arXiv subjects

Claudio Mayrink Verdun

Publications and source records attributed to Claudio Mayrink Verdun.

At least 19 recordsLinked to original sources

arXiv2026

Evaluating AI Models' Capability to Automate Voice Phishing Attacks

Voice phishing (vishing) attacks have traditionally been limited by the need for human operators. The rapid emergence of high-quality AI voice synthesis and large language models (LLMs) reduces this bottleneck and enables scalable, automated scams. In this paper, we conduct a large-scale survey experiment (N=4100) and qualitative interviews (N=12) to assess U.S. adults' susceptibility to AI-powered voice phishing attacks. Participants were exposed to audio recordings or transcripts of scam scenarios generated using leading voice models such as Llama Full Duplex (Llama FD), Sesame, Gemini, OAI AVM, Play$.$AI, and ElevenLabs and the corresponding human baselines. The results show high compliance rates. Up to 36% of participants would or might comply with phishing requests in the "relative-in-distress" category. Overall compliance rate across all five scam categories was 16.5%, a striking figure given the low cost and high scalability of AI-automated voice phishing. Caller persuasiveness was the strongest predictor of compliance and certain models (most notably Sesame) achieved ratings comparable to human voices, or sometimes even slightly surpassing them. Our economic analysis suggests that while human-operated vishing is unprofitable at US wages, AI-powered vishing appears to be economically viable for several models. The primary risk of present-day AI-enabled vishing thus lies in the economics of automation rather than novel or "superhuman" persuasive techniques, though these cannot be ruled out for future systems. This raises significant concerns for the design of AI systems, consumer protection, and model release policies.

cs.CR↗

arXiv2026

KV Cache Compression Through the Lens of Transform Coding

The key-value (KV) cache stores information from past tokens and is a major memory bottleneck in long-context inference. Existing quantization methods address this bottleneck by representing the KV cache uniformly with lower-precision data types and designing quantization schemes to minimize reconstruction error in the cache itself, without accounting for how that error propagates through attention mechanisms. We prove that, under a white-noise quantization model, the expected attention-aware distortion decomposes into additive key and value contributions that factor across tokens and channels. Building on transform coding and reverse water-filling, which are classical tools from signal processing and rate-distortion theory, we introduce Attention-Aware Transform Coding (AATC), which allocates bits over a calibration set to minimize attention-aware distortion. On Llama-3.1-8B-Instruct and Qwen-2.5-7B-Instruct, evaluated across LongBench, RULER, GSM8K, MMLU-Pro, and MATH-500, our method achieves near-lossless accuracy at approximately $5.8\times$ compression, whereas each baseline degrades in at least some settings.

cs.LG↗

arXiv2026

An Instrument to Evaluate Governance Proposals: AI Policy Analysis at Scale

This paper introduces a policy analysis framework for systematic, transparent assessment of AI governance proposals in an evolving and contested regulatory landscape. AI policy debates often collapse into binary positions that obscure underlying tradeoffs and normative assumptions. The framework structures policy analysis around multiple policy attributes, allowing users to surface priorities and tensions without prescribing outcomes. We use a mixed-methods approach that integrates qualitative insights from subject matter experts with computational text analysis to inform the design of policy attribute rubrics. This quantifies the relative emphasis of different policy objectives and presents them through comparative visualizations that support interpretability and cross-policy comparison. The paper also examines the use of commercial LLMs for rubric-based policy analysis, benchmarking their outputs against a domain-trained rubric-calibrated model with explicitly defined analytical assumptions. Rather than assessing policy effectiveness or desirability, the framework focuses on relevance and alignment across attributes. By making analytical assumptions explicit, including attribute selection, rubric construction, and weighting schemes, the framework enables users to evaluate whether its embedded priorities align with the users' own normative commitments. The approach is jurisdiction-agnostic and intended to support policymakers, analysts, and researchers navigating complex AI governance environments. Contributions: (1) multidimensional policy assessment through empirically grounded rubrics that surface tradeoffs rather than resolving them; (2) a transparent hybrid methodology combining feedback from subject-matter experts with computational validation; and (3) use of domain-trained rubric-calibrated models as a benchmark for comparing different general-purpose large language models.

cs.AI↗

arXiv2026

Reading Between the Dots: Decoding Hidden Computation across Filler Tokens

Frontier LLMs can perform multi-step reasoning over content-free filler tokens like dots or counting sequences, producing correct answers with no visible chain-of-thought (CoT). This is a limit case for behavioral oversight, where surface tokens carry no information about the underlying reasoning. But hidden from the output is not the same as hidden from us. On four task families (fact retrieval, parallel numeric composition, string manipulation, and in-context computation), two open-weights frontier models (DeepSeek V3, Kimi K2) compute over filler tokens in a structured, legible way: attention routes the question through the filler region to the answer, logit-lens readouts show retrieved facts emerging early and their composition crystallizing in late layers, and KV-cache transplants at filler positions causally swap outputs between examples. We introduce an unsupervised decoding pipeline that takes only hidden states as input and recovers intermediate values with 80-95% accuracy (best LLM judge) across both models and all four tasks, without ground-truth labels or training. Hidden computation that defeats behavioral CoT monitoring is, on these tasks, directly readable from the residual stream, suggesting monitorability is a property of the model's full computational trace, not just its surface tokens.

cs.CL↗

arXiv2026

Prioritization of Risks from Artificial Intelligence: A Delphi Study of 272 International Experts

Artificial intelligence poses many risks, ranging from familiar present-day harms to unprecedented and potentially catastrophic ones. Effective risk management requires prioritization: we must understand which risks are most severe, who is most vulnerable, and who is most responsible for addressing them. We report results from a three-round Delphi study conducted late 2025 with 272 international AI experts. Experts rated 24 AI risks on harm probability and severity, sector and actor vulnerability, actor responsibility, and overall concern. Experts estimated the five most severe harms in the next 5 years were likely to come from dangerous capabilities, competitive dynamics, weapons & cyberattacks (including CBRNE), power centralization, and false information. In a business-as-usual scenario, experts judged 18 of 24 risks as having a more than 10% probability of catastrophic outcomes (e.g., more than 1 million deaths or more than USD 100B in financial loss) in the next 5 years (2025-2030). In a scenario where pragmatic mitigations are implemented, experts still judged five risks as having a more than 10% probability of catastrophic outcomes: dangerous capabilities, weapons & cyberattacks, environmental harm, inequality & unemployment, and power centralization. All 24 risks were judged as being more than 5% likely to cause catastrophic outcomes. AI users and the general public were judged the most vulnerable to these risks, but experts assigned the highest responsibility for addressing them to general-purpose AI developers and governance actors (including governments, regulators, and standards bodies). Across most risks, experts identified information, finance, and national security as the most vulnerable sectors. These findings can guide AI risk prioritization and clarify expert expectations about who should bear responsibility for mitigation.

cs.CY↗

arXiv2026

GradPCA: Leveraging NTK Alignment for Reliable Out-of-Distribution Detection

We introduce GradPCA, an Out-of-Distribution (OOD) detection method that exploits the low-rank structure of neural network gradients induced by Neural Tangent Kernel (NTK) alignment. GradPCA applies Principal Component Analysis (PCA) to gradient class-means, achieving more consistent performance than existing methods across standard image classification benchmarks. We provide a theoretical perspective on spectral OOD detection in neural networks to support GradPCA, highlighting feature-space properties that enable effective detection and naturally emerge from NTK alignment. Our analysis further reveals that feature quality -- particularly the use of pretrained versus non-pretrained representations -- plays a crucial role in determining which detectors will succeed. Extensive experiments validate the strong performance of GradPCA, and our theoretical framework offers guidance for designing more principled spectral OOD detectors.

cs.LG↗

arXiv2026

Temporal Sparse Autoencoders: Leveraging the Sequential Nature of Language for Interpretability

Translating the internal representations and computations of models into concepts that humans can understand is a key goal of interpretability. While recent dictionary learning methods such as Sparse Autoencoders (SAEs) provide a promising route to discover human-interpretable features, they often only recover token-specific, noisy, or highly local concepts. We argue that this limitation stems from neglecting the temporal structure of language, where semantic content typically evolves smoothly over sequences. Building on this insight, we introduce Temporal Sparse Autoencoders (T-SAEs), which incorporate a novel contrastive loss encouraging consistent activations of high-level features over adjacent tokens. This simple yet powerful modification enables SAEs to disentangle semantic from syntactic features in a self-supervised manner. Across multiple datasets and models, T-SAEs recover smoother, more coherent semantic concepts without sacrificing reconstruction quality. Strikingly, they exhibit clear semantic structure despite being trained without explicit semantic signal, offering a new pathway for unsupervised interpretability in language models.

cs.CL↗

arXiv2025

HeavyWater and SimplexWater: Distortion-Free LLM Watermarks for Low-Entropy Next-Token Predictions

Large language model (LLM) watermarks enable authentication of text provenance, curb misuse of machine-generated text, and promote trust in AI systems. Current watermarks operate by changing the next-token predictions output by an LLM. The updated (i.e., watermarked) predictions depend on random side information produced, for example, by hashing previously generated tokens. LLM watermarking is particularly challenging in low-entropy generation tasks -- such as coding -- where next-token predictions are near-deterministic. In this paper, we propose an optimization framework for watermark design. Our goal is to understand how to most effectively use random side information in order to maximize the likelihood of watermark detection and minimize the distortion of generated text. Our analysis informs the design of two new watermarks: HeavyWater and SimplexWater. Both watermarks are tunable, gracefully trading-off between detection accuracy and text distortion. They can also be applied to any LLM and are agnostic to side information generation. We examine the performance of HeavyWater and SimplexWater through several benchmarks, demonstrating that they can achieve high watermark detection accuracy with minimal compromise of text generation quality, particularly in the low-entropy regime. Our theoretical analysis also reveals surprising new connections between LLM watermarking and coding theory. The code implementation can be found in https://github.com/DorTsur/HeavyWater_SimplexWater

cs.CR↗

arXiv2025

Inference-Time Reward Hacking in Large Language Models

A common paradigm to improve the performance of large language models is optimizing for a reward model. Reward models assign a numerical score to an LLM's output that indicates, for example, how likely it is to align with user preferences or safety goals. However, reward models are never perfect. They inevitably function as proxies for complex desiderata such as correctness, helpfulness, and safety. By overoptimizing for a misspecified reward, we can subvert intended alignment goals and reduce overall performance, a phenomenon commonly referred to as reward hacking. In this work, we characterize reward hacking in inference-time alignment and demonstrate when and how we can mitigate it by hedging on the proxy reward. We study this phenomenon under Best-of-$n$ (BoN) and Soft Best-of-$n$ (SBoN), and we introduce Best-of-Poisson (BoP) that provides an efficient, near-exact approximation of the optimal reward-KL divergence policy at inference time. We show that the characteristic pattern of hacking as observed in practice (where the true reward first increases before declining) is an inevitable property of a broad class of inference-time mechanisms, including BoN and BoP. To counter this effect, we introduce HedgeTune, an efficient algorithm to find the optimal inference-time parameter. We demonstrate that hedging mitigates reward hacking and achieves superior reward-distortion tradeoffs on math, reasoning, and human-preference setups.

cs.LG↗

arXiv2025

ProofCompass: Enhancing Specialized Provers with LLM Guidance

Language models have become increasingly powerful tools for formal mathematical reasoning. However, most existing approaches rely exclusively on either large general-purpose models or smaller specialized models, each with distinct limitations, while training specialized large models still requires significant computational resources. This paper introduces ProofCompass, a novel hybrid methodology that achieves remarkable computational efficiency by strategically guiding existing specialized prover methods, such as DeepSeek-Prover-v1.5-RL (DSP-v1.5) with a Large Language Model (LLM) without requiring additional model training. The LLM provides natural language proof strategies and analyzes failed attempts to select intermediate lemmas, enabling effective problem decomposition. On the miniF2F benchmark, ProofCompass demonstrates substantial resource efficiency: it outperforms DSP-v1.5 ($54.9\% \rightarrow 55.3\%$) while using 25x fewer attempts ($3200 \rightarrow 128$). Our synergistic approach paves the way for simultaneously improving computational efficiency and accuracy in formal theorem proving.

cs.AI↗

arXiv2025

Multi-Group Proportional Representation for Text-to-Image Models

Text-to-image (T2I) generative models can create vivid, realistic images from textual descriptions. As these models proliferate, they expose new concerns about their ability to represent diverse demographic groups, propagate stereotypes, and efface minority populations. Despite growing attention to the "safe" and "responsible" design of artificial intelligence (AI), there is no established methodology to systematically measure and control representational harms in image generation. This paper introduces a novel framework to measure the representation of intersectional groups in images generated by T2I models by applying the Multi-Group Proportional Representation (MPR) metric. MPR evaluates the worst-case deviation of representation statistics across given population groups in images produced by a generative model, allowing for flexible and context-specific measurements based on user requirements. We also develop an algorithm to optimize T2I models for this metric. Through experiments, we demonstrate that MPR can effectively measure representation statistics across multiple intersectional groups and, when used as a training objective, can guide models toward a more balanced generation across demographic groups while maintaining generation quality.

cs.CV↗

arXiv2025

Optimized Couplings for Watermarking Large Language Models

Large-language models (LLMs) are now able to produce text that is, in many cases, seemingly indistinguishable from human-generated content. This has fueled the development of watermarks that imprint a ``signal'' in LLM-generated text with minimal perturbation of an LLM's output. This paper provides an analysis of text watermarking in a one-shot setting. Through the lens of hypothesis testing with side information, we formulate and analyze the fundamental trade-off between watermark detection power and distortion in generated textual quality. We argue that a key component in watermark design is generating a coupling between the side information shared with the watermark detector and a random partition of the LLM vocabulary. Our analysis identifies the optimal coupling and randomization strategy under the worst-case LLM next-token distribution that satisfies a min-entropy constraint. We provide a closed-form expression of the resulting detection rate under the proposed scheme and quantify the cost in a max-min sense. Finally, we provide an array of numerical results, comparing the proposed scheme with the theoretical optimum and existing schemes, in both synthetic data and LLM watermarking. Our code is available at https://github.com/Carol-Long/CC_Watermark

cs.CR↗

arXiv2025

Soft Best-of-n Sampling for Model Alignment

Best-of-$n$ (BoN) sampling is a practical approach for aligning language model outputs with human preferences without expensive fine-tuning. BoN sampling is performed by generating $n$ responses to a prompt and then selecting the sample that maximizes a reward function. BoN yields high reward values in practice at a distortion cost, as measured by the KL-divergence between the sampled and original distribution. This distortion is coarsely controlled by varying the number of samples: larger $n$ yields a higher reward at a higher distortion cost. We introduce Soft Best-of-$n$ sampling, a generalization of BoN that allows for smooth interpolation between the original distribution and reward-maximizing distribution through a temperature parameter $λ$. We establish theoretical guarantees showing that Soft Best-of-$n$ sampling converges sharply to the optimal tilted distribution at a rate of $O(1/n)$ in KL and the expected (relative) reward. For sequences of discrete outputs, we analyze an additive reward model that reveals the fundamental limitations of blockwise sampling.

cs.IT↗

arXiv2025

Get rid of your constraints and reparametrize: A study in NNLS and implicit bias

Over the past years, there has been significant interest in understanding the implicit bias of gradient descent optimization and its connection to the generalization properties of overparametrized neural networks. Several works observed that when training linear diagonal networks on the square loss for regression tasks (which corresponds to overparametrized linear regression) gradient descent converges to special solutions, e.g., non-negative ones. We connect this observation to Riemannian optimization and view overparametrized GD with identical initialization as a Riemannian GD. We use this fact for solving non-negative least squares (NNLS), an important problem behind many techniques, e.g., non-negative matrix factorization. We show that gradient flow on the reparametrized objective converges globally to NNLS solutions, providing convergence rates also for its discretized counterpart. Unlike previous methods, we do not rely on the calculation of exponential maps or geodesics. We further show accelerated convergence using a second-order ODE, lending itself to accelerated descent methods. Finally, we establish the stability against negative perturbations and discuss generalization to other constrained optimization problems.

math.OC↗

arXiv2025

AI Alignment at Your Discretion

In AI alignment, extensive latitude must be granted to annotators, either human or algorithmic, to judge which model outputs are `better' or `safer.' We refer to this latitude as alignment discretion. Such discretion remains largely unexamined, posing two risks: (i) annotators may use their power of discretion arbitrarily, and (ii) models may fail to mimic this discretion. To study this phenomenon, we draw on legal concepts of discretion that structure how decision-making authority is conferred and exercised, particularly in cases where principles conflict or their application is unclear or irrelevant. Extended to AI alignment, discretion is required when alignment principles and rules are (inevitably) conflicting or indecisive. We present a set of metrics to systematically analyze when and how discretion in AI alignment is exercised, such that both risks (i) and (ii) can be observed. Moreover, we distinguish between human and algorithmic discretion and analyze the discrepancy between them. By measuring both human and algorithmic discretion over safety alignment datasets, we reveal layers of discretion in the alignment process that were previously unaccounted for. Furthermore, we demonstrate how algorithms trained on these datasets develop their own forms of discretion in interpreting and applying these principles, which challenges the purpose of having any principles at all. Our paper presents the first step towards formalizing this core gap in current alignment processes, and we call on the community to further scrutinize and control alignment discretion.

cs.AI↗

arXiv2024

Multi-Group Proportional Representation in Retrieval

Image search and retrieval tasks can perpetuate harmful stereotypes, erase cultural identities, and amplify social disparities. Current approaches to mitigate these representational harms balance the number of retrieved items across population groups defined by a small number of (often binary) attributes. However, most existing methods overlook intersectional groups determined by combinations of group attributes, such as gender, race, and ethnicity. We introduce Multi-Group Proportional Representation (MPR), a novel metric that measures representation across intersectional groups. We develop practical methods for estimating MPR, provide theoretical guarantees, and propose optimization algorithms to ensure MPR in retrieval. We demonstrate that existing methods optimizing for equal and proportional representation metrics may fail to promote MPR. Crucially, our work shows that optimizing MPR yields more proportional representation across multiple intersectional groups specified by a rich function class, often with minimal compromise in retrieval accuracy.

cs.AI↗

arXiv2024

Measuring Progress in Dictionary Learning for Language Model Interpretability with Board Game Models

What latent features are encoded in language model (LM) representations? Recent work on training sparse autoencoders (SAEs) to disentangle interpretable features in LM representations has shown significant promise. However, evaluating the quality of these SAEs is difficult because we lack a ground-truth collection of interpretable features that we expect good SAEs to recover. We thus propose to measure progress in interpretable dictionary learning by working in the setting of LMs trained on chess and Othello transcripts. These settings carry natural collections of interpretable features -- for example, "there is a knight on F3" -- which we leverage into $\textit{supervised}$ metrics for SAE quality. To guide progress in interpretable dictionary learning, we introduce a new SAE training technique, $\textit{p-annealing}$, which improves performance on prior unsupervised metrics as well as our new metrics.

cs.LG↗

arXiv2024

With or Without Replacement? Improving Confidence in Fourier Imaging

Over the last few years, debiased estimators have been proposed in order to establish rigorous confidence intervals for high-dimensional problems in machine learning and data science. The core argument is that the error of these estimators with respect to the ground truth can be expressed as a Gaussian variable plus a remainder term that vanishes as long as the dimension of the problem is sufficiently high. Thus, uncertainty quantification (UQ) can be performed exploiting the Gaussian model. Empirically, however, the remainder term cannot be neglected in many realistic situations of moderately-sized dimensions, in particular in certain structured measurement scenarios such as Magnetic Resonance Imaging (MRI). This, in turn, can downgrade the advantage of the UQ methods as compared to non-UQ approaches such as the standard LASSO. In this paper, we present a method to improve the debiased estimator by sampling without replacement. Our approach leverages recent results of ours on the structure of the random nature of certain sampling schemes showing how a transition between sampling with and without replacement can lead to a weighted reconstruction scheme with improved performance for the standard LASSO. In this paper, we illustrate how this reweighted sampling idea can also improve the debiased estimator and, consequently, provide a better method for UQ in Fourier imaging.

eess.SP↗