Search papers, labs, and topics across Lattice.
3
0
7
E^3RL not only overcomes the autoregressive curse but also enhances LLMs' reasoning capabilities, achieving up to 6.514% better performance than previous state-of-the-art models.
Forget specialized architectures: StepAudio 2.5 proves a single audio-language foundation, shaped by RLHF, can dominate ASR, TTS, and real-time dialogue simultaneously.
LLMs can significantly boost their emotional intelligence simply by role-playing conversations with themselves, iteratively refining their ability to both recognize and express emotions.
