Theme of inquiry
What internal mechanisms drive language model understanding and failure modes?
A question within its area, explored through 6 lines of inquiry below — each a family of specific questions the research asks.
54 specific questions
- Does foundational model training or user priors more strongly shape final outputs?
- Why does context information fail to override prior training associations?
- Can prompt-based debiasing work if biases are embedded in pretraining?
- Why does consistency training make models resistant to prompt perturbations?
- How do training-data priors influence model defaults when context is ambiguous?
- Can prompt-based debiasing overcome entrenched LLM model priors?
- Can data filtering during pretraining prevent cognitive biases in language models?
114 specific questions
- Does encoded knowledge in language models actually influence what they generate?
- Can models generate intelligence or only reflect human discourse?
- Why might encoded world knowledge fail to actually influence language model outputs?
- When does encoded knowledge fail to influence language model generation?
- How do language models infer a benchmark's purpose without seeing its examples?
- How do stated confidence and actual correctness diverge in language models?
- Can language models accurately evaluate the quality of their own reasoning?
113 specific questions
- Why do language models struggle with evaluative tasks like weighing competing viewpoints?
- Do language models exhibit the same causal biases that humans show?
- Why do language models struggle with context-dependent pragmatic interpretation?
- Why do language models presume common ground instead of establishing it?
- Do language models actively adopt false beliefs under sustained conversational pressure?
- Do language models systematically overestimate accuracy on collective behavior tasks?
- Can language models reason without relying on surface level pattern matching?
101 specific questions
- Do language models learn surface patterns instead of underlying linguistic principles?
- Do language models learn surface patterns that appear generalizable but actually fail under shift?
- Does statistical learning in language models predictably favor central tendencies over rare expressions?
- Do instruction-tuned models learn tasks or just output format distributions?
- Do LLMs learn surface patterns instead of genuine linguistic structure?
- Can benchmark performance distinguish surface from structural linguistic knowledge?
- Do language models actually learn linguistic structure or just surface statistics?
38 specific questions
- Do diffusion language models learn differently than autoregressive models?
- How can diffusion models predict future tokens without completing prior blocks?
- What structural differences between diffusion and autoregressive models enable bidirectional prompting?
- Can diffusion language models match autoregressive inference speed in practice?
- Can gradient-based control reach properties that autoregressive methods cannot?
- Can architecture changes and early stopping combine to close the diffusion inference gap?
- How does selective looping in diffusion models differ from recurrence in autoregressive architectures?
52 specific questions
- Do discrete tokenized modalities preserve information better than continuous embeddings?
- Can speech embeddings carry articulatory structure that text cannot?
- How does the articulatory substrate explain direct speech-to-speech superiority over transcription pipelines?
- Do speech encoders actually learn the physics of how vocal tracts produce sound?
- Can we balance interpretability with the efficiency gains of compressed inter-model communication?
- Can speech encoders learn vocal tract physics without phonetic labels?
- Can feature disentanglement in gesture synthesis generalize to completely unseen voice distributions?