Woosung (Reiss) Koh
  • Main
  • Experience
  • News
  • Experience
  • News
  • Projects
    • Pandas
    • PyTorch
    • scikit-learn
  • Blog
    • 🎉 Easily create your own simple yet highly customizable blog
    • 🧠 Sharpen your thinking with a second brain
    • 📈 Communicate your results effectively with the best data visualizations
    • 👩🏼‍🏫 Teach academic courses
    • ✅ Manage your projects
  • Publications
    • Large Language Models Can Control Their Own Attention Span
    • mSFT: Addressing Dataset Mixtures Overfiting Heterogeneously in Multi-task SFT
    • Generative Visual Code Mobile World Models
    • Predicting LLM Reasoning Performance with Small Proxy Model
    • AdaSTaR: Adaptive Data Sampling for Training Self-Taught Reasoners
    • C$^2$: Scalable Auto-Feedback for LLM-based Chart Generation
    • FlickerFusion: Intra-trajectory Domain Generalizing Multi-Agent RL
    • Encoding Temporal Statistical-space Priors via Augmented Representation
    • Curriculum Learning and Imitation Learning for Model-free Control on Financial Time-series
    • Network-based exploratory data analysis and explainable three-stage deep clustering for financial customer profiling
  • Recent & Upcoming Talks
    • Example Talk
  • Teaching
    • Learn JavaScript
    • Learn Python

Large Language Models Can Control Their Own Attention Span

May 22, 2026·
[P2] Namgyu Ho*, Huzama Ahmad*, Woosung Koh*, Se-Young Yun, Tal Schuster, Cicero Nogueira Dos Santos
· 0 min read
Type
Conference paper
Publication
Pre-print
Last updated on May 29, 2026
Long-Context Efficient Inference Sparse Attention
mSFT: Addressing Dataset Mixtures Overfiting Heterogeneously in Multi-task SFT Mar 25, 2026 →

© 2026 Reiss Koh

Published with Hugo Blox Builder — the free, open source website builder that empowers creators.