Search papers, labs, and topics across Lattice.
East China Normal University
3
0
5
Intent-driven rule completion in IoT systems can boost completion rates by 43% while embedding safety and traceability throughout the process.
Current MLLM agents struggle to find GUI defects, but a new benchmark and evaluator reveals the critical bottleneck is detection, and surprisingly, simply integrating the evaluator's verifiers significantly boosts performance without retraining.
Why waste tokens on reasoning when you don't need to? Selective Chain-of-Thought cuts LLM inference costs by up to 47% in medical QA, with minimal accuracy loss.
