Research map / Large language models / Retrieval and knowledge
TweetNERD - End to End Entity Linking Benchmark for Tweets
Shubhanshu Mishra, Aman Saini, Raheleh Makki, Sneha Mehta, Aria Haghighi, Ali Mollahosseini
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use Lookup heuristic instead of GENRE for entity linking with true spans | Entity linking · TweetNERD-OOD EL | F1 strong_all_match: 0.39 -> 0.584 | Improved |
| Use Lookup heuristic instead of REL for entity linking with true spans | Entity linking · TweetNERD-OOD EL | F1 entity_match: 0.463 -> 0.621 | Improved |
| Use GENRE instead of Lookup on the Academic subset (EL) | Entity linking · TweetNERD-Academic EL | F1 entity_match: 0.645 -> 0.636 | No clear effect |
| Use Google NLAI instead of TAGME for end-to-end linking | End2End linking · TweetNERD-OOD End2End | F1 entity_match: 0.402 -> 0.522 | Improved |
| Use NLAI instead of GENRE for end-to-end linking | End2End linking · TweetNERD-OOD End2End | F1 strong_all_match: 0.223 -> 0.313 | Improved |
| Evaluate the same NER system on Academic instead of OOD subset | AllenNLP · TweetNERD NER | F1 strong_mention_match: 0.454 -> 0.552 | Improved |
| Use BERTweet WNUT17 (social media model) on OOD instead of Academic | BERTweet WNUT17 · TweetNERD NER | F1 strong_mention_match: 0.46 -> 0.278 | Hurt |
| Consolidate AMBIGOUS and NOT FOUND as NIL for agreement | TweetNERD annotations · mention-entity majority agreement | fraction with majority agreement: 41\% -> 47\% | Improved |
Closest papers by meaning
- Hierarchical Aligned Multimodal Learning for NER on Tweet Posts (AAAI 2024 · similarity 0.80)
- Small Language Model Makes an Effective Long Text Extractor (AAAI 2025 · similarity 0.80)
- TGB 2.0: A Benchmark for Learning on Temporal Knowledge Graphs and Heterogeneous Graphs (NeurIPS 2024 · similarity 0.79)
- UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognition (ICLR 2024 · similarity 0.79)
- DTGB: A Comprehensive Benchmark for Dynamic Text-Attributed Graphs (NeurIPS 2024 · similarity 0.79)
- Knowledge Graph Completion with Relation-Aware Anchor Enhancement (AAAI 2025 · similarity 0.78)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
