每天自动聚合 AI 领域最新动态
On-policy distillation (OPD) offers a promising way to transfer reasoning capabilities from stronger teacher models, ...
Visual on-policy distillation relies heavily on an informative teacher-student asymmetry, through either a larger, st...
Neural language models trained on large crowdsourced corpora frequently exploit spurious surface patterns tied to tar...
Scientific figures and tables encode essential experimental evidence, yet remain difficult for digital libraries and ...
Implement a ChatGPT-like LLM in PyTorch from scratch, step by step（⭐102694）
This paper reports a single, fully instrumented case study of a large-scale architectural refactoring by an AI coding...
Large language models (LLMs) achieve remarkable performance across natural language tasks, yet they are trained on st...
Machine translation (MT) systems often fail to correctly translate gender, especially when converting from a gender-n...
Large audio-language models (LALMs) have demonstrated strong capabilities in understanding diverse audio inputs. This...
Interactive autoregressive video generation demands both low-latency rollouts and precise online control. Few-step di...
Rib fractures are common and time-consuming to localize on computed tomography (CT). We ask whether fractures detecte...
Test-time compute scaling is a primary driver of performance in large reasoning models (LRMs), but extreme inefficien...