Hacker Newsnew | past | comments | ask | show | jobs | submit | anima-core's submissions login

1. We keep making transformers faster. What if we don't need them at all? (zenodo.org) 1 point by anima-core 48 days ago | past | 1 comment 2. A One-Shot Catastrophe Avoidance Benchmark for RL Agents (zenodo.org) 2 points by anima-core 49 days ago | past | 1 comment 3. Semantic Field Execution: Decoupling Transformers from Runtime Inference (zenodo.org) 1 point by anima-core 54 days ago | past | 1 comment 4. The Inference Efficiency Paradox (substack.com) 2 points by anima-core 54 days ago | past 5. [flagged] Post-transformer inference: 224× compression of Llama-70B with improved accuracy (zenodo.org) 72 points by anima-core 62 days ago | past | 56 comments

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact Search: