Entity linking

Paste any text — each mention is linked to its GPTKB v2 entity, disambiguated by context (the same word can resolve to different entities).

Try:

Pipeline

The linker is LELA (Haffoudhi et al., IJCAI-ECAI 2026), configured against GPTKB v2 — candidate retrieval runs on the KB's observed surface forms, the same disambiguation evidence the browser shows humans.

  1. Detect mentions

    GLiNER (numind/NuNER_Zero-span) — zero-shot NER, 9 labels, threshold 0.5, no fine-tuning.

  2. Gather candidates

    GPTKB v2 observed surface forms ranked by frequency; BM25 fills leftover slots (top-20). LELA's optional Reranker stage is deliberately skipped (NoOp) — surface popularity already floats the right entity to the top.

  3. Disambiguate

    One openai/gpt-oss-120b call per mention, run in parallel, choosing from the numbered candidate list — the model may answer “None” → the mention stays NIL.

    Run a text above to see the exact prompt sent for one of its mentions.

Disambiguation prompt

The exact request sent to openai/gpt-oss-120b for mention — rebuilt with LELA's own message builder, so this is what the model actually received.

System


                

User — the mention marked in context, then the numbered candidate list


                

Model answered