Xiao Zihan

xiaozihan2025

·

AI & ML interests

Paper replication and lightweight experiments. Sharing reproducible results.

Recent Activity

liked a model 1 day ago

deepseek-ai/ESFT-token-translation-lite

liked a model 1 day ago

mistralai/Mistral-7B-v0.1

liked a model 1 day ago

deepseek-ai/ESFT-gate-code-lite

View all activity

Organizations

None yet

upvoted a paper 6 days ago

UltraText Bench: A Comprehensive Bilingual Benchmark for Evaluating Visual Text Rendering in Image Generation

Paper • 2610.09823 • Published 8 days ago • 188

upvoted 3 papers 12 days ago

ClawBench: Can AI Agents Complete Everyday Online Tasks?

Paper • 2604.08523 • Published Apr 9 • 379

RewardHarness: Self-Evolving Agentic Post-Training

Paper • 2605.08703 • Published May 9 • 246

Learning from the Self-future: On-policy Self-distillation for dLLMs

Paper • 2606.18195 • Published Jun 16 • 176

upvoted a paper 29 days ago

Dream-RSI: Recursive Self-Improvement through Evolving Worlds

Paper • 2609.14858 • Published about 1 month ago • 255

upvoted a paper about 1 month ago

What Did I Just Say? Self-Listening for Full-Duplex Speech Models

Paper • 2609.05592 • Published Sep 4 • 25

upvoted 2 papers 2 months ago

TARS: Timestep-Aware Data Scaling for 3D-Free Video Re-Shooting

Paper • 2607.28261 • Published Jul 30 • 18

Qwen-UI-Agent Technical Report: Toward Next-Generation Real-World Centric Foundation GUI Agents

Paper • 2607.28227 • Published Jul 30 • 159

upvoted 8 papers 3 months ago

Progress Reward Modeling for Robotic Learning: A Comprehensive Survey

Paper • 2607.21655 • Published Jul 22 • 114

ABot-World-0: Infinite Interactive World Rollout on a Single Desktop GPU

Paper • 2607.19191 • Published Jul 21 • 116

RynnBrain 1.1: Towards More Capable and Generalizable Embodied Foundation Model

Paper • 2607.17977 • Published Jul 20 • 63

X-Lens: Real-Time Metric Depth Estimation with Heterogeneous Cameras

Paper • 2607.12993 • Published Jul 14 • 7

ResearchStudio-Reel: Automate the Last Mile of Research from Paper to Poster, Video, and Blog

Paper • 2607.04438 • Published Jul 5 • 62

OmniOpt: Taxonomy, Geometry, and Benchmarking of Modern Optimizers

Paper • 2607.04033 • Published Jul 4 • 55

The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning

Paper • 2606.29526 • Published Jun 28 • 52

MemLearner: Learning to Query Context memory for Video World Models

Paper • 2606.31734 • Published Jun 30 • 29

upvoted 3 papers 4 months ago

An Enigma of Artificial Reason: Investigating the Production-Evaluation Gap in Large Reasoning Models

Paper • 2606.01462 • Published May 31 • 6

ABot-Earth 0.5: Generative 3D Earth Model

Paper • 2606.09967 • Published Jun 8 • 178

STRIDE: Training Data Attribution via Sparse Recovery from Subset Perturbations

Paper • 2606.05165 • Published Jun 3 • 4

upvoted a paper 5 months ago

OpenComputer: Verifiable Software Worlds for Computer-Use Agents

Paper • 2605.19769 • Published May 19 • 68