Search papers, labs, and topics across Lattice.
3
53
6
28
Forget passively analyzing model outputs – this new attack actively *trains* the model to regurgitate specific texts, revealing its training data with surprising accuracy.
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.
RewardBench 2 exposes a stark reality check for reward models: they struggle significantly on new, human-generated prompts, yet this difficulty is surprisingly predictive of their actual usefulness in downstream tasks.
