arXiv AI
The Pain Axis: LLMs Represent Self-Directed Harm and Act to Relieve It
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Oct 7
Too Categorical to be Human: Emotion Concepts in LLMs and Humans
arXiv:2508.05880v3 Announce Type: replace-cross Abstract: Understanding human emotions is central to user-facing AI applications, safety alignment, and the simulation of human behavior. As emotional...
arXiv AI
Oct 21
From Encoding to Generation: Towards Empathetic Large Language Models
arXiv:2610.27603v1 Announce Type: new Abstract: Empathetic interaction involves two distinguishable capacities: representing another person's emotional state (cognitive empathy) and generating an app...
arXiv AI
Jun 26
Where Do Models Find Happiness? Emotion Vectors in Open-Source LLMs
arXiv:2606. 26987v1 Announce Type: cross Abstract: Recent work identified emotion vectors in Claude Sonnet 4.
arXiv Computation and Language
Aug 27
When Emotion Becomes Trigger: Emotion-style dynamic Backdoor Attack Parasitising Large Language Models
The paper introduces Paraesthesia, a dynamic backdoor attack that uses emotionally styled inputs as triggers for large language models. By mapping target emotions into a valence–arousal space and rewriting a small subset of clean samples, the attack achieves over 98% success while minimally affecting clean performance. Experiments on four major LLMs show that the trigger cannot be fully explained by token-level cues and remains robust against several filtering and mitigation techniques.
arXiv Machine Learning
Sep 22
Replicating the Geometry of Emotion Representations in a Base Open-Weights Model
Sofroniew et al. (2026) showed that emotion concepts in Claude Sonnet 4.5 are encoded as vectors whose geometry mirrors human affect psychology. This study replicates that finding using the base pretrained model google/gemma-2-27b, generating 205,200 Claude Sonnet 4.5 stories, extracting 171 emotion vectors, and recovering a similar affective circumplex with principal components explaining comparable variance. The analysis further identifies a sharp geometric seam at layers 22‑26, demonstrates that much of the geometry already exists in static token embeddings, and shows that the geometry predicts token‑level co‑activation with high correlation.
arXiv AI
Aug 20
Nine Emotion Centroids: A Label-Free Valence Axis That Transfers Across Four Modalities
The paper demonstrates that a single internal direction in modern language models—called the valence axis (V-axis)—captures how positive or negative a sentence feels. By using only nine emotion category names and 50 short narrative paragraphs per emotion, the authors identify this axis via principal component analysis of frozen encoder embeddings, achieving 93% of supervised performance on SST‑2 and strong correlations with human valence ratings across images, audio, and brain recordings. The method transfers across modalities without target‑modality labels, but works only for continuous attributes and is specific to certain model families.