Applications · speech and audio papers, 2023–2026

Key findings

• Applications · speech and audio went from 117 papers in 2023 to 823 in 2026, roughly 7.0x in three years.
• ICLR publishes the most Applications · speech and audio work: 898 papers from 2023 to 2026, 52% of the four-venue total.
• The fastest-growing overlap is with Applications · computer vision: papers tagged with both rose from 0 in 2023 to 292 in 2026.
• Among the 10 topics under Application Domains, it ranks #3 by 2026 paper count; the fastest grower in the group, Applications · health medicine, is up 61x versus 7.0x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026(Strongly) Replicable Distribution Testers imply High Probability Distribution Testers
• ICML 2026(Be Cautious!) Bio-Foundation Models Are Not Yet Robust to Biologically Plausible Perturbations and ML Transformations
• ICLR 2026A Scene is Worth a Thousand Features: Feed-Forward Camera Localization from a Collection of Image Features
• CVPR 2026Globscope Toward a Global View of the Loss Landscape
• NeurIPS 2025AHa-Bench: Benchmarking Audio Hallucinations in Large Audio-Language Models
• ICML 2025A Variational Framework for Improving Naturalness in Generative Spoken Language Models
• ICLR 2025A Multiscale Frequency Domain Causal Framework for Enhanced Pathological Analysis
• CVPR 2025Blurred LiDAR for Sharper 3D: Robust Handheld 3D Scanning with Diffuse LiDAR and RGB
• NeurIPS 2024Acoustic Volume Rendering for Neural Impulse Response Fields
• ICML 2024$\bf{\Phi}_\textrm{Flow}$: Differentiable Simulations for PyTorch, TensorFlow and Jax
• ICLR 2024Dynamic Neural Response Tuning
• CVPR 2024DiVAS: Video and Audio Synchronization with Dynamic Frame Rates
• NeurIPS 2023[Re] Exploring the Role of Grammar and Word Choice in Bias Toward African American English (AAE) in Hate Speech Classification
• ICML 2023Accuracy on the Curve: On the Nonlinear Correlation of ML Performance Between Data Subpopulations
• ICLR 2023An efficient encoder-decoder architecture with top-down attention for speech separation
• CVPR 20233D Line Mapping Revisited
• NeurIPS 2026Approximate Matrix–Vectors Under a Bounded $\ell_1$ Assumption and Applications to Kernel Matrices
• ICML 2026AudioChat: Unified Audio Storytelling, Editing, and Understanding with Transfusion Forcing
• ICLR 2026A Stitch in Time Saves Nine: Proactive Self-Refinement for Language Models
• CVPR 2026Hierarchical Codec Diffusion for Video-to-Speech Generation

All Applications · speech and audio papers

72 papers, page 2 of 2

51. ICML 2025Otter: Generating Tests from Issues to Validate SWE Patches
52. ICML 2025Overcoming Non-monotonicity in Transducer-based Streaming Generation
53. ICML 2025OWLS: Scaling Laws for Multilingual Speech Recognition and Translation Models
54. ICML 2025PDE-Transformer: Efficient and Versatile Transformers for Physics Simulations
55. ICML 2025Playmate: Flexible Control of Portrait Animation via 3D-Implicit Space Guided Diffusion
56. ICML 2025Policy Filtration for RLHF to Mitigate Noise in Reward Models
57. ICML 2025Poly2Vec: Polymorphic Fourier-Based Encoding of Geospatial Objects for GeoAI Applications
58. ICML 2025QuanONet: Quantum Neural Operator with Application to Differential Equation
59. ICML 2025Reducing Tool Hallucination via Reliability Alignment
60. ICML 2025RestoreGrad: Signal Restoration Using Conditional Denoising Diffusion Models with Jointly Learned Prior
61. ICML 2025Rethink GraphODE Generalization within Coupled Dynamical System
62. ICML 2025ReverB-SNN: Reversing Bit of the Weight and Activation for Spiking Neural Networks
63. ICML 2025SING: Spatial Context in Large Language Model for Next-Gen Wearables
64. ICML 2025Sortformer: A Novel Approach for Permutation-Resolved Speaker Supervision in Speech-to-Text Systems
65. ICML 2025Steering Protein Language Models
66. ICML 2025Supervised Contrastive Learning from Weakly-Labeled Audio Segments for Musical Version Matching
67. ICML 2025The Brain's Bitter Lesson: Scaling Speech Decoding With Self-Supervised Learning
68. ICML 2025The dark side of the forces: assessing non-conservative force models for atomistic machine learning
69. ICML 2025TraceGrad: a Framework Learning Expressive SO(3)-equivariant Non-linear Representations for Electronic-Structure Hamiltonian Prediction
70. ICML 2025UniMate: A Unified Model for Mechanical Metamaterial Generation, Property Prediction, and Condition Confirmation
71. ICML 2025Voronoi-grid-based Pareto Front Learning and Its Application to Collaborative Federated Learning
72. ICML 2025XAttnMark: Learning Robust Audio Watermarking with Cross-Attention

FAQ

How many Applications · speech and audio papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

1,722 in total: 117 in 2023, 277 in 2024, 505 in 2025, 823 in 2026.

How many Applications · speech and audio papers are at NeurIPS 2026?

NeurIPS 2026 has 153 Applications · speech and audio papers, versus 98 the year before.

How many Applications · speech and audio papers are at ICLR 2026?

ICLR 2026 has 451 Applications · speech and audio papers, versus 292 the year before.

How many Applications · speech and audio papers are at ICML 2026?

ICML 2026 has 171 Applications · speech and audio papers, versus 72 the year before.

How many Applications · speech and audio papers are at CVPR 2026?

CVPR 2026 has 48 Applications · speech and audio papers, versus 43 the year before.

Which conference publishes the most Applications · speech and audio papers?

ICLR, with 898 papers from 2023 to 2026.