Victor Grösser

orinmemory

victorgrosserAI

AI & ML interests

💻 Tech & Homelab · Servers, hardware, networking · 🤖 AI, coding & SpaceX 🚀 · ⚖️ Legal service provider. Building edtech and my own AI agent team on my homelab. Married, cat person 🐱, building from Groningen 🇳🇱

Recent Activity

commentedon a paper about 18 hours ago

CPU-ARC: Agentic Reasoning Caching for on-CPU Reasoning Reuse

commentedon a paper about 18 hours ago

How Much Training-Inference Mismatch Can On-Policy Distillation Tolerate?

commentedon a paper 5 days ago

EvoMemState: Evolving Memory State for Personal Agents

View all activity

Organizations

None yet

commented 2 papers about 18 hours ago

CPU-ARC: Agentic Reasoning Caching for on-CPU Reasoning Reuse

Paper • 2611.09891 • Published 5 days ago • 10 •

How Much Training-Inference Mismatch Can On-Policy Distillation Tolerate?

Paper • 2611.11450 • Published 3 days ago • 11 •

commented 2 papers 5 days ago

EvoMemState: Evolving Memory State for Personal Agents

Paper • 2611.03743 • Published 9 days ago • 36 •

Harness-S: Learning from Sibling Harnesses for Generalizable Agent Training

Paper • 2611.04277 • Published 8 days ago • 68 •

commented 2 papers 6 days ago

On the Generalization Limitation of On-Policy Distillation for Post-RL Consolidation

Paper • 2611.05609 • Published 8 days ago • 24 •

What Does Multi-Turn Reinforcement Learning Really Improve in LLM Agents?

Paper • 2611.06483 • Published 7 days ago • 39 •

commented 2 papers 7 days ago

Order-Only On-Policy Distillation

Paper • 2611.02112 • Published 10 days ago • 36 •

Axion 1.0: Open Agent Foundation Models, Simplified

Paper • 2611.03233 • Published 9 days ago • 50 •

commented 2 papers 8 days ago

BYOD: Build Your Own DLM - Efficient Autoregressive-to-Diffusion Language Model Conversion

Paper • 2610.41574 • Published 12 days ago • 11 •

CoworkOS:Scaling Agent Coworkers for Real-world Automation

Paper • 2611.03178 • Published 9 days ago • 100 •

commented 2 papers 12 days ago

Multi-Token Prediction Done Right

Paper • 2610.38331 • Published 14 days ago • 23 •

WorkFactory: Turning Professional Intents into Executable Work for Scalable Agent Training

Paper • 2610.37161 • Published 15 days ago • 41 •

commented 2 papers 13 days ago

TAO: Tool-Aware Joint Scheduling for LLM-Based Agentic Serving

Paper • 2610.37230 • Published 15 days ago • 26 •

DTCA: Denoising Token Credit Assignment for Reinforcement Learning in Diffusion Language Models

Paper • 2610.35201 • Published 16 days ago • 15 •

commented 2 papers 14 days ago

What Effort Level Should I Choose? On the Economics of Reasoning Effort in LLM Agents

Paper • 2610.34491 • Published 17 days ago • 22 •

Memory Tax: Dissecting the Net Value of Agent Memory

Paper • 2610.34073 • Published 17 days ago • 27 •

commented 2 papers 15 days ago

Extreme Compression of Per-Layer Embeddings via Cross-Layer Latent Vector Quantization

Paper • 2610.31309 • Published 20 days ago • 13 •

Trajectory2Gym: Decision-Grounded Compositional Interaction Synthesis for Tool-Use Agents

Paper • 2610.33568 • Published 18 days ago • 39 •

commented 2 papers 22 days ago

CoworkEnv: Synthesizing Training Environments for Long-Horizon Cowork Agents

Paper • 2610.24960 • Published 24 days ago • 45 •

Self-Evolving Coding Agents via Harness-Policy Optimization

Paper • 2610.22791 • Published 26 days ago • 49 •

Victor Grösser

AI & ML interests

Recent Activity

Organizations

orinmemory's activity