Google Launches PaperBanana-Interact for Scientific Diagram Refinement
google · hf · 2026-09-01
Google released PaperBanana-Interact, a multi-turn benchmark for scientific diagram refinement with human feedback. The accompanying multi-agent system improves scientific diagram generation by reducing quality drift and feature forgetting across revision turns.
More from Research
- TAAL Fixes Early Beam Pruning in Generative Recommendation — _reachsumit · 2026-09-01
- Strong/Weak Graph Edges Encode Opposite Relations in CF — _reachsumit · 2026-09-01
- REIGN Enables Efficient Context-Length Scaling for Retrieval — _reachsumit · 2026-09-01
- IBM Proposes Spatial Matryoshka Training for Multi-Granularity Document Retrieval — _reachsumit · 2026-09-01
- Doc-REFRAG: coarse-compress then selectively expand for faster, more accurate multi-image RAG — _reachsumit · 2026-09-01
- Google Releases RSLM: Training-Free Vector Quantization for ANN Search — _reachsumit · 2026-09-01