Search papers, labs, and topics across Lattice.
Affiliation:
3
0
4
0
Idiolectal paraphrasing transforms how models learn reasoning by allowing them to express complex thoughts in their own unique language, leading to significant performance gains.
Context-Calibrated DPO can cut object hallucination in MLLMs by 36% while maintaining reasoning performance, revealing a critical gap in how existing methods leverage context.
Iterative refinement of search queries in a continuous latent space leads to a dramatic increase in video retrieval accuracy and reasoning efficiency.
