Search papers, labs, and topics across Lattice.
2
0
5
3
Forget coarse sequence-level hacks: LenVM lets you precisely dial in token generation length, boosting a 7B model's length accuracy from 30.9 to 64.8 and crushing closed-source rivals.
Forget retraining: DOME efficiently adapts generative retrieval models to new documents by directly editing the model's docID mapping, achieving comparable performance with 40% less training.
