Track the frontier, from condensed matter to cosmology.
Search papers, labs, and topics across Lattice.
Affiliation:
7
0
13
3
ShieldVLA is proposed, a safety-aligned fine-tuning framework for VLA models based on Hamilton-Jacobi (HJ) reachability that reduces cumulative safety cost by 57% on average and improves task success rate by +0.13 over SafeVLA.
SNIPER achieves a remarkable CRAFT score of 0.98, ensuring near-perfect adherence to compression budgets while maintaining model performance.
Training agents in deep, evolving environments can dramatically enhance their performance, with a 9B model achieving a 30.6% accuracy increase through targeted design.
Hierarchical visual concepts learned through cascaded sparse autoencoders could revolutionize how we interpret and manipulate MLLM outputs.
Stop wasting time on manual LLM domain adaptation: AutoAdapt automates the process and boosts accuracy by 25% over existing AutoML methods.
A 4B parameter SLM can now rival frontier agent performance in complex tool-use environments, thanks to a novel reinforcement finetuning framework that teaches it how to strategically acquire context and execute actions.
Agentic LLMs can be taught to refuse harmful actions with up to 50% greater success, even zero-shot across diverse models and tasks, by explicitly learning when *not* to act.