paper-with-me

홈 › Papers

When Vision Speaks for Sound

2026-05-13 · Xiaofei Wen, Wenjie Jacky Mo, Xingyu Fu, Rui Cai, Tinghui Zhu, Wendi Li, Yanan Xie, Muhao Chen, Peng Qi arxiv

Despite rapid progress in video-capable MLLMs, we find that their apparent audio understanding in videos is often vision-driven: models rely on visual cues to infer or hallucinate acoustic information, rather than verifying the audio stream. This issue appears across both state-of-the-art open-source omni models and leading closed-source models from providers such as Google and OpenAI. We characterize this failure mode as an audio-visual Clever Hans effect, in which models appear (falsely) audio-grounded, but actually exploit visual-acoustic correlations without verifying whether the audio and visual streams are truly aligned. To systematically study this behavior, we introduce Thud, an intervention-driven probing framework based on three counterfactual audio edits: Shift, which tests temporal synchronization; Mute, which tests sound existence; and Swap, which tests audio-visual consistency. Beyond diagnosis, we further study a two-stage alignment recipe: intervention-derived preference pairs teach audio verification, while event-level general video preferences regularize the model against over-specialization. Our best 10K-sample recipe improves average performance across the three intervention dimensions by 28 percentage points, while slightly improving performance on general video and audio-visual QA benchmarks.

📄 PDF Abstract BibTeX arXiv:2605.16403

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

SpeakStream: Streaming Text-to-Speech with Interleaved Data

2025-05-25 · Richard He Bai, Zijin Gu, Tatiana Likhomanenko, Navdeep Jaitly

The latency bottleneck of traditional text-to-speech (TTS) systems fundamentally hinders the potential of streaming large language models (LLMs) in conversational AI. These TTS systems, typically trained and inferenced o…

Decodertext-to-speechText to Speech

Should you use a probabilistic duration model in TTS? Probably! Especially for spontaneous speech

2024-06-08 · Shivam Mehta, Harm Lameris, Rajiv Punmiya, Jonas Beskow 외

Converting input symbols to output audio in TTS requires modelling the durations of speech sounds. Leading non-autoregressive (NAR) TTS models treat duration modelling as a regression problem. The same utterance is then …

regression

E-BERT: Efficient-Yet-Effective Entity Embeddings for BERT

2019-11-09 · Findings of the Association for Computational Linguistics 2020 · Nina Poerner, Ulli Waltinger, Hinrich Schütze

We present a novel way of injecting factual knowledge about entities into the pretrained BERT model (Devlin et al., 2019): We align Wikipedia2Vec entity vectors (Yamada et al., 2016) with BERT's native wordpiece vector s…

Entity EmbeddingsEntity LinkingLanguage ModellingQuestion Answering+1

TokDrift: When LLM Speaks in Subwords but Code Speaks in Grammar

2025-10-16 · Yinxi Li, Yuntian Deng, Pengyu Nie arxiv

Large language models (LLMs) for code rely on subword tokenizers, such as byte-pair encoding (BPE), learned from mixed natural language text and programming language code but driven by statistics rather than grammar. As …

MOSS-VL Technical Report

2026-08-15 · Pengyu Wang, Chenkun Tan, Shaojun Zhou, Qirui Zhou 외 hf

We present MOSS-VL, an open vision-language model family that treats real-time interaction -- perceiving while it speaks -- as a first-class capability. It is co-designed across the stack: the language decoder attends to…