TRACE (LRM Safety) - Reasoning Trace Evidence: leaderboard

TRACE (Wu et al., 2026) tests guardrail models on 5,000 prompt, reasoning-trace and final-response triples from four large reasoning models in English and Chinese, with safety labels and evidence spans from a majority of three LLM annotators. This board scores the evidence span a guardrail extracts to justify its judgment of the reasoning trace, by token-level F1 against the annotated evidence in percent; higher is better.

Metric: Token-level F1 (%; overlap between the evidence span a guardrail extracts for its judgment of the reasoning trace and the annotated evidence, TRACE English and Chinese). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 4 models tracked.

Top models

Interactive version: theaggregate.ai/benchmark?slug=trace-lrm-safety-reasoning-trace-evidence · How It Works · Data refreshed daily, snapshot 2026-11-12.