Zero-Shot Attribution for Large Language Models: A Distribution Testing Approach

• URL: http://arxiv.org/abs/2506.20197v1
• Date: Wed, 25 Jun 2025 07:37:16 GMT
• Title: Zero-Shot Attribution for Large Language Models: A Distribution Testing Approach
• Authors: Clément L. Canonne, Yash Pote, Uddalok Sarkar,
• Abstract summary: We investigate the problem of attributing code generated by language models using hypothesis testing to leverage established techniques and guarantees.<n>We introduce $mathsfAnubis$, a zero-shot attribution tool that frames attribution as a distribution testing problem.
• Score: 19.455425068600665
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: A growing fraction of all code is sampled from Large Language Models (LLMs). We investigate the problem of attributing code generated by language models using hypothesis testing to leverage established techniques and guarantees. Given a set of samples $S$ and a suspect model $\mathcal{L}^*$, our goal is to assess the likelihood of $S$ originating from $\mathcal{L}^*$. Due to the curse of dimensionality, this is intractable when only samples from the LLM are given: to circumvent this, we use both samples and density estimates from the LLM, a form of access commonly available. We introduce $\mathsf{Anubis}$, a zero-shot attribution tool that frames attribution as a distribution testing problem. Our experiments on a benchmark of code samples show that $\mathsf{Anubis}$ achieves high AUROC scores ( $\ge0.9$) when distinguishing between LLMs like DeepSeek-Coder, CodeGemma, and Stable-Code using only $\approx 2000$ samples.

Related papers

• Blackbox Dataset Inference for LLM [27.02176845242058]
  Training large language models can involve personally identifiable information and copyrighted material.<n>This paper explores textitdataset inference, which aims to detect if a suspect model used a victim dataset $mathcalD$ in training.
  arXiv  Detail & Related papers  (2025-07-04T14:45:41Z)
• Outsourced diffusion sampling: Efficient posterior inference in latent spaces of generative models [65.71506381302815]
  We propose amortize the cost of sampling from a posterior distribution of the form $p(mathbfxmidmathbfy) propto p_theta(mathbfx)$.<n>For many models and constraints, the posterior in noise space is smoother than in data space, making it more suitable for amortized inference.
  arXiv  Detail & Related papers  (2025-02-10T19:49:54Z)
• Neural network learns low-dimensional polynomials with SGD near the information-theoretic limit [75.4661041626338]
  We study the problem of gradient descent learning of a single-index target function $f_*(boldsymbolx) = textstylesigma_*left(langleboldsymbolx,boldsymbolthetarangleright)$<n>We prove that a two-layer neural network optimized by an SGD-based algorithm learns $f_*$ with a complexity that is not governed by information exponents.
  arXiv  Detail & Related papers  (2024-06-03T17:56:58Z)
• Some Notes on the Sample Complexity of Approximate Channel Simulation [2.4554686192257424]
  Channel simulation algorithms can efficiently encode random samples from a prescribed target distribution $Q$ and find applications in machine learning-based lossy data compression. This paper considers approximate schemes with a fixed runtime instead. We exploit global-bound, depth-limited A* coding to ensure $mathrmTV[Q Vert P] leq epsilon$ and maintain optimal coding performance with a sample complexity of only $expbig((D_KL[Q Vert P] + o(1)) big/ epsilonbig
  arXiv  Detail & Related papers  (2024-05-07T14:44:41Z)
• Testable Learning with Distribution Shift [9.036777309376697]
  We define a new model called testable learning with distribution shift. We obtain provably efficient algorithms for certifying the performance of a classifier on a test distribution. We give several positive results for learning concept classes such as halfspaces, intersections of halfspaces, and decision trees.
  arXiv  Detail & Related papers  (2023-11-25T23:57:45Z)
• Testing with Non-identically Distributed Samples [20.74768558932617]
  We examine the extent to which sublinear-sample property testing and estimation applies to settings where samples are independently but not identically distributed. Even with $Theta(k/varepsilon2)$ samples from each distribution, $textbfp_mathrmavg$ is sufficient to learn $textbfp_mathrmavg$ to within error $varepsilon$ in TV distance.
  arXiv  Detail & Related papers  (2023-11-19T01:25:50Z)
• Stochastic Approximation Approaches to Group Distributionally Robust Optimization and Beyond [89.72693227960274]
  This paper investigates group distributionally robust optimization (GDRO) with the goal of learning a model that performs well over $m$ different distributions. To reduce the number of samples in each round from $m$ to 1, we cast GDRO as a two-player game, where one player conducts and the other executes an online algorithm for non-oblivious multi-armed bandits. In the second scenario, we propose to optimize the average top-$k$ risk instead of the maximum risk, thereby mitigating the impact of distributions.
  arXiv  Detail & Related papers  (2023-02-18T09:24:15Z)
• Robust Testing in High-Dimensional Sparse Models [0.0]
  We consider the problem of robustly testing the norm of a high-dimensional sparse signal vector under two different observation models. We show that any algorithm that reliably tests the norm of the regression coefficient requires at least $n=Omegaleft(min(slog d,1/gamma4)right) samples.
  arXiv  Detail & Related papers  (2022-05-16T07:47:22Z)
• Tight Bounds on the Hardness of Learning Simple Nonparametric Mixtures [9.053430799456587]
  We study the problem of learning nonparametric distributions in a finite mixture. We establish tight bounds on the sample complexity for learning the component distributions in such models.
  arXiv  Detail & Related papers  (2022-03-28T23:53:48Z)
• The Sample Complexity of Robust Covariance Testing [56.98280399449707]
  We are given i.i.d. samples from a distribution of the form $Z = (1-epsilon) X + epsilon B$, where $X$ is a zero-mean and unknown covariance Gaussian $mathcalN(0, Sigma)$. In the absence of contamination, prior work gave a simple tester for this hypothesis testing task that uses $O(d)$ samples. We prove a sample complexity lower bound of $Omega(d2)$ for $epsilon$ an arbitrarily small constant and $gamma
  arXiv  Detail & Related papers  (2020-12-31T18:24:41Z)
• Optimal Testing of Discrete Distributions with High Probability [49.19942805582874]
  We study the problem of testing discrete distributions with a focus on the high probability regime. We provide the first algorithms for closeness and independence testing that are sample-optimal, within constant factors.
  arXiv  Detail & Related papers  (2020-09-14T16:09:17Z)
• Model-Free Reinforcement Learning: from Clipped Pseudo-Regret to Sample Complexity [59.34067736545355]
  Given an MDP with $S$ states, $A$ actions, the discount factor $gamma in (0,1)$, and an approximation threshold $epsilon > 0$, we provide a model-free algorithm to learn an $epsilon$-optimal policy. For small enough $epsilon$, we show an improved algorithm with sample complexity.
  arXiv  Detail & Related papers  (2020-06-06T13:34:41Z)
• Locally Private Hypothesis Selection [96.06118559817057]
  We output a distribution from $mathcalQ$ whose total variation distance to $p$ is comparable to the best such distribution. We show that the constraint of local differential privacy incurs an exponential increase in cost. Our algorithms result in exponential improvements on the round complexity of previous methods.
  arXiv  Detail & Related papers  (2020-02-21T18:30:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.