Skip to content

Author

Sheng-Bo Cai

6 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Nov 2026

Speech Tokenizers are Vulnerable: Transferable Semantic Attack and Robust Tokenizer

Speech tokenizers serve as speech front ends, encoding input speech signals into representations for downstream models, such as automatic speech recognition (ASR) systems and large audio-language models (LALMs). In this paper, we find that speech tokenizers are fragile at the semantic level: small perturbations can dis...

Zhi-Sheng Zhang, Yi-Fan Mi, Yi-Xuan Zhou et al. · 0 citations
Preprint Oct 2026

Jarvis: Coordination Deserves a Brain of Its Own in Agentic Interactive Models

Interactive models increasingly hand complex reasoning and tool use to a slow brain in the background, leaving a realtime front model free to sustain full-duplex dialogue. Yet the user may refine, cancel or replace a request while its task runs, so a plan or result that made sense a moment ago goes stale. Each new utte...

Yu-Xiang Wang, Sheng-Bo Cai, Zhi-Xian Zhao et al. · 0 citations
#machine learning Preprint Oct 2026

AURAL: Adaptive Latent Reasoning with Joint Chunk for Speech Language Models

AURAL is introduced, which models a distribution over multiple plausible reasoning continuations in latent space and jointly predicts chunks of future states to reduce sequential forward passes and reasoning latency and shows that harder questions elicit more latent reasoning steps.

Yu-Xiang Wang, Kun-Yu Feng, Yuan-Cheng Wang et al. · 4 citations

Thinking in Depth, Speaking Directly: Recurrent Latent Reasoning for Paralinguistically Grounded Spoken Dialogue

LoopSLM builds on looped Transformers for latent reasoning, reusing a decoder block to refine hidden states with acoustic grounding at every pass, and further narrows the perception-reasoning gap by separating learning to reason from learning to respond, enabling direct inference without CoT.

Sheng-Bo Cai, Yu-Xiang Wang, Jing-Ran Xie et al. · 2 citations
#machine learning Preprint Sep 2026

EvoAudio: Recursive Self-Improvement for Audio Understanding

EvoAudio is proposed, a recursive self-improvement system for audio understanding that is the first to evolve the model, waveforms, questions, and difficulty in one closed loop.

Yu-Xiang Wang, Sheng-Bo Cai, Ying Shen et al. · 2 citations · ⚡2

ParaBridge: Bridging Paralinguistic Perception and Dialogue Behavior in Speech Language Models

ParaBridge is proposed, an on-policy self-distillation method that turns a brittle inference-time scaffold into stable model behavior and generalizes to unseen paralinguistic cues, transfers from safety-oriented training to empathy-oriented dialogue, and works on a different SLM backbone.

Yuxiang Wang, Qin-Ke Ni, Sheng-Bo Cai et al. · 2 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.