arXiv subjects

Ming-Hao Hsu

Publications and source records attributed to Ming-Hao Hsu.

10 recordsLinked to original sources

arXiv2026

Entity Binding Failures in Speech LLM Reasoning: Diagnosis and Chain-of-Thought Intervention

Speech Large Language Models (SLLMs) underperform their text counterparts on complex reasoning. We reveal that this gap is not a uniform cognitive deficit. Evaluating two architecturally diverse SLLMs, we show speech-to-text (S2T) matches or exceeds text-to-text (T2T) on spatial, syntactic, and factual tasks. Yet on logical tasks requiring entity tracking, S2T accuracy collapses to chance. We diagnose this as an entity binding failure: continuous speech features blur precise entity-property associations during implicit reasoning. To validate this diagnosis, we introduce Entity-Aware Chain-of-Thought (EA-CoT), a lightweight inference-time intervention forcing SLLMs to enumerate entities and bind them to claims before reasoning. EA-CoT bridges the gap, even when spoken names are misrecognized, yielding up to a 24.4 percentage-point accuracy gain. Ablations confirm the gains stem from explicit semantic binding, reframing the gap as an elicitation failure rather than a missing capability.

cs.CL↗

arXiv2026

Is Text All You Need? Text as a Universal Information Bottleneck for Speech LLMs

Large language models (LLMs) provide a powerful reasoning backbone for speech understanding, but integrating continuous acoustic signals into a frozen LLM remains challenging. Existing speech-to-LLM interfaces typically operate at two extremes: either enforcing near-discrete token alignment, which benefits transcription but loses paralinguistic information, or learning unconstrained continuous representations, which can drift away from the LLM's input space and degrade autoregressive decoding. In this work, we propose Convex Gate (C-Gate), a speech-to-LLM bridge that constrains all speech representations to lie within the LLM's input embedding manifold with an architectural convex-hull constraint. Concretely, each frame is represented as a convex combination of token embeddings, ensuring compatibility with the pretrained LLM while preserving continuous expressivity. Across automatic speech recognition (ASR) and emotion recognition, C-Gate achieves strong joint performance, improving LibriSpeech WER by up to 48.7% relative while matching or exceeding single-task emotion accuracy. Beyond performance, our analysis reveals a key insight: information is not carried by discrete token identities, but by time-resolved trajectories in the embedding space. Causal interventions confirm that both the trajectory structure and alignment to the pretrained embedding manifold are critical for performance. These results suggest that geometry, rather than token discreteness, is the fundamental design factor in speech-to-LLM interfaces, and provide a controlled regime for studying multimodal integration in frozen LLMs. We release the checkpoint, per-sample outputs, mechanism dumps, and intervention suite for replication.

cs.CL↗

arXiv2026

Anatomy of the Modality Gap: Dissecting the Internal States of End-to-End Speech LLMs

Recent advancements in Large Speech-Language Models have significantly bridged the gap between acoustic signals and linguistic understanding. However, a persistent performance disparity remains in speech-based input tasks compared to direct text inference. In this paper, we investigate the dynamic roots of this modality gap beyond static geometric alignment, analyzing how speech and text representations evolve layer-by-layer. We evaluate four open-weight end-to-end models on SpeechMMLU and VoiceBench BBH. Using cross-layer CKA analysis with speech-text token alignment, we find that speech representations exhibit a broad cross-layer alignment band, attributable to the redundant nature of speech where semantic content spans multiple frames. We show that these alignment patterns are structurally stable across different analysis configurations. Crucially, simple statistical calibration is insufficient and can be detrimental when applied at the input layer, indicating that the modality gap is not a mere distribution shift. Overall, our results suggest that the bottleneck lies in condensing redundant speech into stable late-layer decisions, motivating future solutions that operate at the token or temporal granularity instead of feature-level matching.

cs.CL↗

arXiv2026

Linear Script Representations in Speech Foundation Models Enable Zero-Shot Transliteration

Multilingual speech foundation models such as Whisper are trained on web-scale data, where data for each language consists of a myriad of regional varieties. However, different regional varieties often employ different scripts to write the same language, rendering speech recognition output also subject to non-determinism in the output script. To mitigate this problem, we show that script is linearly encoded in the activation space of multilingual speech models, and that modifying activations at inference time enables direct control over output script. We find the addition of such script vectors to activations at test time can induce script change even in unconventional language-script pairings (e.g. Italian in Cyrillic and Japanese in Latin script). We apply this approach to inducing post-hoc control over the script of speech recognition output, where we observe competitive performance across all model sizes of Whisper.

cs.CL↗

arXiv2025

TASLA: Text-Aligned Speech Tokens with Multiple Layer-Aggregation

We propose Text-Aligned Speech Tokens with Multiple Layer-Aggregation (TASLA), which is a text-aligned speech tokenization framework that aims to address the problem that under a low-frame-rate and text-aligned regime, single-source speech tokens may lose acoustic details during reconstruction. On the other hand, this paper further explains how different encoder layers collaborate to capture comprehensive acoustic features for tokenization. Previous work, TASTE, proposed the text-aligned speech tokenization framework, which is a LM-friendly architecture, but struggles to capture acoustic details. We address this trade-off with two components: Multi-Layer Dynamic Attention (MLDA), which lets each text position adaptively mix shallow/deep features from a frozen speech encoder, and Finite Scalar Quantization (FSQ), a simple per-dimension discretization with smooth optimization. At about 2.62 Hz (tokens/s), TASLA consistently improves prosody and achieves competitive quality over TASTE on in-domain (LibriSpeech) and OOD (EXPRESSO, Voxceleb) sets. We further demonstrate that dynamic layer mixing is correlated with spectral flux and explains why MLDA preserves prosody under a low frame rate with extreme feature compression.

cs.SD↗

arXiv2025

SMILE: Speech Meta In-Context Learning for Low-Resource Language Automatic Speech Recognition

Automatic Speech Recognition (ASR) models demonstrate outstanding performance on high-resource languages but face significant challenges when applied to low-resource languages due to limited training data and insufficient cross-lingual generalization. Existing adaptation strategies, such as shallow fusion, data augmentation, and direct fine-tuning, either rely on external resources, suffer computational inefficiencies, or fail in test-time adaptation scenarios. To address these limitations, we introduce Speech Meta In-Context LEarning (SMILE), an innovative framework that combines meta-learning with speech in-context learning (SICL). SMILE leverages meta-training from high-resource languages to enable robust, few-shot generalization to low-resource languages without explicit fine-tuning on the target domain. Extensive experiments on the ML-SUPERB benchmark show that SMILE consistently outperforms baseline methods, significantly reducing character and word error rates in training-free few-shot multilingual ASR tasks.

eess.AS↗

arXiv2024

GSQA: An End-to-End Model for Generative Spoken Question Answering

In recent advancements in spoken question answering (QA), end-to-end models have made significant strides. However, previous research has primarily focused on extractive span selection. While this extractive-based approach is effective when answers are present directly within the input, it falls short in addressing abstractive questions, where answers are not directly extracted but inferred from the given information. To bridge this gap, we introduce the first end-to-end Generative Spoken Question Answering (GSQA) model that empowers the system to engage in abstractive reasoning. The challenge in training our GSQA model lies in the absence of a spoken abstractive QA dataset. We propose using text models for initialization and leveraging the extractive QA dataset to transfer knowledge from the text generative model to the spoken generative model. Experimental results indicate that our model surpasses the previous extractive model by 3% on extractive QA datasets. Furthermore, the GSQA model has only been fine-tuned on the spoken extractive QA dataset. Despite not having seen any spoken abstractive QA data, it can still closely match the performance of the cascade model. In conclusion, our GSQA model shows the potential to generalize to a broad spectrum of questions, thus further expanding the spoken question answering capabilities of abstractive QA. Our code is available at https://voidful.github.io/GSQA

cs.CL↗

arXiv2024

Exploring In-Context Learning of Textless Speech Language Model for Speech Classification Tasks

Ever since the development of GPT-3 in the natural language processing (NLP) field, in-context learning (ICL) has played an essential role in utilizing large language models (LLMs). By presenting the LM utterance-label demonstrations at the input, the LM can accomplish few-shot learning without relying on gradient descent or requiring explicit modification of its parameters. This enables the LM to perform various downstream tasks in a black-box manner. Despite the success of ICL in NLP, little work is exploring the possibility of ICL in speech processing. This study is the first work exploring ICL for speech classification tasks with textless speech LM. We first show that the current speech LM lacks the ICL capability. We then perform warmup training on the speech LM, equipping the LM with demonstration learning capability. This paper explores and proposes the first speech LM capable of performing unseen classification tasks in an ICL manner.

eess.AS↗

arXiv2024

Diffusion Model-Augmented Behavioral Cloning

Imitation learning addresses the challenge of learning by observing an expert's demonstrations without access to reward signals from environments. Most existing imitation learning methods that do not require interacting with environments either model the expert distribution as the conditional probability p(a|s) (e.g., behavioral cloning, BC) or the joint probability p(s, a). Despite the simplicity of modeling the conditional probability with BC, it usually struggles with generalization. While modeling the joint probability can improve generalization performance, the inference procedure is often time-consuming, and the model can suffer from manifold overfitting. This work proposes an imitation learning framework that benefits from modeling both the conditional and joint probability of the expert distribution. Our proposed Diffusion Model-Augmented Behavioral Cloning (DBC) employs a diffusion model trained to model expert behaviors and learns a policy to optimize both the BC loss (conditional) and our proposed diffusion model loss (joint). DBC outperforms baselines in various continuous control tasks in navigation, robot arm manipulation, dexterous manipulation, and locomotion. We design additional experiments to verify the limitations of modeling either the conditional probability or the joint probability of the expert distribution, as well as compare different generative models. Ablation studies justify the effectiveness of our design choices.

cs.LG↗

arXiv2022

Controllable User Dialogue Act Augmentation for Dialogue State Tracking

Prior work has demonstrated that data augmentation is useful for improving dialogue state tracking. However, there are many types of user utterances, while the prior method only considered the simplest one for augmentation, raising the concern about poor generalization capability. In order to better cover diverse dialogue acts and control the generation quality, this paper proposes controllable user dialogue act augmentation (CUDA-DST) to augment user utterances with diverse behaviors. With the augmented data, different state trackers gain improvement and show better robustness, achieving the state-of-the-art performance on MultiWOZ 2.1

cs.CL↗