RACE

anchor

Findings

• IC-264All 18 evaluated LLMs fail to abstain when the provided context lacks the answer, with performance gaps of 13.6% to 68.4% relative to the original context [source]
• IC-265Model families show extreme variation in detecting conflicting answers in inconsistent contexts, with phi-3 series at 5.8% average accuracy versus GPT-4 series at 89.35% [source]
• IC-762GPT-4 and GPT-3.5 outperform humans in generation but underperform in discriminative (selective) evaluation across 10 of 13 language tasks [eval]