Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs · HackerPapers discussions

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs — mentioned by fpgaminer

Fri, 07 Feb 2025 23:13:38 GMT

A lot of problems jump out to me with this article, particularly with the explanation of multi-modal LLMs. I'll say that I _do_ agree with the thrust of the article. Don't trust LLMs. But they probably should have argued legitimate issues with VLM based OCR, rather than try to talk about how VLMs are somehow fundamentally flawed or something. > LLMs process images through high-dimensional embeddings, essentially creating abstract representations that prioritize semantic understanding over precise character recognition. This isn't true. CLIP and its derivatives don't prioritize semantic…

Paper and discussion history. A mention is not an endorsement.

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs — mentioned by tucnak

Mon, 27 Jan 2025 20:28:09 GMT

I'm still not sure why they keep LeCun at Facebook; his single most-cited contribution to the field in 2024 has been with NYU[0], not Facebook. What is his role at Facebook exactly, has he explained it? I recall him making all the wrong predictions in 2023 what's changed? Chollet is similarly a mystery to me; it feels like these guys were busy riffing CNN's when the Transformer came about and since then have been trying to far-out in search of gold. [0]: https://arxiv.org/abs/2406.16860

Paper and discussion history. A mention is not an endorsement.

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs — mentioned by ellisbrown

Tue, 25 Jun 2024 20:46:16 GMT

arXiv: https://arxiv.org/abs/2406.16860 Website: https://cambrian-mllm.github.io Code: https://github.com/cambrian-mllm/cambrian Checkpoints: https://huggingface.co/collections/nyu-visionx/cambrian-1-mo... Data: https://huggingface.co/collections/nyu-visionx/cambrian-data... Benchmark: https://huggingface.co/datasets/nyu-visionx/CV-Bench

Paper and discussion history. A mention is not an endorsement.