#language models

2 stories · 0 in 7 days · 0 in 30 days

AI auditing 1 any-to-bench 1 evaluation 1 grading 1 measurement artifacts 1 model evaluation 1 rubric 1 self-improvement 1

Stories, products and papers tagged "language models", newest first.

8月21日

PapersSelected10:42

Phantom Gains in Language Model Self-Improvement

The research reveals critical flaws in measuring AI model improvements, affecting how researchers and developers evaluate self-training methods in their work.

#Qwen3-8B #LoRA #self-training #language models

arXiv cs.AI@Cheng Xu et al. (4)Source

8月19日

Papers11:43

Chinese content小模型按评分标准批改试卷与大模型同样可靠

研究发现小模型按评分标准批改试卷和大模型一样可靠，评分标准比模型智能更重要。

#rubric #language models #grading #evaluation

arXiv cs.AI@Jhen-Ke LinSource