Time-To-Inconsistency: A Survival Analysis of Large Language Model Robustness to Adversarial Attacks

• URL: http://arxiv.org/abs/2510.02712v1
• Date: Fri, 03 Oct 2025 04:26:10 GMT
• Title: Time-To-Inconsistency: A Survival Analysis of Large Language Model Robustness to Adversarial Attacks
• Authors: Yubo Li, Ramayya Krishnan, Rema Padman,
• Abstract summary: We present the first comprehensive survival analysis of conversational AI robustness, analyzing 36,951 conversation turns across 9 state-of-the-art LLMs.<n>We find that abrupt, prompt-to-prompt(P2P) semantic drift is catastrophic, dramatically increasing the hazard of conversational failure.<n>AFT models with interactions demonstrate superior performance, achieving excellent discrimination and exceptional calibration.
• Score: 8.86745721473138
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Language Models (LLMs) have revolutionized conversational AI, yet their robustness in extended multi-turn dialogues remains poorly understood. Existing evaluation frameworks focus on static benchmarks and single-turn assessments, failing to capture the temporal dynamics of conversational degradation that characterize real-world interactions. In this work, we present the first comprehensive survival analysis of conversational AI robustness, analyzing 36,951 conversation turns across 9 state-of-the-art LLMs to model failure as a time-to-event process. Our survival modeling framework-employing Cox proportional hazards, Accelerated Failure Time, and Random Survival Forest approaches-reveals extraordinary temporal dynamics. We find that abrupt, prompt-to-prompt(P2P) semantic drift is catastrophic, dramatically increasing the hazard of conversational failure. In stark contrast, gradual, cumulative drift is highly protective, vastly reducing the failure hazard and enabling significantly longer dialogues. AFT models with interactions demonstrate superior performance, achieving excellent discrimination and exceptional calibration. These findings establish survival analysis as a powerful paradigm for evaluating LLM robustness, offer concrete insights for designing resilient conversational agents, and challenge prevailing assumptions about the necessity of semantic consistency in conversational AI Systems.

Related papers

• Mitigating Conversational Inertia in Multi-Turn Agents [47.35031006899519]
  We identify conversational inertia, a phenomenon where models exhibit strong diagonal attention to previous responses.<n>We propose Context Preference Learning to calibrate model preferences to favor low-inertia responses over highinertia ones.
  arXiv  Detail & Related papers  (2026-02-03T15:47:32Z)
• The Anatomy of Conversational Scams: A Topic-Based Red Teaming Analysis of Multi-Turn Interactions in LLMs [3.7304174114240545]
  We study novel risks in multi-turn conversational scams that single-turn safety evaluations fail to capture.<n>We evaluate eight state-of-the-art models in English and Chinese.<n>Results reveal that scam interactions follow recurrent escalation patterns, while defenses employ verification and delay mechanisms.
  arXiv  Detail & Related papers  (2026-01-06T16:06:04Z)
• MORE: Multi-Objective Adversarial Attacks on Speech Recognition [39.77140497042348]
  Large-scale automatic speech recognition (ASR) models such as Whisper have expanded their adoption across diverse real-world applications.<n> robustness against even minor input perturbations is therefore critical for maintaining reliable performance in real-time environments.<n>We introduce MORE, a multi-objective repetitive doubling encouragement attack, which jointly degrades recognition accuracy and inference efficiency.
  arXiv  Detail & Related papers  (2026-01-05T07:27:57Z)
• Drift No More? Context Equilibria in Multi-Turn LLM Interactions [58.69551510148673]
  contexts drift is the gradual divergence of a model's outputs from goal-consistent behavior across turns.<n>Unlike single-turn errors, drift unfolds temporally and is poorly captured by static evaluation metrics.<n>We show that multi-turn drift can be understood as a controllable equilibrium phenomenon rather than as inevitable decay.
  arXiv  Detail & Related papers  (2025-10-09T04:48:49Z)
• FLEXI: Benchmarking Full-duplex Human-LLM Speech Interaction [49.83226596963294]
  Speech-computer human interaction enables real-time spoken dialogue systems.<n>Modelling and benchmarking these models remains a fundamental challenge.<n>We introduce FLEXI, the first benchmark for full-human spoken interaction.
  arXiv  Detail & Related papers  (2025-09-26T11:57:42Z)
• Investigating and Enhancing the Robustness of Large Multimodal Models Against Temporal Inconsistency [59.05753942719665]
  We propose a novel temporal robustness benchmark (TemRobBench) to assess the robustness of models.<n>We evaluate 16 mainstream LMMs and find that they exhibit over-reliance on prior knowledge and textual context in adversarial environments.<n>We design panoramic direct preference optimization (PanoDPO) to encourage LMMs to incorporate both visual and linguistic feature preferences simultaneously.
  arXiv  Detail & Related papers  (2025-05-20T14:18:56Z)
• Detect, Explain, Escalate: Low-Carbon Dialogue Breakdown Management for LLM-Powered Agents [30.13634341221476]
  Large Language Models (LLMs) are transforming numerous applications, but their susceptibility to conversational breakdowns remains a critical challenge undermining user trust.<n>This paper introduces a "Detect, Explain, Escalate" framework to manage dialogue breakdowns in LLM-powered agents, emphasizing low-carbon operation.
  arXiv  Detail & Related papers  (2025-04-26T07:51:05Z)
• Temporal Context Awareness: A Defense Framework Against Multi-turn Manipulation Attacks on Large Language Models [0.0]
  Large Language Models (LLMs) are increasingly vulnerable to sophisticated multi-turn manipulation attacks.<n>This paper introduces the Temporal Context Awareness framework, a novel defense mechanism designed to address this challenge.<n>Preliminary evaluations on simulated adversarial scenarios demonstrate the framework's potential to identify subtle manipulation patterns.
  arXiv  Detail & Related papers  (2025-03-18T22:30:17Z)
• Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models [53.580928907886324]
  Reasoning-Augmented Conversation is a novel multi-turn jailbreak framework.<n>It reformulates harmful queries into benign reasoning tasks.<n>We show that RACE achieves state-of-the-art attack effectiveness in complex conversational scenarios.
  arXiv  Detail & Related papers  (2025-02-16T09:27:44Z)
• Exchange-of-Thought: Enhancing Large Language Model Capabilities through Cross-Model Communication [76.04373033082948]
  Large Language Models (LLMs) have recently made significant strides in complex reasoning tasks through the Chain-of-Thought technique. We propose Exchange-of-Thought (EoT), a novel framework that enables cross-model communication during problem-solving.
  arXiv  Detail & Related papers  (2023-12-04T11:53:56Z)

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.