Search papers, labs, and topics across Lattice.
4
1
9
3
Minor architectural tweaks can lead to a staggering 47% drop in long context performance, challenging assumptions about model design.
Poisoning pretraining data through public discussion interfaces poses a significant threat, with the potential for undetectable harmful behaviors in language models.
VLMs that ace math problems still flunk at understanding *how* students go wrong, highlighting a critical gap for AI in education.
By reusing existing data mixture ratios and only recomputing for affected domains, Olmix slashes compute costs by 74% without sacrificing downstream task performance during iterative LM development.
