ReplayForge
⏪
Turn a screen recording into an evidence-linked bug report
Hector Henriquez PRO
SeedOfEvil
AI & ML interests
Real-world AI Applications & Integrations, Generative AI & Large Language Models, AI in Gaming & Interactive Experiences.
Recent Activity
published a Space about 1 hour ago
SeedOfEvil/ReplayForge updated a Space about 2 hours ago
SeedOfEvil/ReplayForge published a Space about 21 hours ago
SeedOfEvil/TRELLIS-2-Local-Privacy-MirrorOrganizations
None yet
published a
Space about 1 hour ago
updated a
Space about 2 hours ago
published a
Space about 21 hours ago
updated a
Space about 23 hours ago
Just please bring back the first AI.
9
#5 opened 1 day ago
by 
Matejstein
liked a
dataset 1 day ago
updated a
Space 1 day ago
published a
Space 1 day ago
Remove undisclosed prompt logging that secretly saves user prompts and images
2
#2 opened 2 days ago
by 
signsur4739379373
Bring back the early pro realism ai.
4
#1 opened 3 days ago
by Matejstein
updated a
Space 2 days ago
published a
Space 26 days ago
upvoted an article about 1 year ago
Article
Reachy Mini - The Open-Source Robot for Today's and Tomorrow's AI Builders
thomwolf, matthieu-lapeyre
• • 805 liked a
Space about 1 year ago
liked a
model about 1 year ago
liked 3
models over 1 year ago
reacted to clefourrier's post with 👍 over 1 year ago
Post
2765
Gemma3 family is out! Reading the tech report, and this section was really interesting to me from a methods/scientific fairness pov.
Instead of doing over-hyped comparisons, they clearly state that **results are reported in a setup which is advantageous to their models**.
(Which everybody does, but people usually don't say)
For a tech report, it makes a lot of sense to report model performance when used optimally!
On leaderboards on the other hand, comparison will be apples to apples, but in a potentially unoptimal way for a given model family (like some user interact sub-optimally with models)
Also contains a cool section (6) on training data memorization rate too! Important to see if your model will output the training data it has seen as such: always an issue for privacy/copyright/... but also very much for evaluation!
Because if your model knows its evals by heart, you're not testing for generalization.
Instead of doing over-hyped comparisons, they clearly state that **results are reported in a setup which is advantageous to their models**.
(Which everybody does, but people usually don't say)
For a tech report, it makes a lot of sense to report model performance when used optimally!
On leaderboards on the other hand, comparison will be apples to apples, but in a potentially unoptimal way for a given model family (like some user interact sub-optimally with models)
Also contains a cool section (6) on training data memorization rate too! Important to see if your model will output the training data it has seen as such: always an issue for privacy/copyright/... but also very much for evaluation!
Because if your model knows its evals by heart, you're not testing for generalization.
liked a
Space over 1 year ago