INNOVATION

→ View original post on X — @alexjc

Building More in the Open: Open Source Development Approach

By

@alexjc

–

09 April 2023 18h07

Yeah, maybe that's better than the alternative. More stuff built in the open?

→ View original post on X — @marek_rosa

Local LLMs on Gaming GPUs: Cost and Future Potential

By

@marek_rosa

–

09 April 2023 17h58

Currently it’s not, it would costs dollars per hour per player. But I expect this will change soon when small but capable LLMs will run locally on player’s GPU.

→ View original post on X — @marek_rosa

Pre-trained LLMs for NPC Game Control via Prompting

By

@marek_rosa

–

09 April 2023 17h52

You need pre-trained LLM which doesn’t need to be fine-tuned for this game (e.g. GPT3.5) and you prompt it with NPC’s observations, personality traits, memories and description of possible game actions/goals.
This is the principle. Then the devil is in the details… to make it

Pythia: Suite for Analyzing LLMs Across Training and Scaling

By

–

09 April 2023 17h41

9/ Pythia – a suite for analyzing LLMs across training and scaling; includes 16 LLMs trained on public data and ranging in size from 70M to 12B parameters.

SegGPT: Generalist Segmentation Model with In-Context Learning

By

–

09 April 2023 17h41

10/ SegGPT – unifies segmentation tasks into a generalist model through an in-context framework that supports different kinds of data.https://t.co/PiocmACjfo
— DAIR.AI (@dair_ai) 9 avril 2023

10/ SegGPT – unifies segmentation tasks into a generalist model through an in-context framework that supports different kinds of data.

Self-Improving Code LLMs Boost Performance Through Synthetic Data

By

–

09 April 2023 17h41

7/ Self-Improving Code LLMs- generates pseudo data from knowledge gained through pre-training & fine-tuning; adds the data to the training dataset for the next step; shows that different code generation frameworks can be improved in performance.

Baize: Open-Source Chat Model Fine-Tuned with LoRA

By

–

09 April 2023 17h41

5/ Baize – an open-source chat model fine-tuned with LoRA. Leverages 100K dialogs generated from ChatGPT chatting with itself; it releases the dialogs along with 7B, 13B, and 30B parameter models.

Segment Anything Model releases billion mask segmentation dataset

By

–

09 April 2023 17h41

1/ Segment Anything Model – a set of resources for image segmentation; releases the largest segmentation dataset with over 1B masks on 11M licensed images; the model’s zero-shot performance is competitive with or superior to fully supervised results.https://t.co/sJvKhmeECe
— DAIR.AI (@dair_ai) 9 avril 2023

1/ Segment Anything Model – a set of resources for image segmentation; releases the largest segmentation dataset with over 1B masks on 11M licensed images; the model’s zero-shot performance is competitive with or superior to fully supervised results.