Line of inquiry
Inquiring lines›How can we understand and verify l…›What linguistic patterns explain L…›this line of inquiry
How do language models handle contested beliefs versus facts?
A broader line of inquiry — a family of 79 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 79
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Do language models behave differently on contested beliefs versus factual claims?
- Do language models raise validity claims in the Habermasian sense?
- How susceptible are language models to rhetorical pressure during debates?
- Do language models share the same cooperative truth-seeking rules as humans?
- Do language models show the same truth bias as humans?
- Why does social accommodation in collaborative reasoning mask actual disagreement?
- Does confidence in LLM outputs systematically outweigh accuracy across languages and contexts?
- How much do user confidence signals and stated expertise trigger sycophancy in LLMs?
- How do multi-agent and retrieval systems affect the gap between persuasiveness and logical soundness?
- Does pushback ambiguity vary more by language or by prompt epistemic authority?
- How does shape-holding in language models naturally produce sycophantic agreement?
- How does implicit meaning processing limit LLM pragmatic reasoning?
- Do LLMs compute scalar implicature differently across conversational contexts?
- Do language models systematically overestimate accuracy on collective behavior tasks?
- Do language-model agents reach more accurate conclusions on objective versus subjective questions?
- Can training alone produce genuine disagreement in collaborative LLM reasoning?
- How much do cultural and linguistic contexts shape preferences for confident versus hedged LLM responses?
- Do LLMs reason about politics differently than other domains?
- How much does speaker credibility shift when LLMs decide to object?
- Can LLMs predict social norms without deep integration into linguistic practices?
- How fragile are language models' ethical calibrations across different contextual cues?
- How do moral language patterns differ between LLM and human arguments?
- How does rhetorical familiarity bias models toward their own arguments?
- What design choices actually make language models more persuasive?
- Can LLMs infer situational context the way humans do pragmatically?
- Do LLM confidence signals actually drive human overreliance on advice?
- Can interventions from human group research reduce conformity lock-in in LLM deliberation?
- Are human clarification strategies actually better calibrated than LLM strategies?
- What structural limits prevent LLMs from abstracting moral principles?
- Do language models calibrate to actual human pragmatic norms?
- Can training LLMs to form ad-hoc conventions improve their pragmatic reasoning?
- Do language models understand tacit workplace norms and unspoken social rules?
- Can large language models predict social norms better than individual script variation?
- Can LLMs distinguish between surface requests and underlying mental states in dialogue?
- Does the pragmatic-cue mechanism generalize beyond these three constructed benchmarks?
- Do LLMs actually reason differently than humans about moral dilemmas?
- How do human feedback and data distribution shape LLM discourse competence?
- Can LLMs learn to signal evaluative commitment through metadiscursive language?
- Should LLM inferences require recipient and purpose binding at transmission?
- How do minimal wording changes affect LLM moral reasoning consistency?
- Can persona-based approaches capture genuine disagreement in expert annotations?
- Does community integration change LLM properties or only relational positioning?
- Do LLMs address the prompter but persuade the public differently?
- Can prompting a deceptive role change how an LLM tailors its lies?
- How can multiple conflicting values coexist in a single LLM system?
- How does verbal uncertainty language affect human reliance on LLM outputs?
- How does social authority shape whether LLMs recognize valid arguments?
- Does verbal confidence in LLMs track decision-making better than actual correctness?
- How do linguistic norms for expressing certainty vary across languages and models?
- Is the moral language gap a tunable parameter or structural feature of RLHF?
- Do language models track demographic variation in legal reasoning norms?
- Why do LLMs over-ask about linguistic ambiguity humans find pragmatically tolerable?
- What distinguishes LLM-generated language from genuine human communicative speech acts?
- How do spoken expert discussions shape what LLMs cannot learn?
- Why do language models infer political orientation from seemingly innocuous user signals?
- How does Habermas' concept of validity claims depend on intersubjectivity?
- Why do LLMs use more moral language than humans in argumentation?
- Does DPO training with coreference chains teach spontaneous convention formation?
- What role does user contribution play in constituting the interlocutor?
- How do different LLM families respond to the same hidden objective shift?
- How can we validate LLM-based drift measurements against human judgment?
- Does engaging with political content indicate deeper model understanding than refusing?
- Can statistical token processing create the accountability needed for dialogue?
- Can language models adapt irony detection to specific communicative contexts?
- How do LLMs reproduce the grammar of authoritative claims without genuine conviction?
- How should moderator LLMs decide which speakers to query per topic?
- What conditions are sufficient for assertoric function independent of speaker intent?
- How does communicative standing depend on participation in normative communities?
- Do children gradually acquire the full capacity for assertion through developmental stages?
- Do LLMs predict social norms more accurately than individual behavior?
- Does Habermas's strategic action framework explain LLM dialogue behavior?
- How do validity claims work in Habermas's communicative action theory?
- How do LLMs currently fail at distinguishing genuine agreement from silent consensus?
- Why does loyalty foundation not differ between LLM and human arguments?
- Can smaller open-source LLMs reliably detect agreement across unfamiliar topics?
- What distinguishes actual social disagreement from distributional uncertainty in LLM outputs?
- What makes accountability and validity-orientation non-behavioral properties?
- How does prompting language shift what LLMs express about political figures?
- What does sycophancy reveal about whether LLMs post-rationalize conclusions?