Entity · paper

Creativity, honesty and designed forgetting emerge in small hyperbolic language models

paperactiveprovisionalcreativity-honesty-and-designed-forgetting-emerge-in-small-hyperbolic-language-models-19b75cc3·1 events·first seen 10h ago

Aliases: Creativity, honesty and designed forgetting emerge in small hyperbolic language models

Co-occurring entities

Fleiss' Kappa

More like this (12)

Human Creativity Benchmark Decomposing Factual Sycophancy in Language Models: How Size and Instruction Tuning Shape Robustness Speaking the Language of Science: Toward a General-Purpose Generative Foundation Model for the Natural Sciences Message Passing Language Models Space-Efficient Language Generation in the Limit Apparent Psychological Profiles of Large Language Models are Largely a Measurement Artifact Co-LMLM: Continuous-Query Limited Memory Language Models Where Do Models Find Happiness? Emotion Vectors in Open-Source LLMs Verbalizable Representations Form a Global Workspace in Language Models Self-Compacting Language Model Agents encoder-only language models Recursive Language Models (RLMs)

Recent events (1)

5arXiv · cs.CL·10h ago·source ↗

Small hyperbolic language models enable creativity, honesty detection, and designed forgetting for companion AI

Researchers present a suite of three small language models (146M to 3B parameters) built on a hyperbolic geometric substrate, targeting trustworthy companion AI. A 146M behavioral auditor achieves 90.7% binary-compliance accuracy and outperforms a frontier zero-shot judge (AUROC 0.804 vs 0.721) in detecting sycophancy, dependence-fostering, and confabulated memories across unseen generator families. A creative frame-seeder wins 100% of 311 pairwise comparisons over prompting baselines, and a memory OS implements exponential decay-based 'designed forgetting.' The work proposes a small-model alternative to frontier-scale approaches for companion AI safety and personalization.

Evaluation and Benchmarking AI Safety Research Creativity, honesty and designed forgetting emerge in small hyperbolic language models Fleiss' Kappa +1 more