Theme of inquiry
How do language models represent meaning and build world understanding?
A question within its area, explored through 7 lines of inquiry below — each a family of specific questions the research asks.
95 specific questions
- Do language models learn surface patterns instead of underlying linguistic principles?
- Why do LLMs understand efficient language but fail to produce it?
- Do language models encode deep syntactic structure or only surface-level patterns?
- Do language models actually learn linguistic structure or just surface statistics?
- Do language models learn surface patterns that appear generalizable but actually fail under shift?
- Do newer language models diverge further from human lexical patterns?
- Do LLMs learn surface patterns instead of genuine linguistic structure?
59 specific questions
- Do language models actively adopt false beliefs under sustained conversational pressure?
- Can language models correct false assumptions or only reinforce them?
- Why do language models presume common ground instead of establishing it?
- Why do language models struggle with context-dependent pragmatic interpretation?
- Why do language models struggle with evaluative tasks like weighing competing viewpoints?
- Why do language models prefer accommodating false information over rejecting it?
- Do language models systematically overestimate accuracy on collective behavior tasks?
27 specific questions
- Can language models develop world models that ground meaning in causal reality?
- Do language models build world models or just task-specific heuristics?
- Can language models learn internal world models without explicit environment specifications?
- Do language models need words to think or just latent structure?
- Can LLM semantic representations exist without causally influencing their generation output?
- Do LLMs rely on surface statistical patterns instead of causal structure?
- What prevents LLM representations from causally influencing generation outputs?
29 specific questions
- Why do language models fail at implicit discourse relations while handling explicit connectives?
- Why do LLMs perform better on explicit discourse connectives than implicit relations?
- Why do explicit discourse connectives work when implicit relations fail?
- Why do explicit discourse connectives help LLMs but implicit relations cause failures?
- Can explicit connectives compensate for missing intentional tracking in LLMs?
- Can language models distinguish explicit from implicit discourse relations?
- Does chain-of-thought prompting overcome implicit meaning deficits in text analysis?
27 specific questions
- Why do language models fail when semantic content is stripped away?
- Why do LLMs fail at semantic generalization despite grammatical accuracy?
- Do LLMs struggle more with semantic accuracy than syntactic correctness across domains?
- Can LLMs improve at metaphor if they handle decoupled semantics better?
- What semantic information is necessary to preserve for sound LLM reasoning?
- Do metaphors work by decoupling meaning from linguistic associations?
- Why do LLMs fail at implicit elements in literary and poetic text?
39 specific questions
- Why does latent-level prediction beat token-level prediction for reasoning?
- Does the token prediction framing actually capture what human reasoning does?
- Why does token-level gradient targeting matter more than aggregate loss?
- Why do standard next-token prediction models struggle with conversational initiative?
- Can next-token prediction train models to optimize for communication efficiency?
- Does next-token prediction alone produce genuine functional language competence?
- How do meta-tokens help models learn when to generate reasoning versus commit predictions?
34 specific questions
- How does the articulatory substrate explain direct speech-to-speech superiority over transcription pipelines?
- Can speech embeddings carry articulatory structure that text cannot?
- Can feature disentanglement in gesture synthesis generalize to completely unseen voice distributions?
- Do speech models learn the articulatory processes that produce acoustic signals?
- Do speech encoders actually learn the physics of how vocal tracts produce sound?
- How does causal multimodal modeling differ from encoder-decoder architectures?
- Do discrete tokenized modalities preserve information better than continuous embeddings?