Shaobai Jiang

shaobaij's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted 8 papers 1 day ago

Frontier Language Models Struggle to Copy: Text Can Be Better Viewed in 2D

Paper • 2607.16072 • Published 25 days ago • 1

Bad Memory: Evaluating Prompt Injection Risks from Memory in Agentic Systems

Paper • 2607.14611 • Published 26 days ago • 1

PAST-Bench: Benchmarking the Foundations of Recursive Self-Improvement in Personal Agents

Paper • 2608.04003 • Published 7 days ago • 33

When Attention Goes Blind: Numerical Failure in ALiBi Positional Encodings

Paper • 2608.03994 • Published 7 days ago • 8

Distilled Reinforcement Learning for LLM Post-training

Paper • 2607.17247 • Published 23 days ago • 11

LLM-as-a-Coach: Experiential Learning for Non-Verifiable Tasks

Paper • 2607.18110 • Published 22 days ago • 16

NVIDIA-labs OO Agents: Native Python Object-Oriented Agents

Paper • 2607.20709 • Published 20 days ago • 34

Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes

Paper • 2608.05000 • Published 6 days ago • 58

upvoted 6 papers 2 days ago

DiffusionGemma Technical Report

Paper • 2608.00146 • Published 11 days ago • 36

ContinualSkillBench: Can LLM Agents Truly Evolve Their Capabilities?

Paper • 2608.03874 • Published 7 days ago • 14

Permission Denied: Policy-Graded Evaluation of Coding Agents in Hardened Environments

Paper • 2608.02670 • Published 9 days ago • 1

DataSpace: Benchmarking Data Agents for Verifiable Analytics over Heterogeneous Workspaces

Paper • 2608.03451 • Published 7 days ago • 30

Harness-R1: Learning to Edit Executable Runtime Harnesses from Agent Failure Trajectories

Paper • 2608.02276 • Published 8 days ago • 3

From RLVR to RLSVR: Task Transformation Induces Self-Verifiable Rewards for Open-Ended LLM Self-Improvement

Paper • 2607.23802 • Published 16 days ago • 105

upvoted 6 papers 3 days ago

Partition, Prompt, Aggregate: Statistical Self-Consistency in Language Models

Paper • 2607.15277 • Published 26 days ago • 10

MetaView: Monocular Novel View Synthesis with Scale-Aware Implicit Geometry Priors

Paper • 2607.12000 • Published 29 days ago • 40

LATO.2: Factorized 3D Mesh Generation with Vertex and Topology Flow

Paper • 2607.10623 • Published about 1 month ago • 16

Hierarchical Denoising For Multi-Step Visual Reasoning

Paper • 2607.15278 • Published 26 days ago • 7

From Pixels to States: Rethinking Interactive World Models as Game Engines

Paper • 2607.14076 • Published 27 days ago • 37

BadWAM: When World-Action Models Dream Right but Act Wrong

Paper • 2607.15207 • Published 26 days ago • 54