Wenya Wang

A simple post-hoc geometric calibration method recovers lost out-of-domain capabilities and boosts in-domain performance after SFT, even on large, diverse models.

Achieve 25x speedup in World Action Models with <1% performance drop by enabling one-step action generation and asynchronous expert execution.

LLM agents can now dynamically curate multimodal memory on-demand, achieving optimal performance-cost-latency trade-offs by intelligently choosing between pre-computed and runtime-generated evidence.

VideoLMs can significantly enhance their temporal understanding by leveraging a counterfactual approach that reveals hidden dynamics in video data.

Traditional interpretability fails to predict task-critical mechanisms before training, but this new framework bridges that gap, enabling more effective fine-tuning strategies.

Pruning 77.8% of visual tokens without losing performance could revolutionize the efficiency of multimodal large language models.

The fragmented field of world modeling can now be unified under a "levels x laws" taxonomy, revealing critical gaps in autonomous model revision and decision-centric evaluation.