NEBench R1 converts U.S. NRC reactor-operator licensing examinations into a 7,546-question…
论文发布危机咨询评测套件CrisisBench,以200段去标识化热线对话构造1,085个单轮任务,并用热线元数据生成1,185个患者LLM,配合经专家校准的23项LLM裁判。五个…
FMT (Fixed-step Multiscale Tree) redesigns lossless image coding as a coarse-to-fine multi…
M3OPD-Reasoner presents an open post-training recipe for vision-language models. Starting …
Xing 4 extends the TeleChat lineage with an Ascend-oriented co-design of model architectur…
FastFrames advances autoregressive video diffusion from offline rendering toward genuinely…
Heave-Ho presents a decentralized architecture for long-horizon loco-manipulation across m…
TCP4D addresses robot manipulation learning from monocular RGB videos by representing acti…
Pi4 unifies single images, unordered image collections, and videos under a roughly 0.5B-pa…
The paper reframes model construction as weight-space programming: instead of optimizing p…
MemeEconomy introduces simulated economic survival pressure into a multimodal LLM-agent ma…