mlsanityData to 7 Oct 2026

Continuous Interpretive Steering for Scalar Diversity

arXiv 2604.07006 ↗First posted 8 Apr 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Pragmatic inference is inherently graded. Different lexical items give rise to pragmatic enrichment to different degrees. Scalar implicature exemplifies this property through scalar diversity, where implicature strength varies across scalar items. However, evaluations of pragmatic inference in large language models (LLMs) often rely on prompt-based manipulations. Beyond prompt-level effects, this study introduces Continuous Interpretive Steering (CIS), a method that probes graded pragmatic interpretation by treating activation-level steering strength as a continuous experimental variable. To support this analysis, this study introduces a new dataset, GraSD, which encodes graded scalar diversity. Experiments on four LLMs show that uniform activation steering increases pragmatic interpretations globally but collapses item-level variation, whereas graded activation steering yields differentiated interpretive shifts aligned with scalar diversity grades. It indicates that graded sensitivity is encoded in the representation space and can be systematically recovered through controlled intervention. Together, CIS and GraSD provide a principled framework for evaluating graded pragmatic sensitivity in LLMs.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 17 of its 31 references that are in the map. First posted on arXiv 8 Apr 2026.

Builds on

17 of 31 references.

References · 17

  1. 2024The Llama 3 Herd of Modelscited by 14,192
  2. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,928
  3. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,239
  4. 2023Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Promptingcited by 1,081
  5. 2023Steering Language Models With Activation Engineeringcited by 824
  6. 2024OLMo: Accelerating the Science of Language Modelscited by 599
  7. 2021Do Prompt-Based Models Really Understand the Meaning of their Prompts?cited by 324
  8. 2024Semantics-Adaptive Activation Intervention for LLMs via Dynamic Steering Vectorscited by 61
  9. 2025Steering Large Language Model Activations in Sparse Spacescited by 58
  10. 2022The Goldilocks of Pragmatic Understanding: Fine-Tuning Strategy Matters for Implicature Resolution by LLMscited by 57
  11. 2025Interpretable Steering of Large Language Models with Feature Guided Activation Additionscited by 32
  12. 2025Reading Between the Prompts: How Stereotypes Shape LLM's Implicit Personalizationcited by 24
5 more
  1. 2025Improving Reasoning Performance in Large Language Models via Representation Engineeringcited by 21
  2. 2025Non-literal Understanding of Number Words by Language Modelscited by 9
  3. 2024Pragmatic Competence Evaluation of Large Language Models for the Korean Languagecited by 8
  4. 2025Mitigating Memorization in LLMs using Activation Steeringcited by 7
  5. 2024Pragmatic inference of scalar implicature by LLMscited by 5

Built on by

Not cited yet.