LongPIBench - Resume Screening: leaderboard

Metric: Attack success rate (%; share of 100 long synthetic resumes screened against a job description; success when the decision is qualified, under the authority-spoofing prompt injection embedded in the context and the default attack goal, no defense; default inference settings, up to 20,000 output tokens (32,768 for Qwen3-8B)). Source: arxiv.org. Saturation forecast: Estimated already saturated. 8 models tracked.

Top models

#ModelScore
1GPT-4o76
2deepseek-llm-7B-chat97
3GPT-4.198
4Llama 3.2 3B Instruct98
5Llama 3.1 8B Instruct100
6Qwen 3 8B100

Interactive version: theaggregate.ai/benchmark?slug=longpibench-resume-screening · How It Works · Data refreshed daily, snapshot 2026-09-29.