Adithya S K(@adithya_s_k):I love benchmarks like these, but I don't really get the point of one that can't be reproduced openly. Code2Video: 16 models, evaluated by HeyGen, scored by a judge only HeyGen has. It's not a benchmark you can run. It's a result you can read.

Adithya S K

@adithya_s_k

Scaling RL Envs @huggingface 🤗 • Founded @cognitivelab_ai Prev : Research @MSFTResearch • ML @apple • 22

参加 June 2020

2.4K フォロー中 15.4K ファン

Adithya S K@adithya_s_k

2026.09.21 19:51

I love benchmarks like these, but I don't really get the point of one that can't be reproduced openly. Code2Video: 16 models, evaluated by HeyGen, scored by a judge only HeyGen has. It's not a benchmark you can run. It's a result you can read.

HyperFrames@HyperFrames_

2026.09.21 17:17

if you tried hyperframes, you would know the agentic video stack is being built on code-gen but SWE-bench code v.s. code-to-video that feels alive are two different things we built the Code2Video Bench, in collab w/ Google DeepMind & @Kaggle frontier labs can finally get good at agentic video tasks