Today's large language models (LLMs) routinely generate coherent, grammatical and seemingly meaningful paragraphs of text.
This achievement has led to speculation that these networks are -- or will soon become -- "thinking machines", capable of performing tasks that require abstract knowledge and reasoning.
Here, we review the capabilities of LLMs by considering their performance on two different aspects of language use: 'formal linguistic competence', which includes knowledge of rules and patterns of a given language, and 'functional linguistic competence', a host of cognitive abilities required for language understanding and use in the real world.
Drawing on evidence from cognitive neuroscience, we show that formal competence in humans relies on specialized language processing mechanisms, whereas functional competence recruits multiple extralinguistic capacities that comprise human thought, such as formal reasoning, world knowledge, situation modeling, and social cognition.
In line with this distinction, LLMs show impressive (although imperfect) performance on tasks requiring formal linguistic competence, but fail on many tests requiring functional competence.
Based on this evidence, we argue that (1) contemporary LLMs should be taken seriously as models of formal linguistic skills; (2) models that master real-life language use would need to incorporate or develop not only a core language module, but also multiple non-language-specific cognitive capacities required for modeling thought.
Overall, a distinction between formal and functional linguistic competence helps clarify the discourse surrounding LLMs' potential and provides a path toward building models that understand and use language in human-like ways.
"Electronic Control Units (ECUs), which are responsible for the smooth operation of a vehicle’s engine and predominantly sought from China, are embedded with SIM cards before being sent to car manufacturers as sealed units, according to a serving security source.
The manufacturers fit the parts in cars without opening them because of various warranty and commercial agreements, the serving security source said."
And further down the page:
"How SIM cards can be used to track vehicles
SIM cards are commonplace in vehicles, installed by car manufacturers for connectivity purposes, or simply to feed back data on the performance of a car. Most cars will have at least a 2G connection. But intelligence officials warn they have created an increased vulnerability to eavesdropping from hostile states.
Most commercial trackers are essentially SIM cards attached to batteries and every geolocation tracker that sends data via a cellular network requires one. A built-in navigation system in a car with the ability to give live traffic updates will need a SIM card to connect to a signal.
However, the SIMs allegedly embedded in ECUs by Chinese companies are put there without the car manufacturers’ knowledge and appear to give the Chinese suppliers the ability to connect to the car and collect data – including where it has been, how long it has been stopped in a certain place, and even the way in which it has been driven."
Today's large language models (LLMs) routinely generate coherent, grammatical and seemingly meaningful paragraphs of text.
This achievement has led to speculation that these networks are -- or will soon become -- "thinking machines", capable of performing tasks that require abstract knowledge and reasoning.
Here, we review the capabilities of LLMs by considering their performance on two different aspects of language use: 'formal linguistic competence', which includes knowledge of rules and patterns of a given language, and 'functional linguistic competence', a host of cognitive abilities required for language understanding and use in the real world.
Drawing on evidence from cognitive neuroscience, we show that formal competence in humans relies on specialized language processing mechanisms, whereas functional competence recruits multiple extralinguistic capacities that comprise human thought, such as formal reasoning, world knowledge, situation modeling, and social cognition.
In line with this distinction, LLMs show impressive (although imperfect) performance on tasks requiring formal linguistic competence, but fail on many tests requiring functional competence.
Based on this evidence, we argue that (1) contemporary LLMs should be taken seriously as models of formal linguistic skills; (2) models that master real-life language use would need to incorporate or develop not only a core language module, but also multiple non-language-specific cognitive capacities required for modeling thought.
Overall, a distinction between formal and functional linguistic competence helps clarify the discourse surrounding LLMs' potential and provides a path toward building models that understand and use language in human-like ways.