Code Understanding is a Bottleneck for Coding Agents
Research
Research News
New papers, breakthroughs and academic AI research
VFold: Symmetry-Aware Cross-Layer Value Cache Compression
Research
NavGPT-3: Harnessing Context in a Hierarchical Navigation Runtime
Research
From Geometry to Generalization: Why Row Normalization Can Beat Adam and Muon
Research
SV-TAD: Native Sparse Convs for Efficient Temporal Action Detection
Research
MRCert: Towards Post-deployment Patch Robustness Certification for Adversarially Patched Samples via Type-specific Masking
Research
TestJack: Should you trust the results in coding benchmarks? Agentic Coding Benchmarks Auditing via Evaluator Evolution
Research
SpatialOPSD: Self-Distilling Spatial Intelligence from Verified Coding Agent Traces
Research
From Investigation Failures to Reliable SOC Agents: Understanding and Improving LLM-Based Alert Triage
Research
Predicting Alignment Generalization with Value Representations
Research
Residual spectral instabilities in representation learning
Research
NP-Hardness of Minimizing Neurons in Two-Hidden-Layer ReLU Neural Networks
Research