CARES: A Controlled Synthetic Benchmark of Speaker Reactions to Sound
cs.SD, cs.LG
Submitted: 2026-10-07
Updated: 2026-10-07
Code: https://github.com/marcelgibier/CARES
Project page: https://qwenlm.github.io
Terminology
Sources
- Audio Flamingo Next: Next-Generation Open Audio-Language Models for Speech, Sound, and Music
- Qwen3-Omni Technical Report
- Kimi-Audio Technical Report
- MiMo-Audio: Audio Language Models are Few-Shot Learners
- MiDashengLM: Efficient Audio Understanding with General Audio Captions
- Spoken DialogSum: An Emotion-Rich Conversational Dataset for Spoken Dialogue Summarization
- Generating Synthetic Doctor-Patient Conversations for Long-form Audio Summarization
- Can Large Audio Language Models Understand Audio Well? Speech, Scene and Events Understanding Benchmark for LALMs
- From Sounds to Scenes: A Benchmark for Evaluating Context-Aware Auditory Scene Understanding in Large Audio Language Models
- Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity
- Scaling Speech Technology to 1,000+ Languages
- C-Pack: Packed Resources For General Chinese Embeddings
Related papers
- Few-Shot Open-Set Audio Classification via Transductive Prototype Refinement and Class Logit Enhancement
- Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems
- Remix the Timbre: Diffusion-Based Style Transfer Across Polyphonic Stems
- AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
- SoundWeaver: Compositional Warm-Starting for Text-to-Audio Diffusion Serving
- WASIL: In-the-Wild Arabic Spoken Interactions with LLMs