MultiSoc-4D benchmark shows LLMs annotating Bengali social media exhibit instruction-induced label collapse, preferring fallback labels and missing 79% of hate speech and 75% of sarcasm instances despite high agreement but near-zero kappa.
Bogdanov et al
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4verdicts
UNVERDICTED 4representative citing papers
DiZiNER improves zero-shot NER by having multiple LLMs annotate texts and using a supervisor to refine instructions from their disagreements, reaching SOTA on 14 of 18 benchmarks with +8 F1 gains.
Opir introduces efficient multi-task encoder models trained on a 996-category safety taxonomy that match or exceed larger baselines on most safety benchmarks while using under 100M parameters for edge variants.
GLiNER-Relex unifies NER and RE in one zero-shot transformer-based model that achieves competitive results on CoNLL04, DocRED, FewRel, and CrossRE.
citing papers explorer
-
MultiSoc-4D: A Benchmark for Diagnosing Instruction-Induced Label Collapse in Closed-Set LLM Annotation of Bengali Social Media
MultiSoc-4D benchmark shows LLMs annotating Bengali social media exhibit instruction-induced label collapse, preferring fallback labels and missing 79% of hate speech and 75% of sarcasm instances despite high agreement but near-zero kappa.
-
DiZiNER: Disagreement-guided Instruction Refinement via Pilot Annotation Simulation for Zero-shot Named Entity Recognition
DiZiNER improves zero-shot NER by having multiple LLMs annotate texts and using a supervisor to refine instructions from their disagreements, reaching SOTA on 14 of 18 benchmarks with +8 F1 gains.
-
Opir: Efficient Multi-Task Safety Classification for Toxicity, Jailbreaks, Hate Speech, and Harmful Content
Opir introduces efficient multi-task encoder models trained on a 996-category safety taxonomy that match or exceed larger baselines on most safety benchmarks while using under 100M parameters for edge variants.
-
GLiNER-Relex: A Unified Framework for Joint Named Entity Recognition and Relation Extraction
GLiNER-Relex unifies NER and RE in one zero-shot transformer-based model that achieves competitive results on CoNLL04, DocRED, FewRel, and CrossRE.