TriggerBench - Negative Clean PM Accuracy (40K Context): leaderboard
Metric: PM accuracy (%) on the 289 negative controls, where the context makes the reminder unnecessary and restraint is correct; dialogues stretched to 40K tokens with unrelated turns injected between constraint and trigger; TriggerBench: 1,265 prospective-memory tasks from 488 dialogue blueprints in five dimensions (state tracking, temporal grounding, logical adherence, attention recovery, safe coding); a constraint is stated early in a dialogue and a later trigger turn calls for acting on it without being asked; GPT-4o (T=0) judges whether the reply fulfils the proactive intent; temperature 0.6; higher is better. Source: arxiv.org. Saturation forecast: Around December 2027. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 235B A22B 2507 Instruct | 82.7 |
| 2 | GPT-4.1 | 72.32 |
| 3 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 67.48 |
| 4 | GPT-5.2 (High) | 39.8 |
Interactive version: theaggregate.ai/benchmark?slug=triggerbench-negative-clean-pm-accuracy-40k-context · How It Works · Data refreshed daily, snapshot 2026-09-29.