All papers¶
Every lesson on the site, newest first. Difficulty is rated for a reader who is intermediate in deep learning and strongest in speech AI; see how to read this site for the scale.
| Paper | Type | Difficulty | Tags | Added |
|---|---|---|---|---|
| Ming-3.0-tts-tiny Ming-3.0-tts-tiny (open-weight model release, no paper) (no paper: model release) |
Model release (no paper) | 3/5 Stretch | Speech, TTS, Generative models, Multilingual | 2026-10-20 |
| DeepInfer DeepInfer: High-performance, large-scale inference framework for DeepSeek models (open-source release, no paper) (no paper: model release) |
System or infrastructure | 3/5 Stretch | LLMs, Systems, Efficiency | 2026-10-20 |
| ByteRobust Robust LLM Training Infrastructure at ByteDance (arXiv 2509.16293) |
System or infrastructure | 2/5 Comfortable | LLMs, Systems, Robustness, Efficiency | 2026-10-20 |
| Mind the Accent Gap Mind the Accent Gap: British Accent Robustness in Speech-Driven Financial Voice Assistants (arXiv 2610.06587) |
Benchmark or dataset | 1/5 Warm-up | Speech, ASR, Benchmarks, Robustness, Agents | 2026-10-15 |
| Llama 4 Herd The Llama 4 Herd: Architecture, Training, Evaluation, and Deployment Notes (arXiv 2601.11659) |
Survey | 2/5 Comfortable | LLMs, Multimodal, Efficiency, Distillation | 2026-10-15 |
| OmniVoice OmniVoice: Towards Omnilingual Zero-Shot Text-to-Speech with Diffusion Language Models (arXiv 2604.00688) |
New method or architecture | 3/5 Stretch | Speech, TTS, Multilingual, Generative models, Diffusion | 2026-10-14 |
| AudioLBM Audio Super-Resolution with Latent Bridge Models (arXiv 2509.17609) |
New method or architecture | 3/5 Stretch | Speech, Audio, Generative models, Diffusion | 2026-10-14 |
| AudioSR AudioSR: Versatile Audio Super-resolution at Scale (arXiv 2309.07314) |
New method or architecture | 2/5 Comfortable | Speech, Audio, Generative models, Diffusion | 2026-10-14 |
| Breeze TTS 2 Breeze TTS 2 (open-weight model release, no paper) (no paper: model release) |
Model release (no paper) | 2/5 Comfortable | Speech, TTS, Generative models, Efficiency | 2026-10-11 |
| OPD Scaling Laws Scaling Properties of Same-Family On-Policy Distillation (arXiv 2609.32722) |
Empirical study | 3/5 Stretch | LLMs, Distillation, Scaling laws, Reinforcement learning, Reasoning | 2026-10-04 |
| ICL for Robots In-Context Learning for Robots: Methods and Applications (arXiv 2609.36012) |
Survey | 3/5 Stretch | Robotics, In-context learning, Agents, Multimodal | 2026-10-02 |
| Duplex-MPE Duplex-MPE: Benchmarking Multi-Party Interaction in Full-Duplex Dialogue (arXiv 2609.31948) |
Benchmark or dataset | 2/5 Comfortable | Speech, Benchmarks, Evaluation, Datasets | 2026-09-30 |
| SchrodingerRepo Schrödinger's Code Repository: Have LLMs Learned SWE-bench or Memorized It? (arXiv 2609.27891) |
Benchmark or dataset | 2/5 Comfortable | LLMs, Agents, Code, Benchmarks, Evaluation, Robustness | 2026-09-30 |
| Disaggregated Quantization Disaggregated Quantization: Specializing LLM Prefill and Decode (arXiv 2609.26333) |
New method or architecture | 3/5 Stretch | LLMs, Efficiency, Systems | 2026-09-30 |