Concept · 1 paper · updated 2026-11-12
Stance detection
Classify whether a text favours, opposes or ignores a target or claim, then use that to flag false content: as evidence stance in fact-checking or as crowd and media stance in rumour pipelines. One survey is the only source here; its scores come from other papers and are not comparable. Trade-off: label mismatch and small data against evidence access and accuracy.
misinformationfact-checkingrumoursnlptext-classification
Markdown4 min read
AI-compiled overview, updated 2026-11-12, not yet reviewed by a human. Every claim cites its source; check it when a detail matters.
Overview
Stance detection asks whether a text agrees with a target, disagrees with it, or takes neither side. The target may be a claim, a headline or a topic, and it may be implied rather than stated. When the target is implicit the task looks much like sentiment analysis 2103.00242 §3, so keep the two apart when you pick a dataset 2103.00242 §2.
In mis- and disinformation work, stance becomes a veracity signal in one of two ways. Either the stance of the evidence towards a claim decides whether the claim holds 2103.00242 §3.1, or the stance of the crowd and the media towards a rumour feeds a decision about it 2103.00242 §3.2.
The invariant all variants share is a classification of a (text, target) pair into one label from a fixed set. The label sets differ by dataset: Favor, Against and Neither in the general definition; positive, negative, discuss and unrelated for FNC-1; SUPPORTED, REFUTED and NOT ENOUGH INFO for FEVER; Support, Deny, Query and Comment for rumours 2103.00242 §2 2103.00242 §3.1 2103.00242 §3.2. Merge labels only with an explicit mapping.
Variants
The variants below all come from one survey, the Hardalov et al. survey. The survey runs no experiments of its own, so each variant is described by the results it reports. The underlying papers have no cards on this wiki, so their content is not used here.
Stance classification on a labelled dataset
Given a pair such as a headline and its article body, predict the label. The survey's main example is FNC-1, with 75K instances 2103.00242 tab. 1. Early systems used hand-made features: words, embeddings and sentiment lexica. In one analysis, lexical features mattered most and sentiment did not help 2103.00242 §4. Pre-trained Transformers changed this. On FNC-1, RoBERTa raised Disagree from 18 to 58 F1 and Agree from 50 to 70 F1 2103.00242 §4. Card: Hardalov et al. survey.
Fact-checking as stance (evidence retrieval and reasoning)
Retrieve evidence, then judge the claim against it. In FEVER, the evidence is Wikipedia sentences 2103.00242 tab. 1. The survey lists these top FEVER scores: 64.2 for a neural semantic matching network, then 67.1 and about 70 for graph neural networks over the evidence, with a better retriever and graph propagation 2103.00242 §4. The survey notes that FEVER-style checking differs from stance detection: it scores evidence against a claim, not an author's stance 2103.00242 §3.1. Card: Hardalov et al. survey.
Fact-checking without retrieval
A pre-trained language model scores the claim directly, through masked-token scoring or through the perplexity of the claim given the document. There is no retrieval step, but the approach is prone to pattern biases and performs below supervised approaches. Few-shot training helps 2103.00242 §4.
Threaded stance in rumour pipelines
For replies in a conversation, the survey covers tree-structured models over threads, branch-wise LSTMs, and Tree and Hierarchical Transformers for rumour debunking. Pre-training on stance data, and per-class pre-training data built from related tasks, improved RumourEval results 2103.00242 §4. Position in the thread and timing carry signal: denying posts appear later, and the share of querying posts rises over time while their absolute number falls 2103.00242 §4.
Multi-dataset and cross-domain learning
The survey covers multi-task learning over several stance datasets, and cross-domain learning over 16 stance datasets with a domain-adaptation architecture named MoLE. It also reports pattern-exploiting training across 12 languages 2103.00242 §4 2103.00242 §5. Combining datasets with different label sets needs a way to relate the labels, such as label embeddings 2103.00242 §5. See multi-task learning.
When to use which
The survey gives no decision rules. The table reads its reported results against common constraints and is guidance, not a tested recommendation.
| Situation | Use | Why |
|---|---|---|
| Labelled pairs in one label set, and fine-tuning is feasible | Pre-trained Transformer classifier | Large gains on FNC-1; the learning rate was the most important hyperparameter 2103.00242 §4 |
| Labelled pairs, but few or skewed examples | Sampling, class weighting, auxiliary-task data, or several datasets | The survey's listed mitigations for small, imbalanced datasets 2103.00242 §5 |
| Claim to check against a document collection | Retrieval, then a stance or inference classifier | Top FEVER-score systems use this pipeline 2103.00242 §4 |
| No evidence index available | Language-model checking without retrieval | Needs no retrieval step, but below supervised approaches and prone to pattern biases 2103.00242 §4 |
| Replies in a conversation thread | Thread-aware models | Thread depth and timing carry signal 2103.00242 §4 |
| Few labelled examples in a language | Pattern-exploiting training, or multi-dataset learning | Reported useful in low- and full-resource settings 2103.00242 §4 2103.00242 §5 |
| Need a reason for each decision | Partial explanations only | Memory networks, attention and topics give partial explanations; attention weights may mislead 2103.00242 §5 |
Implementation sketch
The survey gives no code. The sketch shows the two shapes described above. The aggregation step is not stated in the survey.
# Stance of one (text, target) pair: one label from a fixed set [§2]
# Input: text, target (the target may be implicit) [§3]
# Output: a label from the dataset's own label set [§2]
def stance(text, target, model):
# Encode the pair; the survey does not fix the encoder [§4]
features = encode(text, target)
# Return the most likely label from the dataset's label set [§2]
return argmax(model.predict(features))# Fact-checking as stance, formulation (a) [§3.1]
# Input: claim (text); evidence (list of texts)
# Output: veracity label
def verify_claim(claim, evidence, model):
# Stance of each evidence text towards the claim [§3.1]
stances = [stance(claim, e, model) for e in evidence]
# How the stances become one verdict is not stated [§3.1]
return aggregate(stances)Contradictions
- Arabic FC and AraStance. The text gives 52 F1 for AraStance and 78 F1 for Arabic FC 2103.00242 §4. The table gives 52 to Arabic FC and 78 to AraStance 2103.00242 tab. 2. The survey does not resolve this. Check the AraStance paper before you quote either number.
- RumourEval '19. The text gives 57.9 F1 for the best stance result 2103.00242 §4. The table gives 61.9 for the same paper and dataset 2103.00242 tab. 2. No explanation is given.
- Comparability of the table. The scores use different datasets, metrics and splits, and the survey says they cannot be compared directly 2103.00242 §4 2103.00242 tab. 2. Do not rank methods by that table.
- Leaderboard row. The FEVER row comes from a public leaderboard, not from a paper 2103.00242 tab. 2.
Papers
- Hardalov et al. survey: maps stance formulations for false-content detection, tabulates 15 datasets and the methods tried on them, and lists open issues.
Source and license
Compiled from the papers listed on this page, all published under CC BY, CC BY-SA or CC0, with citations. See each paper's page for its authors and license.