Track new results across pure and applied mathematics.
Search papers, labs, and topics across Lattice.
4
0
7
A novel bandit algorithm finds the optimal self-speculative decoding configuration for VLMs in real-time, boosting inference speed by up to 11% without retraining.
Congestion can flip the effectiveness of personalization strategies, revealing that SFT and ICL are not always straightforward choices for LLM users.
INFUSER outperforms a frozen 32B model with just an 8B co-evolving generator, showcasing the power of adaptive question generation in self-evolution.
Muon outperforms Adam and SGD by yielding features that are not only more robust but also transfer more effectively across tasks.