VFR-MISE: An Asymmetric Dual-Evidence Verification–Fallback Framework for Few-Shot Stressor Extraction from Social Media
Preprint
- 7 September 2026
- preprint
- Published by MDPI AG in Preprints.org
Abstract
Few-shot stressor extraction from temporally shifting social-media text is challenged by unstable span boundaries and by a mismatch between structural confidence and task-semantic validity. We propose VFR-MISE, an asymmetric dual-evidence verification–fallback framework. VFR-MISE retains the temporal few-shot and knowledge-inheritance paradigm of MISE, strengthens candidate generation with DeBERTa-v3-base, and constructs two candidate-level signals: a guideline-conditioned verifier for semantic evidence and BIOES emission aggregation for structural evidence. Semantic evidence governs candidate admission; structural evidence is consulted only in the rejected region for constrained recovery, after which deterministic span merging produces the final output. In controlled reruns using identical data splits, training budgets, and Exact-span evaluation, replacing RoBERTa-base with DeBERTa-v3-base increases cross-shot F1 from 0.6841 to 0.7236 (+3.95 percentage points). With the DeBERTa-v3 backbone fixed, the complete VFR-MISE reaches 0.7248. These results suggest that stronger candidate representations and asymmetric dual-evidence decision refinement provide complementary benefits, improving the precision–recall trade-off while retaining an auditable decision process.Keywords
All Related Versions
- Version 1, 2026-09-07, Preprints
- Published version: Big Data and Cognitive Computing, 10 (9), 315.
This publication has 28 references indexed in Scilit:
- MISE: Meta-knowledge Inheritance for Social Media-Based Stressor Estimation2025
- Conformal Prediction for Natural Language Processing: A SurveyTransactions of the Association for Computational Linguistics, 2024
- Few-Shot Named Entity Recognition: An Empirical Baseline Study2021
- SpanNER: Named Entity Re-/Recognition as Span Prediction2021
- Few-NERD: A Few-shot Named Entity Recognition Dataset2021
- Locate and Label: A Two-stage Identifier for Nested Named Entity Recognition2021
- Simple and Effective Few-Shot Named Entity Recognition with Structured Nearest Neighbor Learning2020
- Calibrating Structured Output Predictors for Natural Language ProcessingProceedings of the conference. Association for Computational Linguistics. Meeting, 2020
- A Taxonomy of Ethical Tensions in Inferring Mental Health States from Social Media2019
- Data Statements for Natural Language Processing: Toward Mitigating System Bias and Enabling Better ScienceTransactions of the Association for Computational Linguistics, 2018