Entity · paper

Denial of Deadline: Network-Driven Accuracy Collapse in Distributed Inference Pipelines

paperactiveprovisionaldenial-of-deadline-network-driven-accuracy-collapse-in-distributed-inference-pipelines-2d50e535·1 events·first seen 3d ago

Aliases: Denial of Deadline: Network-Driven Accuracy Collapse in Distributed Inference Pipelines

Co-occurring entities

HOTA

More like this (12)

The Illusion of Robustness: Aggregate Accuracy Hides Prediction Flips under Task-Irrelevant Context Rethinking Inference-Time Scaling in Local Computer-Use Agents: Failure Modes and Compute Tradeoffs Contagion Networks: Evaluator Bias Propagation in Multi-Agent LLM Systems Accelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques One-Step Gradient Delay is Not a Barrier for Large-Scale Asynchronous Pipeline Parallel LLM Pretraining Dynamic-Probabilistic Consistency Gap Score Accuracy Along the Forward Diffusion Does Not Certify Numerical Stability in Diffusion Sampling When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models Pessimism's Paradox: Conservative Offline Training Amplifies Reward Hacking During Online Adaptation in Reasoning Models Criticality-Based Guard Rail Validation for AI Agent Decisions in Autonomous Telecom Networks Agentic AI Pipelines Neural Collapse Is Forbidden: Information Floors in Language Models

Recent events (1)

6arXiv · cs.AI·3d ago·source ↗

Denial-of-Deadline attack causes accuracy collapse in distributed two-tier inference pipelines

A new arXiv paper introduces 'Denial of Deadline', a class of network-level workload attacks that exploit shared resource contention in fast-path/slow-path inference architectures to push slow-path predictions past latency deadlines, causing the merger to discard them. The authors demonstrate the attack on an edge-cloud multi-object tracking pipeline for autonomous driving, where ~4,000 burst-shaped requests inflate p99 latency from 92ms to 2s and reduce tracking quality by up to 18.7 HOTA points. The attack requires no access to model weights or victim data, exposing a new attack surface in emerging distributed inference systems.

Inference Economics AI Safety Research Denial of Deadline: Network-Driven Accuracy Collapse in Distributed Inference Pipelines HOTA