While Vision Language Models (VLMs) learn conceptual representations, in the form of generalized knowledge, during training, they are typically used to analyze individual instances. When evaluation instances are atypical, this paradigm results in tension between two priors in the model. The first is a pragmatic prior that the textual and visual input are both relevant, arising from VLM finetuning on congruent inputs; the second is a semantic prior that the conceptual representation is generally true for instances of the category. In order to understand how VLMs trade off these priors, we introduce a new evaluation dataset, VISaGE, consisting of both typical and exceptional images. In carefully balanced experiments, we show that conceptual understanding degrades when the assumption of congruency underlying the pragmatic prior is violated with incongruent images. This effect is stronger than the effect of the semantic prior when querying about individual instances.

Pre-trained language models have achieved remarkable success across diverse applications but remain susceptible to spurious, concept-driven correlations that impair robustness and fairness. In this work, we introduce CURE, a novel and lightweight framework that systematically disentangles and suppresses conceptual shortcuts while preserving essential content information. Our method first extracts concept-irrelevant representations via a dedicated content extractor reinforced by a reversal network, ensuring minimal loss of task-relevant information. A subsequent controllable debiasing module employs contrastive learning to finely adjust the influence of residual conceptual cues, enabling the model to either diminish harmful biases or harness beneficial correlations as appropriate for the target task. Evaluated on the IMDB and Yelp datasets using three pre-trained architectures, CURE achieves an absolute improvement of +10 points in F1 score on IMDB and +2 points on Yelp, while introducing minimal computational overhead. Our approach establishes a flexible, unsupervised blueprint for combating conceptual biases, paving the way for more reliable and fair language understanding systems.

Human languages vary widely in how they encode information within circumscribed semantic domains (e.g., time, space, color, human body parts and activities), but little is known about the global structure of semantic information and nothing about its relation to human communication. We first show that across a sample of ~1,000 languages, there is broad variation in how densely languages encode information into their words. Second, we show that this language information density is associated with a denser configuration of semantic information. Finally, we trace the relationship between language information density and patterns of communication, showing that informationally denser languages tend toward (1) faster communication, but (2) conceptually narrower conversations within which topics of conversation are discussed at greater depth. These results highlight an important source of variation across the human communicative channel, revealing that the structure of language shapes the nature and texture of human engagement, with consequences for human behavior across levels of society.

When performing a task alone, humans achieve a certain level of performance. When humans are assisted by a tool or automation to perform the same task, performance is enhanced-- augmented. Recently developed cognitive systems are able to perform cognitive processing at or above the level of a human in some domains. When humans work collaboratively with such "cogs" in a human/cog ensemble, we expect augmentation of cognitive processing to be evident and measurable. This paper shows the degree of cognitive augmentation depends on the nature of the information the cog contributes to the ensemble. Results of an experiment are reported showing conceptual information is the most effective type of information resulting in increases in cognitive accuracy, cognitive precision, and cognitive power.

This paper presents a computational model for conceptual shifts, based on a novelty metric applied to a vector representation generated through deep learning. This model is integrated into a co-creative design system, which enables a partnership between an AI agent and a human designer interacting through a sketching canvas. The AI agent responds to the human designer's sketch with a new sketch that is a conceptual shift: intentionally varying the visual and conceptual similarity with increasingly more novelty. The paper presents the results of a user study showing that increasing novelty in the AI contribution is associated with higher creative outcomes, whereas low novelty leads to less creative outcomes.

The frame problem in representing natural-language information is discussed. It is argued that the problem is not restricted to problem-solving-type situations, in which it has mostly been studied so far, but also has a broader significance. A new solution to the frame problem, which arose within a larger system for representing natural-language information, is described. The basic idea is to extend the predicate calculus notation with a special operator, Unless, with peculiar properties. Some difficulties with Unless are described. THE FRAME PROBLEM This paper proposes a method for handling the frame problem in representing conceptual, or natural-language-type information.

The frame problem in representing natural-language information is discussed. It is argued that the problem is not restricted to problem-solving-type situations, in which it has mostly been studied so far, but also has a broader significance. A new solution to the frame problem, which arose within a larger system for representing natural-language information, is described. The basic idea is to extend the predicate calculus notation with a special operator, Unless, with peculiar properties. Some difficulties with Unless are described.

Report #SP-1395, System Development Corporation, Santa Monica, CA.