DISCURSIVE

30 karmajoined October 9, 2015On HN

Recent submissions

Nvidia's Openness Promise Leaves Out Hosting, Gating and Licensing topicqueue.substack.com
1DISCURSIVE2d1 comment
Perceptron Ships 36B Robot Weights Under Apache-2.0 topicqueue.substack.com
2DISCURSIVE9ddiscuss
Model Success Rate May Count the Wrong Runs topicqueue.substack.com
1DISCURSIVE16ddiscuss
RoboDojo Scores 32 of 34 Robot Policies Under 10% Success topicqueue.substack.com
2DISCURSIVE23ddiscuss
Dyna-2: A Million Hours of Human Video, Zero Robot Frames topicqueue.substack.com
1DISCURSIVE30ddiscuss
Gemini Robotics Released One Policy from Feet to Fingertips topicqueue.substack.com
2DISCURSIVE1modiscuss
5k Hours of Tactile Data, and Four Ways to Get It topicqueue.substack.com
1DISCURSIVE1modiscuss
Nvidia Puts a 4B World Model on the Robot [Weekly Physical AI Roundup]topicqueue.substack.com
2DISCURSIVE1modiscuss
Turning robot video into training-ready data eventual.ai
2DISCURSIVE1modiscuss
Xiaomi Opens a 38B World Model Built to Generate Robot Data topicqueue.substack.com
4DISCURSIVE1modiscuss
Querying Physical AI Data with Daft eventual.ai
1DISCURSIVE2modiscuss
LeRobot v0.6.0 Adds World Models, Reward Models, and an Open GR00T topicqueue.substack.com
1DISCURSIVE2modiscuss
A Push on Tactile Data, and a Warning That the Benchmark Was Leaking topicqueue.substack.com
2DISCURSIVE2modiscuss
Hours of Humanoid Teleop, Recorded in Real Homes topicqueue.substack.com
3DISCURSIVE2modiscuss
Knowledge curation (not search) is the AI big data problem daft.ai
3DISCURSIVE8modiscuss
Cutting LLM Batch Inference Time by Half with Dynamic Prefix Bucketing daft.ai
2DISCURSIVE9modiscuss
Agentic systems are just query engines for unstructured data daft.ai
3DISCURSIVE9modiscuss
Benchmark: Spark vs. Ray Data vs. Daft on Multimodal Workloads daft.ai
1DISCURSIVE11modiscuss
Show HN: Flotilla, the Distributed Engine for Multimodal Pipelines daft.ai
2DISCURSIVE11modiscuss
Profiling multimodal workloads: lessons from Daft daft.ai
1DISCURSIVE11modiscuss
Processing 24T tokens for LLM training with 0 crashes (what made it possible)daft.ai
1DISCURSIVE1ydiscuss
We Hit 100% GPU Utilization–and Then Made It 3× Faster by Not Using It daft.ai
17DISCURSIVE1y11 comments
Elasticsearch Was Great, but Vector Databases Are the Future thenewstack.io
16DISCURSIVE1y7 comments
Have you faced any challenges of using Elasticsearch for vector search?
1DISCURSIVE1ydiscuss
BGE-M3 vs. Splade zilliz.com
1DISCURSIVE2ydiscuss
Enhance information retrieval with learned sparse embeddings zilliz.com
1DISCURSIVE2ydiscuss
The emerging trend of vector databases zilliz.com
2DISCURSIVE2ydiscuss
The Future of Vector Database zilliz.com
1DISCURSIVE2ydiscuss
Milvus Supports NVDIA's Graph Index zilliz.com
2DISCURSIVE2ydiscuss
Zilliz Cloud Jan 2024 Launch zilliz.com
1DISCURSIVE2ydiscuss

All submissions and comments on Hacker News.