Exploring the
Frontiers of AI
Deep dives into machine learning, neural architectures, and the future of artificial intelligence.
Latest Post
Retaining by doing the role of on Policy data in mitigating forgetting
Source: “Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting,” arXiv: arXiv:2510.18874.
Latest Publications
Agent Based automated claim matching with instruction Following llms
Source: “Agent-based Automated Claim Matching with Instruction-following LLMs,” arXiv: arXiv:2510.23924.
The 10,000x explosion reproducing deepseek’s mhc at scale
The 10,000x Explosion: Reproducing DeepSeek’s mHC at Scale
Mhc manifold Constrained hyper Connections
Source: “mHC: Manifold-Constrained Hyper-Connections,” arXiv: arXiv:2512.24880.
Conftuner training large language models to express their confidence verbally
Source: “ConfTuner: Training Large Language Models to Express Their Confidence Verbally,” arXiv:2508.18847.
The zero temperature myth why greedy doesn't always mean same
The Zero Temperature Myth: Why “Greedy” Doesn’t Always Mean “Same”
Halogen fantastic llm hallucinations and where to find them
Source: “HALoGEN: Fantastic LLM Hallucinations and Where to Find Them,” arXiv: arXiv:2501.08292.
Why diffusion models don’t memorize the role of implicit dynamical regularization in training
Source: “Why Diffusion Models Don’t Memorize: The Role of Implicit Dynamical Regularization in Training,” The Thirty-ninth Annual Conference on Neural Information Processing Systems
Soft thinking unlocking the reasoning potential of llms in continuous concept space
Source: “Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space,” arXiv: arXiv:2505.15778.
Does reinforcement learning really incentivize reasoning capacity in llms beyond the base model
Source: “Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?,” The Thirty-ninth Annual Conference on Neural Information Processing Systems
1000 layer networks for self Supervised rl scaling depth can enable new goal Reaching capabilities
Source: “1000 Layer Networks for Self-Supervised RL: Scaling Depth Can Enable New Goal-Reaching Capabilities” The Thirty-ninth Annual Conference on Neural Information Processing Systems
Artificial hivemind the open Ended homogeneity of language models (and beyond)
Source: “Artificial Hivemind: The Open-Ended Homogeneity of Language Models (and Beyond),” The Thirty-ninth Annual Conference on Neural Information Processing Systems Datasets and Benchmarks Track Code: https://github.com/liweijiang/artificial-hivemind Dataset: INFINITY-CHAT Collection
Gated attention for large language models non Linearity, sparsity, and attention Sink Free
Source: “Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free,” The Thirty-ninth Annual Conference on Neural Information Processing Systems
Lightmem Lightweight and efficient memory Augmented generation
Source: “LightMem: Lightweight and Efficient Memory-Augmented Generation,” arXiv: arXiv:2510.18866
The cockpit of ai a beginner’s guide to llm parameters
When you use an LLM (Large Language Model) through an API like OpenRouter, you aren’t just sending a text message and hoping for the best. You actually have access to a “cockpit” of dials and...
Inside vllm how this amazing engine makes large models lightning fast
Source: Inside vLLM: Anatomy of a High-Throughput LLM Inference System - Aleksa Gordić
Brillm Brain Inspired Large Language Model
Source: “BriLLM: Brain-inspired Large Language Model,” arXiv: arXiv:2503.11299
Revisiting Long Context Modeling From Context Denoising Perspective
Source: “Revisiting Long-context Modeling from Context Denoising Perspective,” arXiv: arXiv:2510.05862