Search

Any time Past 24 hours Past week Past monthSort:Relevance Newest

Benchmark

MMLU-Redux

Reported in 3 papers on Zotpaper. The latest, each as the paper states it:

Role prompts help language models only under the right conditions0.87 accuracy
Post-training multi-token prediction heads match joint pretraining with far less data2.64 MAL
Draft-KV ensures the receiver actually uses the sender's latent message78.04% accuracy

All benchmarks →

1 result for “MMLU-Redux”

Research · Agents
Draft-KV ensures the receiver actually uses the sender's latent message
Draft-KV: Learning Useful Latent Communication Between Language Models
By sending key-value states from the sharer's draft, a lightweight interface boosts a 0.5B receiver to 78% on MMLU-Redux, and message-replacement tests confirm the gain comes from genuine information use.
MMLU-Redux: 78.04% accuracy (as reported)
29 Sept 2026City University of Hong Kong, University of Science and Technology of China, University of Electronic Science and Technology of China