Yang Gao

PL-HCL transforms the detection of inconsistencies in Agent Skills, boosting accuracy from 45% to nearly 90% and enhancing user trust in skill selection.

VLMs can achieve substantial improvements in reasoning performance using only unlabeled data through a novel self-reflective training framework inspired by human cognition.

CI-MSE dramatically improves the correlation between offline validation and real-world performance, making it a game-changer for robot policy evaluation.

Automated judges for LLM jailbreak assessments are unreliable, with significant discrepancies in ASR outcomes based on the judge type used.

FTP-1 not only excels on familiar tactile sensors but also achieves unprecedented success on unseen setups, redefining the potential for cross-sensor generalization in robotic manipulation.

Simulated evaluations can mislead policy rankings, but our findings reveal how to better align simulations with real-world performance.

MOSS-Audio achieves state-of-the-art performance in audio understanding tasks by effectively integrating temporal cues and deep acoustic features, setting a new benchmark for audio-language models.

Arithmetic errors in LLMs stem from geometric slippages in internal computations, revealing a surprising fragility in their handling of fundamental math.

Modeling dynamic 3D scenes gets a serious upgrade: DeGO's deformable Gaussians and foundation model distillation boost occupancy prediction accuracy by 13.5% on human-centric instances.

Forget scaling laws – this zero-shot navigation agent beats million-sample trained models by structurally unifying language, vision, and robot actions within the reasoning capabilities of pre-trained MLLMs.