PaliGemma 2: A Family of Versatile VLMs for Transfer — mentioned by xnx

Thu, 05 Dec 2024 20:00:39 GMT

Even more technical detail here: https://arxiv.org/html/2412.03555v1

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

PaliGemma 2: A Family of Versatile VLMs for Transfer — mentioned by sigmar

Thu, 05 Dec 2024 19:43:15 GMT

It is probably hard to come up with good benchmarks for VLMs like this, but I feel like the "Non entailment sentences" benchmark seems ill-suited. The examples for sentences that were non entailment included[1]: "There is a pile of horse manure in front of the horse." Which is true if you mean "in front of the [photosubject] from the perspective of the camera," but I think they marked it as non entailment because the pile is not in front of the horse's face(?) [1] page 20 https://arxiv.org/pdf/2412.03555

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

PaliGemma 2: A Family of Versatile VLMs for Transfer · HackerPapers discussions

PaliGemma 2: A Family of Versatile VLMs for Transfer — mentioned by xnx

PaliGemma 2: A Family of Versatile VLMs for Transfer — mentioned by sigmar