·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Predictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance1h◆Pixel-Level Transformers in Remote Sensing: A Canopy Height Case Study1h◆Explore, Execute, Evolve: A Skill Acquisition and Reuse Loop for Embodied Agents1h◆Making Duplicate Reimbursement Unrepresentable: A Verified Ethereum E-Invoice System for Humans and AI Agents1h◆Is manual software optimization a thing of the past?1h◆HandAnthro: Automated Hand Anthropometry from a Single Image1h◆It's Not What the Image Shows: Irrelevant Context Destabilises VLM Judges Without Informing Them1h◆AgentBug-Smith: Automatically Reproducing Real-World Harness Bugs in Agentic Systems1h◆Learning Beyond What You Sample: Off-Policy-Aware Cross-Model Trajectory Exchange for RLVR1h◆ExceptionDrive: A Planning-Oriented Counterfactual Corner-Case Benchmark for Autonomous Driving1h◆Boids of a Feather Flock Together - Evolving Prey Behaviours Under Different Predator Attack Strategies1h◆Beyond Interaction Capacity: Estimator Scaling with Recursive Models for CTR Prediction1h◆Generated Query Expansion Still Helps Strong Sparse Retrieval: A Controlled Study with SPLADE-v31h◆The Unequal Influence of Bad Advice: Using Training Data Attribution to Modulate Emergent Misalignment1h◆RLX: A Unified Multi-Backend Tensor Compiler and Distributed Runtime in Rust1h◆Rollout-Marginal Distillation for Long-Horizon Autoregressive Video Generation1h◆Does Local Video Understanding Transfer Across Encounters? The EgoGears Benchmark1h◆Dagger: Decoupling-based Model Stealing Attack against Graph Neural Networks1h◆On Trajectory-Aware Training for Masked Diffusion Language Models1h◆$S^3$: Spectral Null-Space Swap Makes Reasoning Models Efficient1h◆Predictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance1h◆Pixel-Level Transformers in Remote Sensing: A Canopy Height Case Study1h◆Explore, Execute, Evolve: A Skill Acquisition and Reuse Loop for Embodied Agents1h◆Making Duplicate Reimbursement Unrepresentable: A Verified Ethereum E-Invoice System for Humans and AI Agents1h◆Is manual software optimization a thing of the past?1h◆HandAnthro: Automated Hand Anthropometry from a Single Image1h◆It's Not What the Image Shows: Irrelevant Context Destabilises VLM Judges Without Informing Them1h◆AgentBug-Smith: Automatically Reproducing Real-World Harness Bugs in Agentic Systems1h◆Learning Beyond What You Sample: Off-Policy-Aware Cross-Model Trajectory Exchange for RLVR1h◆ExceptionDrive: A Planning-Oriented Counterfactual Corner-Case Benchmark for Autonomous Driving1h◆Boids of a Feather Flock Together - Evolving Prey Behaviours Under Different Predator Attack Strategies1h◆Beyond Interaction Capacity: Estimator Scaling with Recursive Models for CTR Prediction1h◆Generated Query Expansion Still Helps Strong Sparse Retrieval: A Controlled Study with SPLADE-v31h◆The Unequal Influence of Bad Advice: Using Training Data Attribution to Modulate Emergent Misalignment1h◆RLX: A Unified Multi-Backend Tensor Compiler and Distributed Runtime in Rust1h◆Rollout-Marginal Distillation for Long-Horizon Autoregressive Video Generation1h◆Does Local Video Understanding Transfer Across Encounters? The EgoGears Benchmark1h◆Dagger: Decoupling-based Model Stealing Attack against Graph Neural Networks1h◆On Trajectory-Aware Training for Masked Diffusion Language Models1h◆$S^3$: Spectral Null-Space Swap Makes Reasoning Models Efficient1h◆
News/AgentBug-Smith: Automatically Reproducing Real-World Harness Bugs in Agentic Systems
arxiv
PublishedSeptember 30, 2026 at 4:00 AM

AgentBug-Smith: Automatically Reproducing Real-World Harness Bugs in Agentic Systems

Source
arxiv.orgfull article ↗
Read on arxiv→
Publisher summary· verbatim

arXiv:2609.37864v1 Announce Type: cross Abstract: Agent harness bugs exhibit unique characteristics and remain challenging for state-of-the-art software agents to repair. Progress in this area is further hindered by existing benchmarks, which contain only a small and fixed number of executable harne

Stay posted· Newsletter

A 5-min weekly brief — top movers, price watch, story of the week.

// no spam · unsubscribe one-click · free forever

Discussion
Source
↗
arxiv
Read original ↗All from arxiv →

No replies yet. Be first.

Source
↗
arxiv
Read original ↗All from arxiv →

Related coverage

More from ARXIV
arxivPredictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance1harxivPixel-Level Transformers in Remote Sensing: A Canopy Height Case Study1harxivExplore, Execute, Evolve: A Skill Acquisition and Reuse Loop for Embodied Agents1harxivMaking Duplicate Reimbursement Unrepresentable: A Verified Ethereum E-Invoice System for Humans and AI Agents1h
The Bubble Brief
WEEKLY

Read AI insights every Tuesday — top movers, new releases, story of the week.

// no spam · unsubscribe one-click · free forever

Originally published on arxiv ↗
Built by Marouane Gazouzi
HomeModelsNews