Tag
This paper proposes Joint Speech-Text Interleaved Pretraining (JSTIP), a pretraining strategy that constructs word-level and segment-level interleaved speech-text sequences to improve ASR entity accuracy and reduce the modality gap between speech and text, showing competitive performance on domain adaptation and zero-shot speech question answering.
Chronicle is a 324M-parameter decoder-only transformer pretrained from scratch on both natural language and time series, achieving competitive performance on NLU and time series classification tasks, and setting new state-of-the-art for frozen-embedding time series classification on UCR/UEA datasets.
This paper introduces ICRL, a framework that jointly trains a solver and critic with reinforcement learning to internalize critique guidance, enabling the solver to improve without external critique. It uses distribution calibration and role-wise group advantage estimation, achieving 6-7 point gains over GRPO on agentic and mathematical reasoning tasks.