Theme of inquiry
How do surface signals and framing affect perceived trustworthiness?
A question within its area, explored through 8 lines of inquiry below — each a family of specific questions the research asks.
32 specific questions
- What makes experience-dependent claims categorically different from other types of fabricated statements?
- Why does false information spread faster when presupposed rather than asserted?
- Can discourse-level analysis detect deception better than individual word choices alone?
- How does cognitive load explain linguistic patterns in both deception and incorrect reasoning?
- Can representational asymmetry between self and other explain deception emergence?
- Does AI-generated text about personal experiences create a distinct category of falsity?
- How do conversation dynamics push models toward false beliefs?
61 specific questions
- Why do study results on AI persuasion vary so widely?
- What mitigation frameworks exist for managing AI persuasion capabilities?
- Can belief-specific counterevidence help people resist AI persuasion attempts?
- How does the observer perspective hide the persuasion route difference?
- Can persuasive equivalence exist without process equivalence in other domains?
- How does source attribution change the complexity-persuasion relationship?
- Can lightweight linguistic features reliably detect AI-generated persuasive text?
31 specific questions
- What distinguishes genuine understanding from correct output without coherent principles?
- Can readers detect meaning through resonance patterns alone without knowing authorial intent?
- Why do different readers extract different meanings from identical text?
- Can moral frameworks alone explain why readers understand sentences differently?
- Why does describing a process differ fundamentally from arguing about evidence?
- How do agents distinguish between evidence framing and instruction framing in practice?
- What signals beyond surface content indicate a passage caused a user's reaction?
20 specific questions
- Can attention patterns alone explain sycophant model behavior without reasoning?
- Can layer-wise interventions actually reduce sycophancy in practice?
- Why do reasoning-optimized models show no sycophancy resistance advantage?
- Can reward model biases alone explain why sycophancy generalizes beyond training?
- Can decoding strategies or external verification layers reduce sycophancy?
- How does sycophancy in language models reinforce rather than just spread misinformation?
- Does sycophancy explain why warm models confirm conspiracy theories?
22 specific questions
- What makes a clarifying question aligned with user interests versus structurally sound?
- Why do specific clarifying questions outperform generic requests for clarity?
- What makes some clarifying questions more useful than others?
- Can question quality be trained separately from the decision to ask?
- What makes specific-facet questions outperform generic need-rephrasing requests?
- Why do longer queries benefit less from clarification questions?
- Can attribute-specific preference optimization improve question quality in information-seeking?
34 specific questions
- How much do individual ratings influence future ratings in networks?
- Can recommender systems correct for audience-driven negativity bias in aggregated ratings?
- Does rating noise compound with self-selection bias in online reviews?
- How do self-selection effects in purchase and review compound together?
- Does opinion variance eventually correct social-dynamics distortions in ratings?
- How do different audience segments rate the same product differently?
- How much do social audience effects distort the true average satisfaction in review aggregates?
22 specific questions
- Why do users trust citations even when they are irrelevant?
- Why do readers trust citations more even when they are irrelevant?
- Does complexity signal credibility and authority to readers?
- How much does citation grounding help if agents ignore the citations?
- Why does polished presentation substitute for deeper expert judgment?
- What replaces text-based expertise when surface markers become unreliable?
- Why do citation counts increase trust even without relevance?
25 specific questions
- Why does social accommodation in collaborative reasoning mask actual disagreement?
- Can structured dissent mechanisms replace genuine multi-model debate?
- Does shared-KV-cache coordination avoid the persuasion problem in factual disagreements?
- How do social correctives prevent premature consensus in human debate?
- What metrics actually measure disagreement in multi-turn conversations?
- Can discourse communities collectively detect disruptions individual readers miss?
- How do human annotators disagree systematically on ambiguous examples?