ACL Findings2023
Causal interventions expose implicit situation models for commonsense language understanding
Takateru Yamakoshi, James L. McClelland, Adele Goldberg, Robert D. Hawkins
Abstract
Accounts of human language processing have long appealed to implicit "situation models" that enrich comprehension with relevant but unstated world knowledge. Here, we apply causal intervention techniques to recent transformer models to analyze performance on the Winograd Schema Challenge (WSC), where a single context cue shifts interpretation of an ambiguous pronoun. We identify a relatively small circuit of attention heads that are responsible for propagating information from the context word that guides which of the candidate noun phrases the pronoun ultimately attends to. We then compare how this circuit behaves in a closely matched "syntactic" control where the situation model is not strictly necessary. These analyses suggest distinct pathways through which implicit situation models are constructed to guide pronoun resolution 1 .
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
1 top-tier paper has cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Locating and Editing Factual Associations in GPT cites Investigating Gender Bias in Language Models Using Causal Mediation Analysis
- ReFT: Representation Finetuning for Language Models cites Causal Abstractions of Neural Networks
- ReFT: Representation Finetuning for Language Models cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
Cited by top-tier papers1
Builds on21
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
- ALBERT: A Lite BERT for Self-supervised Learning of Language RepresentationsZhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel et al.ICLR 2020 · 7,418 citations
- Locating and Editing Factual Associations in GPTKevin Meng, David Bau, Alex Andonian, Yonatan BelinkovNeurIPS 2022 · 3,415 citations
- Investigating Gender Bias in Language Models Using Causal Mediation AnalysisJesse Vig, Sebastian Gehrmann, Yonatan Belinkov, Sharon Qian et al.NeurIPS 2020 · 851 citations
- Causal Abstractions of Neural NetworksAtticus Geiger, Hanson Lu, Thomas Icard, Christopher PottsNeurIPS 2021 · 516 citations
Related papers
- The Sensitivity of Language Models and Humans to Winograd Schema PerturbationsMostafa Abdou, Vinit Ravishankar, Maria Barrett, Yonatan Belinkov et al.ACL 2020 · 1 citation
- Do Context-Aware Translation Models Pay the Right Attention?Kayo Yin, Patrick Fernandes, Danish Pruthi, Aditi Chaudhary et al.ACL 2021
- WinoWhy: A Deep Diagnosis of Essential Commonsense Knowledge for Answering Winograd Schema ChallengeHongming Zhang, Xinran Zhao, Yangqiu SongACL 2020 · 33 citations
- Picturing Ambiguity: A Visual Twist on the Winograd Schema ChallengeBrendan Park, Madeline Janecek, Naser Ezzati-Jivan, Yifeng Li et al.ACL 2024
- Attention-based Contrastive Learning for Winograd SchemasTassilo Klein, Moin NabiEMNLP Findings 2021 · 5 citations
