Symbiotic Architecture for Post-Hoc Audio Extension of Frozen Language Models

arXiv:2609.30784 · cs.SD, cs.CL, eess.AS · Submitted 2026-09-25 · Read on arXiv

cs.SD, cs.CL, eess.AS

Submitted: 2026-09-25

Updated: 2026-09-25

Terminology

• KV cache — 12× in this paper · explained in 3 episodes
• Catastrophic Forgetting — 8× in this paper · explained in 5 episodes
• Embeddings — 6× in this paper · explained in 2 episodes
• Fine-Tuning — 6× in this paper · explained in 5 episodes
• Transformer — 4× in this paper · explained in 2 episodes
• Robustness — 3× in this paper · explained in 5 episodes
• Accuracy — 2× in this paper · explained in 1 episode
• Adaptation — 2× in this paper · explained in 1 episode
• Alignment — 2× in this paper · explained in 1 episode
• Calibration — 2× in this paper · explained in 7 episodes
• Benchmark — 1× in this paper · explained in 2 episodes
• Core — 1× in this paper · explained in 2 episodes
• Decoupling — 1× in this paper · explained in 2 episodes
• Mixture of Experts — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• Sparks of Large Audio Models: A Survey and Outlook
• Gemma 4 Technical Report
• Qwen3 Technical Report
• Multi-Domain Audio Question Answering Benchmark Toward Acoustic Content Reasoning
• Pointer Sentinel Mixture Models
• Training Verifiers to Solve Math Word Problems

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
• 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
• Efficient Audiovisual Speech Processing via MUTUD: Multimodal Training and Unimodal Deployment