Behavioral Fingerprints for LLM Endpoint Stability and Identity

• URL: http://arxiv.org/abs/2603.19022v1
• Date: Thu, 19 Mar 2026 15:23:23 GMT
• Title: Behavioral Fingerprints for LLM Endpoint Stability and Identity
• Abstract summary: Stability Monitor periodically fingerprints an endpoint by sampling outputs from a fixed prompt set.<n>In controlled validation, Stability Monitor detects changes to model family, version, inference stack, quantization, and behavioral parameters.<n>In real-world monitoring of the same model hosted by multiple providers, we observe substantial provider-to-provider and within-provider stability differences.
• Score: 6.5886705394788665
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The consistency of AI-native applications depends on the behavioral consistency of the model endpoints that power them. Traditional reliability metrics such as uptime, latency and throughput do not capture behavioral change, and an endpoint can remain "healthy" while its effective model identity changes due to updates to weights, tokenizers, quantization, inference engines, kernels, caching, routing, or hardware. We introduce Stability Monitor, a black-box stability monitoring system that periodically fingerprints an endpoint by sampling outputs from a fixed prompt set and comparing the resulting output distributions over time. Fingerprints are compared using a summed energy distance statistic across prompts, with permutation-test p-values as evidence of distribution shift aggregated sequentially to detect change events and define stability periods. In controlled validation, Stability Monitor detects changes to model family, version, inference stack, quantization, and behavioral parameters. In real-world monitoring of the same model hosted by multiple providers, we observe substantial provider-to-provider and within-provider stability differences.

Related papers

• CIAware-Bench: Benchmarking Control Intervention Awareness Across Frontier LLMs [100.38986535324284]
  We introduce textbfCIAware-Bench, a benchmark for measuring textbfcontrol textbfintervention (CI) awareness across frontier models.<n>CIAware-Bench tests whether models can distinguish their own trajectories from those modified by a control intervention.
  arXiv  Detail & Related papers  (2026-06-09T16:24:16Z)
• SS-TPT: Stability and Suitability-Guided Test-Time Prompt Tuning for Adversarially Robust Vision-Language Models [7.578599270855317]
  Vision-language models (VLMs) such as CLIP achieve strong zero-shot recognition but remain highly fragile under adversarial perturbations.<n>Recent test-time adaptation defenses improve robustness by leveraging many augmented views, but this leads to impractical slowdown.<n>We present Stability and Suitability-guided Test-time Prompt Tuning (SS-TPT) to evaluate the quality of each augmented view.
  arXiv  Detail & Related papers  (2026-06-05T06:12:48Z)
• Sequential statistical inference for Large Language Models: Representation, validity, and monitoring [10.902456009993898]
  This discussion argues that sequential statistical inference can naturally contribute to LLM trustworthiness.<n>The discussion is organized around three tasks: modeling interactions as dependent processes; validity, developing uncertainty guarantees; and monitoring, using sequential alarms and change-point detection to identify shifts in calibration, hallucination rates, refusal behavior, fairness, or other task-relevant properties.
  arXiv  Detail & Related papers  (2026-05-30T14:20:35Z)
• Explainable Fall Detection for Elderly Care via Temporally Stable SHAP in Skeleton-Based Human Activity Recognition [3.5944383615563615]
  We propose a lightweight and explainable framework for skeleton-based fall detection.<n>The proposed framework combines an efficient LSTM model with T-SHAP, a temporally aware post-hoc aggregation strategy.<n> Experiments show that the proposed framework achieves 94.3% classification accuracy with an end-to-end inference latency below 25 milliseconds.
  arXiv  Detail & Related papers  (2026-04-14T20:15:00Z)
• SMFormer: Empowering Self-supervised Stereo Matching via Foundation Models and Data Augmentation [55.67206878777881]
  We propose SMFormer, a framework integrating more reliable self-supervision guided by the Vision Foundation Model (VFM) and data augmentation.<n> SMFormer achieves state-of-the-art (SOTA) performance among self-supervised methods and even competes on par with supervised ones.
  arXiv  Detail & Related papers  (2026-04-11T13:56:41Z)
• When Sensors Fail: Temporal Sequence Models for Robust PPO under Sensor Drift [64.37959940809633]
  We study robustness of Proximal Policy Optimization (PPO) under temporally persistent sensor failures.<n>We show Transformer-based sequence policies substantially outperform, RNN, and SSMs in robustness, maintaining high returns even when large fractions of sensors are unavailable.
  arXiv  Detail & Related papers  (2026-03-04T22:21:54Z)
• Reasoning Stabilization Point: A Training-Time Signal for Stable Evidence and Shortcut Reliance [0.0]
  We define explanation drift as the epoch-to-epoch change in normalized token attributions on a fixed probe set.<n>RSP is computed from within-run drift dynamics and requires no tuning on out-of-distribution data.
  arXiv  Detail & Related papers  (2026-01-12T17:48:05Z)
• QDNA-ID Quantum Device Native Authentication [0.0]
  QDNA-ID is a trust-chain framework that links physical quantum behavior to digitally verified records.<n>A machine learning engine tracks entropy drift, detects anomalies, and classifies devices based on evolving patterns.<n>QDNA-ID operates as a continuous feedback loop that maintains a persistent chain of trust for quantum computing environments.
  arXiv  Detail & Related papers  (2025-11-21T15:40:59Z)
• Technical note on Sequential Test-Time Adaptation via Martingale-Driven Fisher Prompting [3.5808917363708743]
  M-FISHER is a method for sequential distribution shift detection and stable adaptation in streaming data.<n>For detection, we construct an exponential martingale from non-conformity scores and apply Ville's inequality to obtain time-uniform guarantees on false alarm control.<n>For adaptation, we show that Fisher-preconditioned updates of prompt parameters implement natural gradient descent on the distributional manifold.
  arXiv  Detail & Related papers  (2025-10-04T15:31:26Z)
• Segmented Confidence Sequences and Multi-Scale Adaptive Confidence Segments for Anomaly Detection in Nonstationary Time Series [0.0]
  We introduce and empirically evaluate two novel adaptive thresholding frameworks: Segmented Confidence Sequences (SCS) and Multi-Scale Adaptive Confidence Segments (MACS)<n>Our experiments across Wafer Manufacturing benchmark datasets show significant F1-score improvement compared to traditional percentile and rolling quantile approaches.
  arXiv  Detail & Related papers  (2025-08-08T18:34:54Z)
• Advancing Reliable Test-Time Adaptation of Vision-Language Models under Visual Variations [67.35596444651037]
  Vision-language models (VLMs) exhibit remarkable zero-shot capabilities but struggle with distribution shifts in downstream tasks when labeled data is unavailable.<n>We propose a Reliable Test-time Adaptation (ReTA) method that enhances reliability from two perspectives.
  arXiv  Detail & Related papers  (2025-07-13T05:37:33Z)
• A Poisson-Gamma Dynamic Factor Model with Time-Varying Transition Dynamics [51.147876395589925]
  A non-stationary PGDS is proposed to allow the underlying transition matrices to evolve over time. A fully-conjugate and efficient Gibbs sampler is developed to perform posterior simulation. Experiments show that, in comparison with related models, the proposed non-stationary PGDS achieves improved predictive performance.
  arXiv  Detail & Related papers  (2024-02-26T04:39:01Z)
• Generalized Robust Test-Time Adaptation in Continuous Dynamic Scenarios [18.527640606971563]
  Test-time adaptation (TTA) adapts pre-trained models to test distributions during the inference phase exclusively employing unlabeled test data streams. We propose a Generalized Robust Test-Time Adaptation (GRoTTA) method to effectively address the difficult problem.
  arXiv  Detail & Related papers  (2023-10-07T07:13:49Z)
• Deep Explicit Duration Switching Models for Time Series [84.33678003781908]
  We propose a flexible model that is capable of identifying both state- and time-dependent switching dynamics. State-dependent switching is enabled by a recurrent state-to-switch connection. An explicit duration count variable is used to improve the time-dependent switching behavior.
  arXiv  Detail & Related papers  (2021-10-26T17:35:21Z)

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.