Probing layer-wise embeddings with three prompt-variant families reveals a consistent grounding-reasoning-decoding structure in LLaVA-1.5, LLaVA-Next, and Qwen2-VL, with base LLM architecture shifting layer allocation.
Probing Large Language Models from A Human Behavioral Perspective
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Large Language Models (LLMs) have emerged as dominant foundational models in modern NLP. However, the understanding of their prediction processes and internal mechanisms, such as feed-forward networks (FFN) and multi-head self-attention (MHSA), remains largely unexplored. In this work, we probe LLMs from a human behavioral perspective, correlating values from LLMs with eye-tracking measures, which are widely recognized as meaningful indicators of human reading patterns. Our findings reveal that LLMs exhibit a similar prediction pattern with humans but distinct from that of Shallow Language Models (SLMs). Moreover, with the escalation of LLM layers from the middle layers, the correlation coefficients also increase in FFN and MHSA, indicating that the logits within FFN increasingly encapsulate word semantics suitable for predicting tokens from the vocabulary.
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
How Multimodal LLMs Solve Image Tasks: A Lens on Visual Grounding, Task Reasoning, and Answer Decoding
Probing layer-wise embeddings with three prompt-variant families reveals a consistent grounding-reasoning-decoding structure in LLaVA-1.5, LLaVA-Next, and Qwen2-VL, with base LLM architecture shifting layer allocation.