6 papers
Isotropy Cliffs: The Geometric Signature of Decision-Making in Large Language Models
Okan S. Coskun, Florian Rottach, Carsten Eickhoff +1
We investigate the geometry of decision-making in Multiple Choice Question Answering (MCQA) through the lens of isotropy. Analyzing five open-weight models across diverse datasets,…
When to Call an Apple Red: Humans Follow Introspective Rules, VLMs Don't
Jonathan Nemitz, Carsten Eickhoff, Junyi Jessy Li +3
Understanding when Vision-Language Models (VLMs) will behave unexpectedly, whether models can reliably predict their own behavior, and if models adhere to their introspective reaso…
Pathway to Relevance: How Cross-Encoders Implement a Semantic Variant of BM25
Meng Lu, Catherine Chen, Carsten Eickhoff
Mechanistic interpretation has greatly contributed to a more detailed understanding of generative language models, enabling significant progress in identifying structures that impl…
TRIM: Achieving Extreme Sparsity with Targeted Row-wise Iterative Metric-driven Pruning
Florentin Beck, William Rudman, Carsten Eickhoff
Large Language Models (LLMs) present significant computational and memory challenges due to their extensive size, making pruning essential for their efficient deployment. Existing…
What Do VLMs NOTICE? A Mechanistic Interpretability Pipeline for Gaussian-Noise-free Text-Image Corruption and Evaluation
Michal Golovanevsky, William Rudman, Vedant Palit +2
Vision-Language Models (VLMs) have gained community-spanning prominence due to their ability to integrate visual and textual inputs to perform complex tasks. Despite their success,…
MechIR: A Mechanistic Interpretability Framework for Information Retrieval
Andrew Parry, Catherine Chen, Carsten Eickhoff +1
Mechanistic interpretability is an emerging diagnostic approach for neural models that has gained traction in broader natural language processing domains. This paradigm aims to pro…