Xiu Li

Verifiable temporal grounding in video forensics can drastically improve the detection of AI-generated content, outperforming traditional methods reliant on coarse supervision.

Temporal reconstruction errors can be harnessed to significantly boost video quality, eliminating the need for expensive human annotations in preference optimization.

Extracting interaction cues from a frozen video model enables robots to achieve up to 90.6% success in manipulation tasks without costly rollout processes.

PhysEditWorld reveals that explicit control over physical parameters can transform how game world models interact with their environments, leading to more realistic and manipulable simulations.

SVP-IL boosts success rates on ambiguous language tasks by over 60% with minimal training data, revolutionizing data efficiency in robotic manipulation.

Bridging the perception-reasoning gap in visual planning, MGSD boosts model performance by over 19% while relying solely on visual inference during deployment.

By pretraining a VLA model with goal-conditioned RL, PRTS learns to reason about goal reachability, leading to substantial gains in long-horizon robotic tasks and zero-shot generalization.

Simply plugging in RoTE, a lightweight temporal embedding module, can boost existing Transformer-based sequential recommendation models by over 20% on standard benchmarks.

Achieve state-of-the-art real-world image dehazing by jointly reconstructing the clear scene and scattering variables, even with non-uniform haze and complex lighting.

Existing image editing models fall short when it comes to precise spatial manipulations, but a new benchmark and dataset reveal the path to closing the gap.