Search papers, labs, and topics across Lattice.
7
0
14
SearchEyes achieves state-of-the-art performance in multimodal search by unifying training data, environments, and rewards into a cohesive simulated world.
NA-LoRA reveals that adapting low-rank updates with a focus on gate channel responsiveness can significantly enhance model fine-tuning performance.
Leveraging historical solving traces transforms software engineering agents into self-evolving entities, achieving a 50.40% success rate on SWE-bench Verified after just three iterations.
Agents can spontaneously develop self-referential communication systems, challenging traditional views on language emergence in AI.
LLMs struggle to predict prosecutorial decisions, highlighting a critical blind spot in legal AI's ability to assess criminal liability beyond formally indicted cases.
Achieve 50% parameter reduction in LLaMA-2-7B with minimal performance loss and no fine-tuning, thanks to a new global gating-based structured pruning method.
Optimizing multilingual training? Shapley values reveal the hidden cross-lingual transfer effects that current scaling laws miss, leading to better language mixture ratios.
