Search papers, labs, and topics across Lattice.
Graduate School of Data Science, Seoul National University
2
0
3
Unconstrained decoding in dLLMs can lead to a staggering 90% collapse into answer-only outputs, highlighting a critical flaw in reasoning capabilities.
Hallucinations aren't always due to a lack of knowledge; larger LLMs often "know" the right answer but fail to commit to it, scattering probability across multiple surface forms.
