PANORAMA-NOC4PC-Multimodal: Patentability Judgments with Drawings

PANORAMA-NOC4PC-Multimodal extends a text-only patentability benchmark by adding patent drawings. The dataset is based on the PANORAMA benchmark from NeurIPS 2025 Datasets & Benchmarks. It was uploaded by user sungjae98 to Hugging Face and last updated in June 2026.

Use Cases

• Train multimodal models for patentability classification based on claim text and drawings.
• Benchmark model performance on a task that mirrors real-world patent examiner workflows.
• Study the impact of visual information on legal decision-making tasks.
• Develop models that can reason across text and image modalities in a structured domain.

Strengths

• Extends a recognized NeurIPS 2025 benchmark (PANORAMA) with a new modality.
• Includes the actual drawings that real patent examiners reference, adding realism.
• Focuses on a specific, structured legal task (patentability judgment under §102/§103).

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file formats are unknown, which may limit suitability assessment.

Provenance

Source

Based on the PANORAMA benchmark (Lim et al., NeurIPS 2025 Datasets & Benchmarks).

Collection Method

Multimodal extension of the NOC4PC task, adding drawings to the original text data.

Freshness

Last updated 2026-06-02 12:53:52; freshness should be verified.