Search papers, labs, and topics across Lattice.
School of Automation Science and Engineering, Xi鈥檃n Jiaotong University, China
2
0
5
Over-reliance on agentic decomposition can actually *hurt* audio understanding when a strong audio frontend already provides sufficient information, highlighting the importance of conditional evidence acquisition.
Forget expensive audio-text data collection: TASU2 lets you dial in the perfect amount of noise for training your speech LLM, all from text.
