Yuma Ichikawa

4 papers · 2024–2026 · 4 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (14) 🏆 Keyword Champion (2)

Venues

ACL (1) AISTATS (1) ICLR (1) NeurIPS (1)

Top co-authors

Yamato Arai (1) Akira Sakai (1) Koji Hukushima (1) Yuzi Kanazawa (1) Takumi Nakagawa (1) Naoya Takagi (1)

Research topics

Deep Learning (3) Machine Learning (2) Mathematics & Optimization (2) Models (2) Optimization (2) Artificial Intelligence (1) Architectures (1) Core AI (1) Core Methods (1) Optimization & Theory · Machine Learning (1) Optimization & Theory · Deep Learning (1)

Keywords

continuous relaxation (2) combinatorial optimization (2) representation learning (1) discrete optimization (1) gradient-based sampling (1) disentangled representation (1) variational autoencoder (1) neural network solver (1) latent space (1) local optima (1) posterior collapse (1) kl annealing (1) local optimum (1) long-context modeling (1) multi-resolution encoding (1) neural network (1) recursive generation (1) penalty term (1) hierarchical auto-regressive model (1) memory-efficient generation (1)

Papers

PHOTON: Hierarchical Autoregressive Modeling for Lightspeed and Memory-Efficient Language Generation ACL 2026 Optimization by Parallel Quasi-Quantum Annealing with Gradient-Based Sampling ICLR 2025 Controlling Continuous Relaxation for Combinatorial Optimization NeurIPS 2024 Learning Dynamics in Linear VAE: Posterior Collapse Threshold, Superfluous Latent Space Pitfalls, and Speedup with KL Annealing AISTATS 2024