.00805 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_…

Constraints on Lorentz Invariance Violation from Gamma-ray Burst rest-frame spectral lags using Profile Likelihood
We reanalyze the spectral lag data for 56 Gamma-Ray Bursts (GRBs) in the cosmological rest frame to search for Lorentz Invariance Violation (LIV) using frequentist inference. For this purpose, we use the technique of profile likelihood to deal with the nuisance parameters, corresponding to a constant time lag in the GRB rest frame and an unknown intrinsic scatter, while the parameter of interest is the energy scale for LIV ($E_{QG}$). With this method, we do not obtain a global minimum for $χ^…

This https://arxiv.org/abs/2503.18617 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_…

Scaling Laws for Emulation of Stellar Spectra
Neural network-based emulators for the inference of stellar parameters and elemental abundances represent an increasingly popular methodology in modern spectroscopic surveys. However, these approaches are often constrained by their emulation precision and domain transfer capabilities. Greater generalizability has previously been achieved only with significantly larger model architectures, as demonstrated by Transformer-based models in natural language processing. This observation aligns with ne…

This https://arxiv.org/abs/2409.14202 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_eco…

Mining Causality: AI-Assisted Search for Instrumental Variables
The instrumental variables (IVs) method is a leading empirical strategy for causal inference. Finding IVs is a heuristic and creative process, and justifying its validity -- especially exclusion restrictions -- is largely rhetorical. We propose using large language models (LLMs) to search for new IVs through narratives and counterfactual reasoning, similar to how a human researcher would. The stark difference, however, is that LLMs can dramatically accelerate this process and explore an extreme…

This https://arxiv.org/abs/2504.10667 has been replaced.
link: https://scholar.google.com/scholar?q=a

Beyond Coordinates: Meta-Equivariance in Statistical Inference
Optimal statistical decisions should transcend the language used to describe them. Yet, how do we guarantee that the choice of coordinates - the parameterisation of an optimisation problem - does not subtly dictate the solution? This paper reveals a fundamental geometric invariance principle. We first analyse the optimal combination of two asymptotically normal estimators under a strictly convex trace-AMSE risk. While methods for finding optimal weights are known, we prove that the resulting op…

This https://arxiv.org/abs/2505.24293 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Large Language Models are Locally Linear Mappings
We demonstrate that the inference operations of several open-weight large language models (LLMs) can be mapped to an exactly equivalent linear system for an input sequence without modifying the model weights or altering output predictions. Extending techniques from image diffusion models that exhibit local or piecewise linearity, we strategically alter the gradient computation with respect to a given input sequence for a next-token prediction such that the Jacobian of the model nearly exactly r…

This https://arxiv.org/abs/2505.19931 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_ees…

Accelerating Flow-Matching-Based Text-to-Speech via Empirically Pruned Step Sampling
Flow-matching-based text-to-speech (TTS) models, such as Voicebox, E2 TTS, and F5-TTS, have attracted significant attention in recent years. These models require multiple sampling steps to reconstruct speech from noise, making inference speed a key challenge. Reducing the number of sampling steps can greatly improve inference efficiency. To this end, we introduce Fast F5-TTS, a training-free approach to accelerate the inference of flow-matching-based TTS models. By inspecting the sampling traje…

Reappraising the Elatina series: Solar dynamo clocking and inference of orbital periods
F. Stefani, T. Weier, G. M. Horstmann, G. Mamatsashvili
https://arxiv.org/abs/2506.02628

Reappraising the Elatina series: Solar dynamo clocking and inference of orbital periods
We reconsider the 680 million year old Elatina series of sedimentary laminae from South Australia that show a remarkably stable periodicity with a main period of around 12 years, which is close to the Schwabe cycle, and a second period of 314 years that has been coined Elatina cycle. By analyzing the residuals of the series' minima from a linear trend, and deriving Dicke's ratio, we first show that the series exhibits a high degree of phase stability, except one single break point which may ind…

Breaking the HBM Bit Cost Barrier: Domain-Specific ECC for AI Inference Infrastructure
Rui Xie, Asad Ul Haq, Yunhua Fang, Linsen Ma, Sanchari Sen, Swagath Venkataramani, Liu Liu, Tong Zhang
https://arxiv.org/abs/2507.02654

Breaking the HBM Bit Cost Barrier: Domain-Specific ECC for AI Inference Infrastructure
High-Bandwidth Memory (HBM) delivers exceptional bandwidth and energy efficiency for AI workloads, but its high cost per bit, driven in part by stringent on-die reliability requirements, poses a growing barrier to scalable deployment. This work explores a system-level approach to cost reduction by eliminating on-die ECC and shifting all fault management to the memory controller. We introduce a domain-specific ECC framework combining large-codeword Reed--Solomon~(RS) correction with lightweight …

The Spurious Factor Dilemma: Robust Inference in Heavy-Tailed Elliptical Factor Models
Jiang Hu, Jiahui Xie, Yangchun Zhang, Wang Zhou
https://arxiv.org/abs/2506.05116

The Spurious Factor Dilemma: Robust Inference in Heavy-Tailed Elliptical Factor Models
Factor models are essential tools for analyzing high-dimensional data, particularly in economics and finance. However, standard methods for determining the number of factors often overestimate the true number when data exhibit heavy-tailed randomness, misinterpreting noise-induced outliers as genuine factors. This paper addresses this challenge within the framework of Elliptical Factor Models (EFM), which accommodate both heavy tails and potential non-linear dependencies common in real-world da…

This https://arxiv.org/abs/2411.13280 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_qbi…

Empower Structure-Based Molecule Optimization with Gradient Guided Bayesian Flow Networks
Structure-Based molecule optimization (SBMO) aims to optimize molecules with both continuous coordinates and discrete types against protein targets. A promising direction is to exert gradient guidance on generative models given its remarkable success in images, but it is challenging to guide discrete data and risks inconsistencies between modalities. To this end, we leverage a continuous and differentiable space derived through Bayesian inference, presenting Molecule Joint Optimization (MolJO),…

Pyrefly vs. ty: Comparing Python’s Two New Rust-Based Type Checkers
#types

This https://arxiv.org/abs/2504.15268 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_qfi…

Beating the Correlation Breakdown: Robust Inference, Flexible Scenarios, and Stress Testing for Financial Portfolios
We live in a multivariate world, and effective modeling of financial portfolios, including their construction, allocation, forecasting, and risk analysis, simply is not possible without explicitly modeling the dependence structure of their assets. Dependence structure can drive portfolio results more than the combined effects of other parameters in investment and risk models, but the literature provides relatively little to define the finite-sample distributions of dependence measures in useabl…

EARN: Efficient Inference Acceleration for LLM-based Generative Recommendation by Register Tokens
Chaoqun Yang, Xinyu Lin, Wenjie Wang, Yongqi Li, Teng Sun, Xianjing Han, Tat-Seng Chua
https://arxiv.org/abs/2507.00715

EARN: Efficient Inference Acceleration for LLM-based Generative Recommendation by Register Tokens
Large Language Model-based generative recommendation (LLMRec) has achieved notable success, but it suffers from high inference latency due to massive computational overhead and memory pressure of KV Cache. Existing KV Cache reduction methods face critical limitations: cache compression offers marginal acceleration given recommendation tasks' short decoding steps, while prompt compression risks discarding vital interaction history. Through systematic analysis of attention patterns in LLMRec, we …

Evaluation of "As-Intended" Vehicle Dynamics using the Active Inference Framework
Kazuharu Kidera, Takuma Miyaguchi, Hideyoshi Yanagisawa
https://arxiv.org/abs/2506.00035

Evaluation of "As-Intended" Vehicle Dynamics using the Active Inference Framework
We constructed a computational model of the driver's brain for steering tasks using the active inference framework, grounded in the free energy principle - a theory from computational neuroscience. This model enables quantitative estimation of how accurately the brain learns vehicle dynamics and performs appropriate steering, using a measure called variational free energy. Through driving simulator experiments, we observed strong correlations between variational free energy and both expert driv…

Dissecting the Impact of Mobile DVFS Governors on LLM Inference Performance and Energy Efficiency
Zongpu Zhang, Pranab Dash, Y. Charlie Hu, Qiang Xu, Jian Li, Haibing Guan
https://arxiv.org/abs/2507.02135

Dissecting the Impact of Mobile DVFS Governors on LLM Inference Performance and Energy Efficiency
Large Language Models (LLMs) are increasingly being integrated into various applications and services running on billions of mobile devices. However, deploying LLMs on resource-limited mobile devices faces a significant challenge due to their high demand for computation, memory, and ultimately energy. While current LLM frameworks for mobile use three power-hungry components-CPU, GPU, and Memory-even when running primarily-GPU LLM models, optimized DVFS governors for CPU, GPU, and memory feature…

#AI #honeypots huh

The /llms.txt file – llms-txt
A proposal to standardise on using an /llms.txt file to provide information to help LLMs use a website at inference time.

#AI #honeypots huh

The /llms.txt file – llms-txt
A proposal to standardise on using an /llms.txt file to provide information to help LLMs use a website at inference time.

QuickSilver -- Speeding up LLM Inference through Dynamic Token Halting, KV Skipping, Contextual Token Fusion, and Adaptive Matryoshka Quantization
Danush Khanna, Aditya Kumar Guru, Srivarshinee Sridhar, Zidan Ahmed, Rubhav Bahirwani, Meetu Malhotra, Vinija Jain, Aman Chadha, Amitava Das, Kripabandhu Ghosh
https://arxiv.org/abs/2…

QuickSilver -- Speeding up LLM Inference through Dynamic Token Halting, KV Skipping, Contextual Token Fusion, and Adaptive Matryoshka Quantization
Inference accounts for the majority of latency and energy consumption in large language model (LLM) deployments, often exceeding 90% of total cost. While training-time efficiency has seen extensive progress, runtime optimization remains a key bottleneck, particularly under autoregressive decoding. Existing approaches -- such as pruning, quantization, early exits, and speculative decoding -- often require retraining, architectural changes, or disrupt decoding compatibility. We introduce QuickSil…

This https://arxiv.org/abs/2502.03956 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLO_…

POPACheck: A Model Checker for Probabilistic Pushdown Automata
We present POPACheck, the first model checking tool for probabilistic Pushdown Automata (pPDA) supporting temporal logic specifications. POPACheck provides a user-friendly probabilistic modeling language with recursion that automatically translates into Probabilistic Operator Precedence Automata (pOPA). pOPA are a class of pPDA that can express all the behaviors of probabilistic programs: sampling, conditioning, recursive procedures, and nested inference queries. On pOPA, POPACheck can solve re…

Adaptive Configuration Selection for Multi-Model Inference Pipelines in Edge Computing
Jinhao Sheng, Zhiqing Tang, Jianxiong Guo, Tian Wang
https://arxiv.org/abs/2506.02814

Adaptive Configuration Selection for Multi-Model Inference Pipelines in Edge Computing
The growing demand for real-time processing tasks is driving the need for multi-model inference pipelines on edge devices. However, cost-effectively deploying these pipelines while optimizing Quality of Service (QoS) and costs poses significant challenges. Existing solutions often neglect device resource constraints, focusing mainly on inference accuracy and cost efficiency. To address this, we develop a framework for configuring multi-model inference pipelines. Specifically: 1) We model the de…

This https://arxiv.org/abs/2505.24765 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_qu…

Supervised Quantum Machine Learning: A Future Outlook from Qubits to Enterprise Applications
Supervised Quantum Machine Learning (QML) represents an intersection of quantum computing and classical machine learning, aiming to use quantum resources to support model training and inference. This paper reviews recent developments in supervised QML, focusing on methods such as variational quantum circuits, quantum neural networks, and quantum kernel methods, along with hybrid quantum-classical workflows. We examine recent experimental studies that show partial indications of quantum advantag…

A Survey of LLM Inference Systems
James Pan, Guoliang Li
https://arxiv.org/abs/2506.21901 https://arxiv.org/pdf/2506.21901

A Survey of LLM Inference Systems
The past few years has witnessed specialized large language model (LLM) inference systems, such as vLLM, SGLang, Mooncake, and DeepFlow, alongside rapid LLM adoption via services like ChatGPT. Driving these system design efforts is the unique autoregressive nature of LLM request processing, motivating new techniques for achieving high performance while preserving high inference quality over high-volume and high-velocity workloads. While many of these techniques are discussed across the literatu…

This https://arxiv.org/abs/2405.00295 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csGT_…

Proof of Sampling: A Nash Equilibrium-Based Verification Protocol for Decentralized Systems
This paper introduces the Proof of Sampling (PoSP) protocol, a Nash Equilibrium-based verification mechanism, and its application to decentralized machine learning inference through spML. Our protocol has a pure strategy Nash Equilibrium, compelling rational participants to act honestly. It economically disincentivizes dishonest behavior, making it costly for participants to compromise the network's integrity. In our spML protocol, we apply PoSP to decentralized inference for AI applications vi…

Everybody complaining about getting hammered with #AI traffic seems to think that these are crawlers scraping for training data.
How likely is it that this is a complete misconception and this is all inference time?
Most public companies give their cralwers and RAG agents different user agent strings. But what about security services trawling through their data?

Privacy Leaks by Adversaries: Adversarial Iterations for Membership Inference Attack
Jing Xue, Zhishen Sun, Haishan Ye, Luo Luo, Xiangyu Chang, Ivor Tsang, Guang Dai
https://arxiv.org/abs/2506.02711

Privacy Leaks by Adversaries: Adversarial Iterations for Membership Inference Attack
Membership inference attack (MIA) has become one of the most widely used and effective methods for evaluating the privacy risks of machine learning models. These attacks aim to determine whether a specific sample is part of the model's training set by analyzing the model's output. While traditional membership inference attacks focus on leveraging the model's posterior output, such as confidence on the target sample, we propose IMIA, a novel attack strategy that utilizes the process of generatin…

LLM-Driven Treatment Effect Estimation Under Inference Time Text Confounding
Yuchen Ma, Dennis Frauen, Jonas Schweisthal, Stefan Feuerriegel
https://arxiv.org/abs/2507.02843

LLM-Driven Treatment Effect Estimation Under Inference Time Text Confounding
Estimating treatment effects is crucial for personalized decision-making in medicine, but this task faces unique challenges in clinical practice. At training time, models for estimating treatment effects are typically trained on well-structured medical datasets that contain detailed patient information. However, at inference time, predictions are often made using textual descriptions (e.g., descriptions with self-reported symptoms), which are incomplete representations of the original patient i…

Neural simulation-based inference of the Higgs trilinear self-coupling via off-shell Higgs production
Aishik Ghosh, Maximilian Griese, Ulrich Haisch, Tae Hyoun Park
https://arxiv.org/abs/2507.02032

Neural simulation-based inference of the Higgs trilinear self-coupling via off-shell Higgs production
One of the forthcoming major challenges in particle physics is the experimental determination of the Higgs trilinear self-coupling. While efforts have largely focused on on-shell double- and single-Higgs production in proton-proton collisions, off-shell Higgs production has also been proposed as a valuable complementary probe. In this article, we design a hybrid neural simulation-based inference (NSBI) approach to construct a likelihood of the Higgs signal incorporating modifications from the S…

Variational Inference for Latent Variable Models in High Dimensions
Chenyang Zhong, Sumit Mukherjee, Bodhisattva Sen
https://arxiv.org/abs/2506.01893 https…

Variational Inference for Latent Variable Models in High Dimensions
Variational inference (VI) is a popular method for approximating intractable posterior distributions in Bayesian inference and probabilistic machine learning. In this paper, we introduce a general framework for quantifying the statistical accuracy of mean-field variational inference (MFVI) for posterior approximation in Bayesian latent variable models with categorical local latent variables. Utilizing our general framework, we capture the exact asymptotic regime where MFVI `works' for the celeb…

Simulation-Efficient Cosmological Inference with Multi-Fidelity SBI
Leander Thiele, Adrian E. Bayer, Naoya Takeishi
https://arxiv.org/abs/2507.00514 https:…

Simulation-Efficient Cosmological Inference with Multi-Fidelity SBI
The simulation cost for cosmological simulation-based inference can be decreased by combining simulation sets of varying fidelity. We propose an approach to such multi-fidelity inference based on feature matching and knowledge distillation. Our method results in improved posterior quality, particularly for small simulation budgets and difficult inference problems.

Parallel CPU-GPU Execution for LLM Inference on Constrained GPUs
Jiakun Fan, Yanglin Zhang, Xiangchen Li, Dimitrios S. Nikolopoulos
https://arxiv.org/abs/2506.03296

Parallel CPU-GPU Execution for LLM Inference on Constrained GPUs
Deploying large language models (LLMs) for online inference is often constrained by limited GPU memory, particularly due to the growing KV cache during auto-regressive decoding. Hybrid GPU-CPU execution has emerged as a promising solution by offloading KV cache management and parts of attention computation to the CPU. However, a key bottleneck remains: existing schedulers fail to effectively overlap CPU-offloaded tasks with GPU execution during the latency-critical, bandwidth-bound decode phase…

Grapheme-Coherent Phonemic and Prosodic Annotation of Speech by Implicit and Explicit Grapheme Conditioning
Hien Ohnaka, Yuma Shirahata, Byeongseon Park, Ryuichi Yamamoto
https://arxiv.org/abs/2506.04527

Grapheme-Coherent Phonemic and Prosodic Annotation of Speech by Implicit and Explicit Grapheme Conditioning
We propose a model to obtain phonemic and prosodic labels of speech that are coherent with graphemes. Unlike previous methods that simply fine-tune a pre-trained ASR model with the labels, the proposed model conditions the label generation on corresponding graphemes by two methods: 1) Add implicit grapheme conditioning through prompt encoder using pre-trained BERT features. 2) Explicitly prune the label hypotheses inconsistent with the grapheme during inference. These methods enable obtaining p…

This https://arxiv.org/abs/2505.14884 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Polar Sparsity: High Throughput Batched LLM Inferencing with Scalable Contextual Sparsity
Accelerating large language model (LLM) inference is critical for real-world deployments requiring high throughput and low latency. Contextual sparsity, where each token dynamically activates only a small subset of the model parameters, shows promise but does not scale to large batch sizes due to union of active neurons quickly approaching dense computation. We introduce Polar Sparsity, highlighting a key shift in sparsity importance from MLP to Attention layers as we scale batch size and seque…

Information-Optimal Sensing and Control in High-Intensity Laser Experiments
A. D\"opp, C. Eberle, J. Esslinger, S. Howard, F. Irshad, J. Schroeder, N. Weisse, S. Karsch
https://arxiv.org/abs/2506.04946

Information-Optimal Sensing and Control in High-Intensity Laser Experiments
High-intensity laser systems present unique measurement and optimization challenges due to their high complexity, low repetition rates, and shot-to-shot variations. We discuss recent developments towards a unified framework based on information theory and Bayesian inference that addresses these challenges. Starting from fundamental constraints on the physical field structure, we recently demonstrated how to capture complete spatio-temporal information about individual petawatt laser pulses. Bui…

This https://arxiv.org/abs/2405.05969 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_…

Learned harmonic mean estimation of the Bayesian evidence with normalizing flows
We present the learned harmonic mean estimator with normalizing flows - a robust, scalable and flexible estimator of the Bayesian evidence for model comparison. Since the estimator is agnostic to sampling strategy and simply requires posterior samples, it can be applied to compute the evidence using any Markov chain Monte Carlo (MCMC) sampling technique, including saved down MCMC chains, or any variational inference approach. The learned harmonic mean estimator was recently introduced, where ma…

CLONE: Customizing LLMs for Efficient Latency-Aware Inference at the Edge
Chunlin Tian, Xinpeng Qin, Kahou Tam, Li Li, Zijian Wang, Yuanzhe Zhao, Minglei Zhang, Chengzhong Xu
https://arxiv.org/abs/2506.02847

CLONE: Customizing LLMs for Efficient Latency-Aware Inference at the Edge
Deploying large language models (LLMs) on edge devices is crucial for delivering fast responses and ensuring data privacy. However, the limited storage, weight, and power of edge devices make it difficult to deploy LLM-powered applications. These devices must balance latency requirements with energy consumption and model accuracy. In this paper, we first quantify the challenges of deploying LLMs on off-the-shelf edge devices and then we present CLONE, an in-depth algorithm-hardware co-design at…

Combining Type Inference and Automated Unit Test Generation for Python
Lukas Krodinger, Stephan Lukasczyk, Gordon Fraser
https://arxiv.org/abs/2507.01477 h…

Combining Type Inference and Automated Unit Test Generation for Python
Automated unit test generation is an established research field that has so far focused on statically-typed programming languages. The lack of type information in dynamically-typed programming languages, such as Python, inhibits test generators, which heavily rely on information about parameter and return types of functions to select suitable arguments when constructing test cases. Since automated test generators inherently rely on frequent execution of candidate tests, we make use of these fre…

This https://arxiv.org/abs/2506.01969 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDC_…

FlashMLA-ETAP: Efficient Transpose Attention Pipeline for Accelerating MLA Inference on NVIDIA H20 GPUs
Efficient inference of Multi-Head Latent Attention (MLA) is challenged by deploying the DeepSeek-R1 671B model on a single Multi-GPU server. This paper introduces FlashMLA-ETAP, a novel framework that enhances MLA inference for the single-instance deployment scenario on NVIDIA H20 GPUs. We propose the Efficient Transpose Attention Pipeline (ETAP), which reconfigures attention computation through transposition to align the KV context length with the \(M\)-dimension in WGMMA operations, significa…

Randomization Inference with Sample Attrition
Xinran Li, Peizan Sheng, Zeyang Yu
https://arxiv.org/abs/2507.00795 https://arxiv.org/p…

Randomization Inference with Sample Attrition
Although appealing, randomization inference for treatment effects can suffer from severe size distortion due to sample attrition. We propose new, computationally efficient methods for randomization inference that remain valid under a range of potentially informative missingness mechanisms. We begin by constructing valid p-values for testing sharp null hypotheses, using the worst-case p-value from the Fisher randomization test over all possible imputations of missing outcomes. Leveraging distrib…

Towards Better Attribute Inference Vulnerability Measures
Paul Francis, David Wagner
https://arxiv.org/abs/2507.01710 https://arxiv.o…

Towards Better Attribute Inference Vulnerability Measures
The purpose of anonymizing structured data is to protect the privacy of individuals in the data while retaining the statistical properties of the data. An important class of attack on anonymized data is attribute inference, where an attacker infers the value of an unknown attribute of a target individual given knowledge of one or more known attributes. A major limitation of recent attribute inference measures is that they do not take recall into account, only precision. It is often the case tha…

This https://arxiv.org/abs/2505.07802 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csRO_…

Improving Trajectory Stitching with Flow Models
Generative models have shown great promise as trajectory planners, given their affinity to modeling complex distributions and guidable inference process. Previous works have successfully applied these in the context of robotic manipulation but perform poorly when the required solution does not exist as a complete trajectory within the training set. We identify that this is a result of being unable to plan via stitching, and subsequently address the architectural and dataset choices needed to re…

Quantum Bayesian inference with Suport vector states for intrusion detection
Nayema Mridha, Garrv Sipani, Eva R Gaarder, Shah Haque, Radhika Kuttala, Binay P Akhouri, Mohamad M Al Zein, Eric Howard
https://arxiv.org/abs/2507.00403

Quantum Bayesian inference with Suport vector states for intrusion detection
We present a quantum Bayesian inference method for intrusion detection, using explicitly constructed quantum circuits and statevector simulation. Prior and conditional probabilities are encoded via unitary gates, and posterior distributions are extracted through symbolic post-selection. Applied to a scenario with network spikes, system vulnerabilities, and false alarms, the method yields joint, marginal, and conditional probabilities aligned with causal structure. Our results demonstrate the fe…

This https://arxiv.org/abs/2406.10554 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_sta…

Causal Inference with Outcomes Truncated by Death and Missing Not at Random
In clinical trials, principal stratification analysis is commonly employed to address the issue of truncation by death, where a subject dies before the outcome can be measured. However, in practice, many survivor outcomes may remain uncollected or be missing not at random, posing a challenge to standard principal stratification analyses. In this paper, we explore the identification, estimation, and bounds of the average treatment effect within a subpopulation of individuals who would potentiall…

At the edge of Donsker's Theorem: Asymptotics of multiscale scan statistics
Johann K\"ohne, Fabian Mies
https://arxiv.org/abs/2506.05112 https://

At the edge of Donsker's Theorem: Asymptotics of multiscale scan statistics
For nonparametric inference about a function, multiscale testing procedures resolve the need for bandwidth selection and achieve asymptotically optimal detection performance against a broad range of alternatives. However, critical values strongly depend on the noise distribution, and we argue that existing methods are either statistically infeasible, or asymptotically sub-optimal. To address this methodological challenge, we show how to develop a feasible multiscale test via weak convergence ar…

Bayesian inference of the magnetic field and chemical potential on holographic jet quenching in heavy-ion collisions
Liqiang Zhu, Zhan Gao, Weiyao Ke, Hanzhong Zhang
https://arxiv.org/abs/2506.00340

Bayesian inference of the magnetic field and chemical potential on holographic jet quenching in heavy-ion collisions
Jet quenching is studied in a background magnetic field and a finite baryon chemical potential. The production of energetic partons is calculated using the next-to-leading order (NLO) perturbative Quantum Chromodynamics (pQCD) parton model, while the parton energy loss formula is obtained from the AdS/CFT correspondence incorporating the magnetic field and baryon chemical potential effects. Using Bayesian inference, we systemically compare the theoretical calculations with experimental data for…

FlashMLA-ETAP: Efficient Transpose Attention Pipeline for Accelerating MLA Inference on NVIDIA H20 GPUs
Pencuo Zeren, Qiuming Luo, Rui Mao, Chang Kong
https://arxiv.org/abs/2506.01969

FlashMLA-ETAP: Efficient Transpose Attention Pipeline for Accelerating MLA Inference on NVIDIA H20 GPUs
Efficient inference of Multi-Head Latent Attention (MLA) is challenged by deploying the DeepSeek-R1 671B model on a single Multi-GPU server. This paper introduces FlashMLA-ETAP, a novel framework that enhances MLA inference for the single-instance deployment scenario on NVIDIA H20 GPUs. We propose the Efficient Transpose Attention Pipeline (ETAP), which reconfigures attention computation through transposition to align the KV context length with the \(M\)-dimension in WGMMA operations, significa…

This https://arxiv.org/abs/2501.16007 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

TOPLOC: A Locality Sensitive Hashing Scheme for Trustless Verifiable Inference
Large language models (LLMs) have proven to be very capable, but access to frontier models currently relies on inference providers. This introduces trust challenges: how can we be sure that the provider is using the model configuration they claim? We propose TOPLOC, a novel method for verifiable inference that addresses this problem. TOPLOC leverages a compact locality-sensitive hashing mechanism for intermediate activations, which can detect unauthorized modifications to models, prompts, or pr…

System-performance and cost modeling of Large Language Model training and inference
Wenzhe Guo, Joyjit Kundu, Uras Tos, Weijiang Kong, Giuliano Sisto, Timon Evenblij, Manu Perumkunnil
https://arxiv.org/abs/2507.02456

System-performance and cost modeling of Large Language Model training and inference
Large language models (LLMs), based on transformer architectures, have revolutionized numerous domains within artificial intelligence, science, and engineering due to their exceptional scalability and adaptability. However, the exponential growth in LLM size and complexity has outpaced advancements in compute capacity, memory bandwidth, network performance, and cost efficiency, posing significant challenges to their scalability on distributed systems. To address these limitations, alternative m…

Simulation-Based Inference for Adaptive Experiments
Brian M Cho, Aur\'elien Bibaut, Nathan Kallus
https://arxiv.org/abs/2506.02881 https://

Simulation-Based Inference for Adaptive Experiments
Multi-arm bandit experimental designs are increasingly being adopted over standard randomized trials due to their potential to improve outcomes for study participants, enable faster identification of the best-performing options, and/or enhance the precision of estimating key parameters. Current approaches for inference after adaptive sampling either rely on asymptotic normality under restricted experiment designs or underpowered martingale concentration inequalities that lead to weak power in p…

This https://arxiv.org/abs/2409.18858 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Predicting memorization within Large Language Models fine-tuned for classification
Large Language Models have received significant attention due to their abilities to solve a wide range of complex tasks. However these models memorize a significant proportion of their training data, posing a serious threat when disclosed at inference time. To mitigate this unintended memorization, it is crucial to understand what elements are memorized and why. This area of research is largely unexplored, with most existing works providing a posteriori explanations. To address this gap, we pro…

Causal Inference for Aggregated Treatment
Carolina Caetano, Gregorio Caetano, Brantly Callaway, Derek Dyal
https://arxiv.org/abs/2506.22885 https://…

Causal Inference for Aggregated Treatment
In this paper, we study causal inference when the treatment variable is an aggregation of multiple sub-treatment variables. Researchers often report marginal causal effects for the aggregated treatment, implicitly assuming that the target parameter corresponds to a well-defined average of sub-treatment effects. We show that, even in an ideal scenario for causal inference such as random assignment, the weights underlying this average have some key undesirable properties: they are not unique, the…

This https://arxiv.org/abs/2501.08524 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_…

Deep inference of simulated strong lenses in ground-based surveys
The large number of strong lenses discoverable in future astronomical surveys will likely enhance the value of strong gravitational lensing as a cosmic probe of dark energy and dark matter. However, leveraging the increased statistical power of such large samples will require further development of automated lens modeling techniques. We show that deep learning and simulation-based inference (SBI) methods produce informative and reliable estimates of parameter posteriors for strong lensing syste…

This https://arxiv.org/abs/2505.09999 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDC_…

ServeGen: Workload Characterization and Generation of Large Language Model Serving in Production
With the widespread adoption of Large Language Models (LLMs), serving LLM inference requests has become an increasingly important task, attracting active research advancements. Practical workloads play an essential role in this process: they are critical for motivating and benchmarking serving techniques and systems. However, the existing understanding of real-world LLM serving workloads is limited due to the lack of a comprehensive workload characterization. Prior analyses remain insufficient …

This https://arxiv.org/abs/2406.04655 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_sta…

Bayesian Inference for Spatial-Temporal Non-Gaussian Data Using Predictive Stacking
Analysing non-Gaussian spatial-temporal data requires introducing spatial dependence in generalised linear models through the link function of an exponential family distribution. Unlike in Gaussian likelihoods, inference is considerably encumbered by the inability to analytically integrate out the random effects and reduce the dimension of the parameter space. Iterative estimation algorithms struggle to converge due to the presence of weakly identified parameters. We devise Bayesian inference u…

Two-Sample Covariance Inference in High-Dimensional Elliptical Models
Nina D\"ornemann
https://arxiv.org/abs/2507.02640 https://…

Two-Sample Covariance Inference in High-Dimensional Elliptical Models
We propose a two-sample test for large-dimensional covariance matrices in generalized elliptical models. The test statistic is based on a U-statistic estimator of the squared Frobenius norm of the difference between the two population covariance matrices. This statistic was originally introduced by Li and Chen (2012, AoS) for the independent component model. As a key theoretical contribution, we establish a new central limit theorem for the U-statistics under elliptical data, valid under both t…

Memory Access Characterization of Large Language Models in CPU Environment and its Potential Impacts
Spencer Banasik
https://arxiv.org/abs/2506.01827 https…

Memory Access Characterization of Large Language Models in CPU Environment and its Potential Impacts
As machine learning algorithms are shown to be an increasingly valuable tool, the demand for their access has grown accordingly. Oftentimes, it is infeasible to run inference with larger models without an accelerator, which may be unavailable in environments that have constraints such as energy consumption, security, or cost. To increase the availability of these models, we aim to im- prove the LLM inference speed on a CPU-only environment by modifying the cache architecture. To determine what …

A Frequentist Simulation-Based Inference Treatment of Sterile Neutrino Global Fits
Joshua Villarreal, Julia Woodward, John Hardin, Janet Conrad
https://arxiv.org/abs/2507.01153

A Frequentist Simulation-Based Inference Treatment of Sterile Neutrino Global Fits
A critical challenge in particle physics is combining results from diverse experimental setups that measure the same physical quantity to enhance precision and statistical power, a process known as a global fit. Global fits of sterile neutrino searches, hunts for additional neutrino oscillation frequencies and amplitudes, present an intriguing case study. In such a scenario, the key assumptions underlying Wilks' theorem, a cornerstone of most classic frequentist analyses, do not hold. The metho…

Robust Estimation in Step-Stress Experiments under Exponential Lifetime Distributions
Mar\'ia Jaenada, Juan Manuel Mill\'an, Leandro Pardo
https://arxiv.org/abs/2506.04445

Robust Estimation in Step-Stress Experiments under Exponential Lifetime Distributions
Many modern products exhibit high reliability, often resulting in long times to failure. Consequently, conducting experiments under normal operating conditions may require an impractically long duration to obtain sufficient failure data for reliable statistical inference. As an alternative, accelerated life tests (ALTs) are employed to induce earlier failures and thereby reduce testing time. In step-stress experiments a stress factor that accelerates product degradation is identified and system…

Deep Recommender Models Inference: Automatic Asymmetric Data Flow Optimization
Giuseppe Ruggeri, Renzo Andri, Daniele Jahier Pagliari, Lukas Cavigelli
https://arxiv.org/abs/2507.01676

Deep Recommender Models Inference: Automatic Asymmetric Data Flow Optimization
Deep Recommender Models (DLRMs) inference is a fundamental AI workload accounting for more than 79% of the total AI workload in Meta's data centers. DLRMs' performance bottleneck is found in the embedding layers, which perform many random memory accesses to retrieve small embedding vectors from tables of various sizes. We propose the design of tailored data flows to speedup embedding look-ups. Namely, we propose four strategies to look up an embedding table effectively on one core, and a framew…

This https://arxiv.org/abs/2505.23655 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Keyed Chaotic Tensor Transformations for Secure And Attributable Neural Inference
This work introduces a novel framework for secure and privacy-preserving neural network inference based on keyed chaotic dynamical transformations. The proposed method applies a deterministic, cryptographically seeded chaotic system to tensors, producing non-invertible, user-specific transformations that enable authenticated inference, tensor-level watermarking, and data attribution. This framework offers a scalable and lightweight alternative to conventional cryptographic techniques, and estab…

Flexible Selective Inference with Flow-based Transport Maps
Sifan Liu, Snigdha Panigrahi
https://arxiv.org/abs/2506.01150 https://arx…

Flexible Selective Inference with Flow-based Transport Maps
Data-carving methods perform selective inference by conditioning the distribution of data on the observed selection event. However, existing data-carving approaches typically require an analytically tractable characterization of the selection event. This paper introduces a new method that leverages tools from flow-based generative modeling to approximate a potentially complex conditional distribution, even when the underlying selection event lacks an analytical description -- take, for example,…

DistMLIP: A Distributed Inference Platform for Machine Learning Interatomic Potentials
Kevin Han, Bowen Deng, Amir Barati Farimani, Gerbrand Ceder
https://arxiv.org/abs/2506.02023

DistMLIP: A Distributed Inference Platform for Machine Learning Interatomic Potentials
Large-scale atomistic simulations are essential to bridge computational materials and chemistry to realistic materials and drug discovery applications. In the past few years, rapid developments of machine learning interatomic potentials (MLIPs) have offered a solution to scale up quantum mechanical calculations. Parallelizing these interatomic potentials across multiple devices poses a challenging, but promising approach to further extending simulation scales to real-world applications. In this…

ProjMC$^2$: Scalable and Stable Posterior Inference for Bayesian Spatial Factor Models with Application to Spatial Transcriptomics
Lu Zhang
https://arxiv.org/abs/2506.01098

ProjMC$^2$: Scalable and Stable Posterior Inference for Bayesian Spatial Factor Models with Application to Spatial Transcriptomics
Factor models exhibit a fundamental tradeoff among flexibility, identifiability, and computational efficiency. Bayesian spatial factor models, in particular, face pronounced identifiability concerns and scaling difficulties. To mitigate these issues and enhance posterior inference reliability, this work proposes Projected Markov Chain Monte Carlo (ProjMC$^2$), a novel Markov Chain Monte Carlo (MCMC) sampling algorithm employing projection techniques and conditional conjugacy. ProjMC$^2$ is show…

Keyed Chaotic Tensor Transformations for Secure And Attributable Neural Inference
Peter David Fagan
https://arxiv.org/abs/2505.23655 https://

Keyed Chaotic Tensor Transformations for Secure And Attributable Neural Inference
This work introduces a novel framework for secure and privacy-preserving neural network inference based on keyed chaotic dynamical transformations. The proposed method applies a deterministic, cryptographically seeded chaotic system to tensors, producing non-invertible, user-specific transformations that enable authenticated inference, tensor-level watermarking, and data attribution. This framework offers a scalable and lightweight alternative to conventional cryptographic techniques, and estab…

QPART: Adaptive Model Quantization and Dynamic Workload Balancing for Accuracy-aware Edge Inference
Xiangchen Li, Saeid Ghafouri, Bo Ji, Hans Vandierendonck, Deepu John, Dimitrios S. Nikolopoulos
https://arxiv.org/abs/2506.23934

QPART: Adaptive Model Quantization and Dynamic Workload Balancing for Accuracy-aware Edge Inference
As machine learning inferences increasingly move to edge devices, adapting to diverse computational capabilities, hardware, and memory constraints becomes more critical. Instead of relying on a pre-trained model fixed for all future inference queries across diverse edge devices, we argue that planning an inference pattern with a request-specific model tailored to the device's computational capacity, accuracy requirements, and time constraints is more cost-efficient and robust to diverse scenari…

This https://arxiv.org/abs/2408.06211 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_sta…

Causal Effect Identification and Inference with Endogenous Exposures and a Light-tailed Error
Endogeneity poses significant challenges in causal inference across various research domains. This paper proposes a novel approach to identify and estimate causal effects in the presence of endogeneity. We consider a structural equation with endogenous exposures and an additive error term. Assuming the light-tailedness of the error term, we show that the causal effect can be identified by contrasting extreme conditional quantiles of the outcome given the exposures. Unlike many existing results,…

Ghidorah: Fast LLM Inference on Edge with Speculative Decoding and Hetero-Core Parallelism
Jinhui Wei, Ye Huang, Yuhui Zhou, Jiazhi Jiang, Jiangsu Du
https://arxiv.org/abs/2505.23219

Ghidorah: Fast LLM Inference on Edge with Speculative Decoding and Hetero-Core Parallelism
In-situ LLM inference on end-user devices has gained significant interest due to its privacy benefits and reduced dependency on external infrastructure. However, as the decoding process is memory-bandwidth-bound, the diverse processing units in modern end-user devices cannot be fully exploited, resulting in slow LLM inference. This paper presents Ghidorah, a LLM inference system for end-user devices with the unified memory architecture. The key idea of Ghidorah can be summarized in two steps: 1…

Find a Scapegoat: Poisoning Membership Inference Attack and Defense to Federated Learning
Wenjin Mo, Zhiyuan Li, Minghong Fang, Mingwei Fang
https://arxiv.org/abs/2507.00423

Find a Scapegoat: Poisoning Membership Inference Attack and Defense to Federated Learning
Federated learning (FL) allows multiple clients to collaboratively train a global machine learning model with coordination from a central server, without needing to share their raw data. This approach is particularly appealing in the era of privacy regulations like the GDPR, leading many prominent companies to adopt it. However, FL's distributed nature makes it susceptible to poisoning attacks, where malicious clients, controlled by an attacker, send harmful data to compromise the model. Most e…

Causal Inference in Panel Data with a Continuous Treatment
Zhiguo Xiao, Peikai Wu
https://arxiv.org/abs/2506.23226 https://arxiv.org/…

Causal Inference in Panel Data with a Continuous Treatment
This paper proposes a framework that incorporates the two-way fixed effects model as a special case to conduct causal inference with a continuous treatment. Treatments are allowed to change over time and potential outcomes are dependent on historical treatments. Regression models on potential outcomes, along with the sequentially conditional independence assumptions (SCIAs) are introduced to identify the treatment effects, which are measured by aggre causal responses. Least squares and generali…

LLM-Mesh: Enabling Elastic Sharing for Serverless LLM Inference
Chuhao Xu, Zijun Li, Quan Chen, Han Zhao, Minyi Guo
https://arxiv.org/abs/2507.00507 https:…

LLM-Mesh: Enabling Elastic Sharing for Serverless LLM Inference
The rise of LLMs has driven demand for private serverless deployments, characterized by moderate-scale models and infrequent requests. While existing solutions follow exclusive GPU deployment, we take a step back to explore modern platforms and find that: Emerging CPU architectures with built-in accelerators are capable of serving LLMs but remain underutilized, and both CPUs and GPUs can accommodate multiple LLMs simultaneously. We propose LLM-Mesh, a serverless inference scheme for small-to-…

Synopsis: Secure and private trend inference from encrypted semantic embeddings
Madelyne Xiao, Palak Jain, Micha Gorelick, Sarah Scheffler
https://arxiv.org/abs/2505.23880

Synopsis: Secure and private trend inference from encrypted semantic embeddings
WhatsApp and many other commonly used communication platforms guarantee end-to-end encryption (E2EE), which requires that service providers lack the cryptographic keys to read communications on their own platforms. WhatsApp's privacy-preserving design makes it difficult to study important phenomena like the spread of misinformation or political messaging, as users have a clear expectation and desire for privacy and little incentive to forfeit that privacy in the process of handing over raw data…

SkyLB: A Locality-Aware Cross-Region Load Balancer for LLM Inference
Tian Xia, Ziming Mao, Jamison Kerney, Ethan J. Jackson, Zhifei Li, Jiarong Xing, Scott Shenker, Ion Stoica
https://arxiv.org/abs/2505.24095

SkyLB: A Locality-Aware Cross-Region Load Balancer for LLM Inference
Serving Large Language Models (LLMs) efficiently in multi-region setups remains a challenge. Due to cost and GPU availability concerns, providers typically deploy LLMs in multiple regions using instance with long-term commitments, like reserved instances or on-premise clusters, which are often underutilized due to their region-local traffic handling and diurnal traffic variance. In this paper, we introduce SkyLB, a locality-aware multi-region load balancer for LLM inference that aggregates regi…

This https://arxiv.org/abs/2304.12414 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_sta…

Bayesian Geostatistics Using Predictive Stacking
We develop Bayesian predictive stacking for geostatistical models, where the primary inferential objective is to provide inference on the latent spatial random field and conduct spatial predictions at arbitrary locations. We exploit analytically tractable posterior distributions for regression coefficients of predictors and the realizations of the spatial process conditional upon process parameters. We subsequently combine such inference by stacking these models across the range of values of th…

SiPipe: Bridging the CPU-GPU Utilization Gap for Efficient Pipeline-Parallel LLM Inference
Yongchao He, Bohan Zhao, Zheng Cao
https://arxiv.org/abs/2506.22033

SiPipe: Bridging the CPU-GPU Utilization Gap for Efficient Pipeline-Parallel LLM Inference
As inference workloads for large language models (LLMs) scale to meet growing user demand, pipeline parallelism (PP) has become a widely adopted strategy for multi-GPU deployment, particularly in cross-node setups, to improve key-value (KV) cache capacity and inference throughput. However, PP suffers from inherent inefficiencies caused by three types of execution bubbles-load-imbalance, intra-stage, and inter-stage-which limit pipeline saturation. We present SiPipe, a heterogeneous pipeline des…

Large Language Models for Statistical Inference: Context Augmentation with Applications to the Two-Sample Problem and Regression
Marc Ratkovic
https://arxiv.org/abs/2506.23862

Large Language Models for Statistical Inference: Context Augmentation with Applications to the Two-Sample Problem and Regression
We introduce context augmentation, a data-augmentation approach that uses large language models (LLMs) to generate contexts around observed strings as a means of facilitating valid frequentist inference. These generated contexts serve to reintroduce uncertainty, incorporate auxiliary information, and facilitate interpretability. For example, in the two-sample test, we compare the log-probability of strings under contexts from its own versus the other group. We show on synthetic data that the me…

Reluctant Interaction Inference after Additive Modeling
Yiling Huang, Snigdha Panigrahi, Guo Yu, Jacob Bien
https://arxiv.org/abs/2506.01219 https://

Reluctant Interaction Inference after Additive Modeling
Additive models enjoy the flexibility of nonlinear models while still being readily understandable to humans. By contrast, other nonlinear models, which involve interactions between features, are not only harder to fit but also substantially more complicated to explain. Guided by the principle of parsimony, a data analyst therefore may naturally be reluctant to move beyond an additive model unless it is truly warranted. To put this principle of interaction reluctance into practice, we formula…

Cascadia: A Cascade Serving System for Large Language Models
Youhe Jiang, Fangcheng Fu, Wanru Zhao, Stephan Rabanser, Nicholas D. Lane, Binhang Yuan
https://arxiv.org/abs/2506.04203

Cascadia: A Cascade Serving System for Large Language Models
Recent advances in large language models (LLMs) have intensified the need to deliver both rapid responses and high-quality answers. More powerful models yield better results but incur higher inference latency, whereas smaller models are faster yet less capable. Recent work proposes balancing this latency-quality trade-off using model cascades, which route simpler queries to smaller models and more complex ones to larger models. However, enabling efficient cascade serving remains challenging. Cu…