Internal Generation Record
Internal generation metadata: 344 candidate papers, lightweight restored issue.
Generation Record
This page preserves selected papers, candidate scale, and source-date metadata for traceability. The page only changes presentation, not selected papers, ordering, or counts.
Generation note
This page restores the 2026-06-07 source record from the real 2026-06-05 arXiv candidate pool. It uses a reranked lightweight selection and does not include any mock papers.
Selected papers
- TEVI: Text Conditioned Editing of Visual Representations via Sparse Autoencoders for Improved Vision Language Alignment
- Socratic SWE: Self Evolving Coding Agents via Trace Derived Agent Skills
- Mind the Gap: Disentangling Performance Bottlenecks in Video Instance Segmentation
- Making the Most of Limited Data: Score Aware Training for Text to Music Generation
- RhinoVLA Technical Report
Candidate pool sample
- A robust PPG foundation model using multimodal physiological supervision
- Hierarchical Certified Semantic Commitment for Byzantine Resilient LLM Agent Collaboration
- Closed Form Spectral Regularization for Multi Task Model Merging
- MMAE: A Massive Multitask Audio Editing Benchmark
- Seeing Without Exposing