ReasonAudio benchmark reveals that state-of-the-art text-audio retrieval models struggle with reasoning tasks like negation and duration, and multimodal LLMs lose reasoning ability after contrastive fine-tuning.
hub
Audioclip: Extending clip to image, text and audio
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
hub tools
representative citing papers
KL regularization aligning model predictions with empirical transition patterns improves macro-F1 by 9-42% in next dialogue act prediction on German counselling data and transfers to other datasets.
Introduces the LDD task, ListenForge dataset built from five listening head generation methods, and MANet model that detects listening forgeries via motion inconsistencies guided by audio semantics.
TB-AVA uses text-mediated gated semantic modulation to enable efficient audio-visual alignment, achieving state-of-the-art results on AVE, AVS, and AVVP benchmarks.
LLMs identify mental states in dialogues well but mostly fail to forecast state-consistent future trajectories, except Gemini 3 Pro, with only weak overlap to human inferences.
SCALE disentangles emotion and cause representations in conversations and uses optimal transport for many-to-many global alignment, achieving SOTA on ECPEC benchmarks.
PRISM-CTG is the first large-scale foundation model for cardiotocography that uses multi-view self-supervised learning on unlabeled data to learn transferable representations, outperforming baselines on seven downstream tasks with external validation.
Qwen-Audio trains a unified model on diverse audio and tasks with hierarchical tags to enable strong zero-shot performance on audio understanding benchmarks and multi-turn audio chat.
The RADAR Challenge 2026 provides a multilingual benchmark for audio deepfake detection under media transformations and finds that robust performance remains an open problem.
WorldSpeech supplies 65k hours of multilingual aligned speech data across 76 languages and delivers 63.5% average relative WER reduction after fine-tuning ASR models on 11 typologically diverse languages.
GaborNet replaces sinc functions with Gabor filters in raw-audio neural networks and is tested for audio spoof detection with augmentations in RawNet2 and RawGAT-ST.
R-FLoRA combines Laplacian residual statistics with a frozen vision transformer via gated low-rank adapters, residual fusion, and contrastive alignment to achieve better accuracy and generalization than prior single-image face morphing attack detectors.
Qwen3.5-Omni scales an omnimodal model to hundreds of billions of parameters with 256k context, introduces ARIA for stable speech synthesis, and reports SOTA performance on 215 audio-visual benchmarks while adding multilingual and audio-visual coding capabilities.
Kimi-Audio is an open-source audio foundation model that achieves state-of-the-art results on speech recognition, audio understanding, question answering, and conversation after pre-training on more than 13 million hours of speech, sound, and music data.
Qwen2-Audio is an open-source audio-language model that outperforms prior systems such as Gemini-1.5-pro on audio-centric instruction-following benchmarks after simplified prompt-based pre-training and expanded data.
citing papers explorer
-
Audio Spoof Detection with GaborNet
GaborNet replaces sinc functions with Gabor filters in raw-audio neural networks and is tested for audio spoof detection with augmentations in RawNet2 and RawGAT-ST.