Papers
Deep, long-form breakdowns of recent research in AI, machine learning, security & programming languages.
JustFit: 200K-Token LLM Serving on a 24 GiB Laptop with Just-in-Time State Management
by Yuhua Chen
Coupled Calibration and Learning: Mitigating Teacher Bias in LLM Distillation without Target-Domain Reward Feedback
by Haichen Hu, Yuheng Zhang, David Simchi-Levi
Reduced-Space Multi-Fidelity Bayesian Optimization of Process Simulation Models
by Niki Triantafyllou, Andrea Bernardi, Maria M. Papathanasiou
Right Tool, Right Job: Native-Language Evaluation, Tokenizer Sensitivity, and Methodological Findings from a French-Only BabyLM
by Adam Zachary Wasserman, David Beauchemin
Learning-Guided Planning in Large Dynamic Action Spaces: Budgeted Tree Search for One-to-Many Mobile Charging
by Liang-Ching Tao, Pi-Chung Wang
Talking Head Synthesis with Facial Landmark Guidance via 3D Gaussian Splatting
by Ziheng Yang, Yinfeng Yu, Yongming Li
Transformer-Based Token Fusion and Dynamic Graph Planning for Audio-Visual Navigation
by Shaohang Wu, Yinfeng Yu
World Model Science: Self-Organized Criticality, Weak Chaos, and Metastable Belief Dynamics in Long-Horizon LLM Agents
by Xinyuan Song, Zekun Cai
Knowledge as Orbit: Finite Collections as Phases of an Exactly Periodic Latent Generator
by Siddharth Pal, Viktoria Rojkova
Corrupt Plans, Clean Traces: Evading Chain-of-Thought Monitoring with Plan Injection
by Keertana Chidambaram, Andrew Ilyas, Vasilis Syrgkanis