Search papers, labs, and topics across Lattice.
Xi鈥檃n Jiaotong University
5
0
8
11
DAIS boosts final-answer accuracy by up to 5.6% by leveraging dependency-aware intermediate QA records that enhance reasoning coherence.
Adding depthwise convolutions to Transformers can boost accuracy on downstream tasks while barely increasing model size.
StatLUT achieves artifact-free photorealistic style transfer by decoupling color and structure, outperforming state-of-the-art methods in visual fidelity.
Harness VLA boosts the performance of frozen VLA models by 38.6 percentage points on challenging manipulation tasks without the need for finetuning.
Agent-as-a-Judge can outperform LLM-as-a-Judge in complex environments, but still struggles to reliably verify agent behavior, revealing a critical gap in current LLM-based agent evaluation.
