arXiv Computation and Language By Leandra Fichtel, Janek Prange, Henning Wachsmuth

Attribute-Based Activation Steering of LLMs for Group-Specific Explanation Generation

Read the original on arXiv Computation and Language →

The paper proposes a method to steer large language models (LLMs) to generate explanations tailored to specific target groups. It first identifies group-specific attributes related to explanatory style and knowledge, then uses activation engineering to compute steering vectors that are added to the LLM’s internal activations during inference. Experiments show that this attribute-based steering improves specificity and factuality of explanations compared to prompting and existing steering baselines, and a human study confirms better tailoring to target groups.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv AI
Sep 11

XAI-Arena: Can LLMs Assess the Quality of XAI Explanations?

XAI-Arena proposes using large language models (LLMs) as judges to evaluate the quality of explainable AI (XAI) explanations, aiming for reproducibility, scalability, and multidimensional assessment. The framework assesses dimensions such as simplicity, clarity, task adequacy, trust calibration, actionability, transparency, faithfulness, and overall interpretability across different datasets, models, and stakeholder personas. Human validation shows a strong positive correlation between LLM-generated and human ratings (Spearman's rho = .693, p < .001), supporting the viability of LLM-based evaluations.

By Yanfei Hu Fleischhauer, Alona Zharova, Nadja Klein, Stefan Feuerriegel
arXiv Computation and Language
Sep 7

From Plausible to Actionable: A Position on LLM Self-Explanations

The paper discusses how Large Language Models can produce natural language self‑explanations that appear plausible but may not accurately reflect the model’s reasoning. It critiques current evaluation methods for such explanations and offers practical guidelines to assess their plausibility and faithfulness. Additionally, it argues that evaluation should also consider the actionability of these explanations, showing how they can aid decision‑making for various stakeholders.

By Elize Herrewijnen, Benedetta Muscato, Gizem Gezici, Fosca Giannotti
arXiv Computation and Language
Aug 31

Human Label Variation as Stable Signal: Learning Annotator-Specific Explanation Behavior via Cross-Annotator Preference Optimization

The paper investigates whether large language models can learn and reproduce annotator‑specific label‑explanation behavior, using two sentence‑pair tasks with four annotators each. It finds that individual annotator patterns are weak at the single‑annotation level but become detectable after reducing input‑content effects and aggregating across annotators. The authors propose cross‑annotator preference optimization (CAPO), which improves upon prompting and supervised fine‑tuning by better capturing annotator‑specific reasoning while maintaining stable attribution.

By Beiduo Chen, Pingjun Hong, Ziyun Zhang, Benjamin Roth, Anna Korhonen, Barbara Plank