PAL*M: Property Attestation for Large Generative Models

• URL: http://arxiv.org/abs/2601.16199v2
• Date: Sat, 24 Jan 2026 03:15:41 GMT
• Title: PAL*M: Property Attestation for Large Generative Models
• Authors: Prach Chantasantitam, Adam Ilyas Caulfield, Vasisht Duddu, Lachlan J. Gunn, N. Asokan,
• Abstract summary: We present PAL*M, a property attestation framework for large generative models, illustrated using large language models.<n> PAL*M defines properties across training and inference, leverages confidential virtual machines with security-aware GPUs for coverage of CPU-GPU operations, and proposes using incremental multiset hashing over memory-mapped datasets to efficiently track their integrity.<n>We implement PAL*M on Intel TDX and NVIDIA H100, showing it is efficient, scalable, versatile, and secure.
• Score: 9.787777791892784
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Machine learning property attestations allow provers (e.g., model providers or owners) to attest properties of their models/datasets to verifiers (e.g., regulators, customers), enabling accountability towards regulations and policies. But, current approaches do not support generative models or large datasets. We present PAL*M, a property attestation framework for large generative models, illustrated using large language models. PAL*M defines properties across training and inference, leverages confidential virtual machines with security-aware GPUs for coverage of CPU-GPU operations, and proposes using incremental multiset hashing over memory-mapped datasets to efficiently track their integrity. We implement PAL*M on Intel TDX and NVIDIA H100, showing it is efficient, scalable, versatile, and secure.

Related papers

• CREDIT: Certified Ownership Verification of Deep Neural Networks Against Model Extraction Attacks [54.04030169323115]
  We introduce CREDIT, a certified ownership verification against Model Extraction Attacks (MEAs)<n>We quantify the similarity between DNN models, propose a practical verification threshold, and provide rigorous theoretical guarantees for ownership verification based on this threshold.<n>We extensively evaluate our approach on several mainstream datasets across different domains and tasks, achieving state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-02-23T23:36:25Z)
• Ghost in the Transformer: Tracing LLM Lineage with SVD-Fingerprint [40.05267673405802]
  Large Language Models (LLMs) have rapidly advanced and are widely adopted across diverse fields.<n>Many developers choose to fine-tune or modify existing open-source models.<n>Some falsely claim original training despite clear derivation from public models.<n>This raises pressing concerns about intellectual property protection.
  arXiv  Detail & Related papers  (2025-11-09T13:57:59Z)
• Every Step Counts: Decoding Trajectories as Authorship Fingerprints of dLLMs [63.82840470917859]
  We show that the decoding mechanism of dLLMs can be used as a powerful tool for model attribution.<n>We propose a novel information extraction scheme called the Directed Decoding Map (DDM), which captures structural relationships between decoding steps and better reveals model-specific behaviors.
  arXiv  Detail & Related papers  (2025-10-02T06:25:10Z)
• ZKPROV: A Zero-Knowledge Approach to Dataset Provenance for Large Language Models [46.71493672772134]
  We introduce ZKPROV, a novel cryptographic framework that enables zero-knowledge proofs of LLM provenance.<n>It allows users to verify that a model is trained on a reliable dataset without revealing sensitive information about it or its parameters.<n>Our method cryptographically binds a trained model to its authorized training dataset(s) through zero-knowledge proofs while avoiding proof of every training step.
  arXiv  Detail & Related papers  (2025-06-26T00:49:02Z)
• LASA: Enhancing SoC Security Verification with LLM-Aided Property Generation [7.52190283487474]
  Formal property verification (FPV) provides the capability to model and validate design behaviors.<n>Current practices require significant manual efforts to create such properties, making them time-consuming, costly, and error-prone.<n>This paper presents LASA, a novel framework that leverages LLMs and retrieval-augmented generation (RAG) to produce non-vacuous security properties.
  arXiv  Detail & Related papers  (2025-06-22T01:21:03Z)
• A Collaborative Ensemble Framework for CTR Prediction [73.59868761656317]
  We propose a novel framework, Collaborative Ensemble Training Network (CETNet), to leverage multiple distinct models. Unlike naive model scaling, our approach emphasizes diversity and collaboration through collaborative learning. We validate our framework on three public datasets and a large-scale industrial dataset from Meta.
  arXiv  Detail & Related papers  (2024-11-20T20:38:56Z)
• Data-Juicer Sandbox: A Feedback-Driven Suite for Multimodal Data-Model Co-development [67.55944651679864]
  We present a new sandbox suite tailored for integrated data-model co-development.<n>This sandbox provides a feedback-driven experimental platform, enabling cost-effective and guided refinement of both data and models.
  arXiv  Detail & Related papers  (2024-07-16T14:40:07Z)
• Laminator: Verifiable ML Property Cards using Hardware-assisted Attestations [10.278905067763686]
  Industry has adopted model cards and verifiables to describe various properties training datasets and models.<n>We coin term ML property cards to collectively refer to these various types of cards.<n>To prevent a malicious model provider from including false information in ML property cards, they need to be verifiable.
  arXiv  Detail & Related papers  (2024-06-25T13:36:53Z)
• Dissecting Multimodality in VideoQA Transformer Models by Impairing Modality Fusion [54.33764537135906]
  VideoQA Transformer models demonstrate competitive performance on standard benchmarks. Do these models capture the rich multimodal structures and dynamics from video and text jointly? Are they achieving high scores by exploiting biases and spurious features?
  arXiv  Detail & Related papers  (2023-06-15T06:45:46Z)
• Just Fine-tune Twice: Selective Differential Privacy for Large Language Models [69.66654761324702]
  We propose a simple yet effective just-fine-tune-twice privacy mechanism to achieve SDP for large Transformer-based language models. Experiments show that our models achieve strong performance while staying robust to the canary insertion attack.
  arXiv  Detail & Related papers  (2022-04-15T22:36:55Z)
• Paraphrastic Representations at Scale [134.41025103489224]
  We release trained models for English, Arabic, German, French, Spanish, Russian, Turkish, and Chinese languages. We train these models on large amounts of data, achieving significantly improved performance from the original papers.
  arXiv  Detail & Related papers  (2021-04-30T16:55:28Z)

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.