Toward Reproducible Cross-Backend Compatibility for Deep Learning: A Configuration-First Framework with Three-Tier Verification
This paper presents a configuration-first framework for evaluating cross-backend compatibility in deep learning systems deployed on CPU, GPU, and compiled runtimes. The framework decouples experiments from code using YAML, supports both library and repository models, and employs a three-tier verification protocol covering tensor-level closeness, activation alignment, and task-level metrics. Through 672 checks across multiple models and tolerance settings, we observe that 72.0% of runs pass, with most discrepancies occurring under stricter thresholds. Our results show that detection models and compiled backends are particularly prone to drift, often due to nondeterministic post-processing. We further demonstrate that deterministic adapters and selective fallbacks can substantially improve agreement without significant performance loss. To our knowledge, this is the first unified framework that systematically quantifies and mitigates cross-backend drift in deep learning, providing a reproducible methodology for dependable deployment across heterogeneous runtimes.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
AlignTune: Modular Toolkit for Post-Training Alignment of Large Language Models
Post-training alignment is central to deploying large language models (LLMs), yet practical workflows remain split across backend-specific tools and ad-hoc glue code, making experiments hard to reproduce. We identify bac…
Is Your NPU Ready for LLMs? Dissecting the Hidden Efficiency Bottlenecks in Mobile LLM Inference
Deploying Large Language Models (LLMs) on mobile devices enhances privacy and reduces latency, but is severely bottlenecked by hardware inefficiency. We present the first comprehensive, cross-layer measurement study of m…
Mind the Boundary: Stabilizing Gemini Enterprise A2A via a Cloud Run Hub Across Projects and Accounts
Enterprise conversational UIs increasingly need to orchestrate heterogeneous backend agents and tools across project and account boundaries in a secure and reproducible way. Starting from Gemini Enterprise Agent-to-Agent…
Question AnsweringGeneral KnowledgeMemSecBench: Tracking Agent Memory Poisoning from Persistence to Consequence and Repair
Memory systems allow agents to retain and reuse information from past interactions, but they can also let malicious content persist. A malicious instruction crafted by an attacker may be stored in long-term memory, recal…
Dooly: Configuration-Agnostic, Redundancy-Aware Profiling for LLM Inference Simulation
Selecting the optimal LLM inference configuration requires evaluation across hardware, serving engines, attention backends, and model architectures, since no single choice performs best across all workloads. Profile-base…