# Social Commonsense Reasoning with Multi-Head Knowledge Attention

arXiv: https://arxiv.org/abs/2010.05587 · HTML: https://mlsanity.com/paper/2010.05587/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Oct 2020.

## Built on by (most cited first, 10 of 10)

- [COINS: Dynamically Generating COntextualized Inference Rules for Narrative Story Completion](https://mlsanity.com/paper/2106.02497/) (2021)
- [Self-Consistent Narrative Prompts on Abductive Natural Language Inference](https://mlsanity.com/paper/2309.08303/) (2023)
- [Wiring the 'Why': A Unified Taxonomy and Survey of Abductive Reasoning in LLMs](https://mlsanity.com/paper/2604.08016/) (2026)
- [Generating Hypothetical Events for Abductive Inference](https://mlsanity.com/paper/2106.03973/) (2021)
- [CO-NNECT: A Framework for Revealing Commonsense Knowledge Paths as Explicitations of Implicit Knowledge in Texts](https://mlsanity.com/paper/2105.03157/) (2021)
- [Interactive Model with Structural Loss for Language-based Abductive Reasoning](https://mlsanity.com/paper/2112.00284/) (2021)
- [CIKQA: Learning Commonsense Inference with a Unified Knowledge-in-the-loop QA Paradigm](https://mlsanity.com/paper/2210.06246/) (2022)
- [Knowledge-enhanced Neuro-Symbolic AI for Cybersecurity and Privacy](https://mlsanity.com/paper/2308.02031/) (2023)
- [CHARET: Character-centered Approach to Emotion Tracking in Stories](https://mlsanity.com/paper/2102.07537/) (2021)
- [Go Beyond Plain Fine-tuning: Improving Pretrained Models for Social Commonsense](https://mlsanity.com/paper/2105.05913/) (2021)

## Abstract (arXiv)

Social Commonsense Reasoning requires understanding of text, knowledge about social events and their pragmatic implications, as well as commonsense reasoning skills. In this work we propose a novel multi-head knowledge attention model that encodes semi-structured commonsense inference rules and learns to incorporate them in a transformer-based reasoning cell. We assess the model's performance on two tasks that require different reasoning skills: Abductive Natural Language Inference and Counterfactual Invariance Prediction as a new task. We show that our proposed model improves performance over strong state-of-the-art models (i.e., RoBERTa) across both reasoning tasks. Notably we are, to the best of our knowledge, the first to demonstrate that a model that learns to perform counterfactual reasoning helps predicting the best explanation in an abductive reasoning task. We validate the robustness of the model's reasoning capabilities by perturbing the knowledge and provide qualitative analysis on the model's knowledge incorporation capabilities.