Search papers, labs, and topics across Lattice.
7
0
8
4
Hallucinations in LVLMs can be cut by over 43% without sacrificing grounded object coverage, thanks to a novel verifier-guided approach.
TIGA can generate high-quality, detector-evasive images on-the-fly, bypassing the need for source images or model retraining, which could revolutionize evasion strategies against AIGC detectors.
Watermarking agent memories is now possible without performance degradation or reliance on logs, enabling snapshot-only attribution even after memory migration or leakage.
LVLMs are better at spotting their own mistakes than generating correct answers in the first place, and this self-awareness can be exploited to reduce hallucinations.
Forget external teachers – the best way to boost your RL model's performance is to learn from its future self.
EasyVideoR1 achieves a 1.47 times throughput improvement in video understanding tasks by eliminating redundant video decoding and leveraging a comprehensive task-aware reward system.
Fake news in short videos often betrays itself through subtle inconsistencies between text, visuals, and audio, a weakness MAGIC3 exploits to achieve VLM-level accuracy at a fraction of the cost.
