cd/entity/GapFT· home› entities› GapFT
grep -l @gapft /news/*.json | wc -l → 1

GapFT

mentions 1 type Organization feed RSS

// recent coverage 1 mentions

04:00
2026-09-30
arxiv.org
large-language-models

Learning from the Gap Between Pass@K and Pass@1

A new post-training method called GapFT improves single-sample decoding accuracy of Llama-3.1-8B by 14.4 points on LogiQA 2.0 and 13.9 points on ReClor over the source model, according to an arXiv pap…

// co-occurs with top 5 entities
Llama-3.1-8B 1 Qwen2.5-7B 1 LogiQA 2.0 1 ReClor 1 arXiv 1
// topics top 4 topics
large language models 1 machine learning 1 ai research 1 natural language processing 1