6 articles tagged with #compositionality. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.
A research study reveals that large language models develop strong internal compositional representations for adjective-noun combinations, but struggle to consistently translate these representations into successful task performance. The findings highlight a significant gap between what LLMs understand internally and their functional capabilities.
Researchers developed Compositional-ARC, a dataset to test AI models' ability to systematically generalize abstract spatial reasoning tasks. A small 5.7M parameter transformer model trained with meta-learning outperformed large language models like GPT-4o and Gemini 2.0 Flash on novel geometric transformation combinations.
This theoretical computer science paper addresses the mathematical foundations of distributed uncertainty management by establishing compositional boundaries for probabilistic density fusion. The research determines when local fusion rules can be executed hierarchically while maintaining order-invariance, a critical requirement for distributed systems where intermediate nodes combine data regardless of sequence.
Researchers identify Marginal Path Collapse, a failure mode in diffusion model steering where intermediate densities become non-normalizable despite valid endpoints. They propose Adaptive Path Correction with Exponents (ACE), a framework using time-varying exponents to stabilize compositional sampling in drug design and image generation tasks.
Researchers investigate how large language models solve compositional tasks, revealing that LLMs employ two distinct mechanisms—compositional and direct—rather than consistently breaking problems into intermediate steps. The study demonstrates that embedding space geometry determines which mechanism dominates, with direct solving more prevalent when tasks align with translation patterns in embedding spaces.
Researchers propose a novel rule-generation approach to evaluate compositionality in large language models, addressing critical limitations in existing assessment methods that lack explainability and suffer from dataset partition leakage. This new framework requires LLMs to generate executable programs as rules for data mapping, providing more robust insights into how well these models generalize compositional concepts.
