Bibliographic record
Abstract
Personalized trip planning is becoming increasingly important in modern urban mobility, as travelers expect contextaware, efficient, and adaptive recommendations tailored to their individual needs. This paper investigates how agentic AI can enable such personalization through autonomous reasoning and dynamic itinerary generation. A case study in the Greater Toronto Area demonstrates how specialized agents collaboratively generate persona-specific itineraries while enforcing guardrails for input validation and output verification. Experimental evaluation across multiple large language models (LLMs) and large reasoning models (LRMs) reveals distinct performance tradeoffs: GPT-4 models offer faster responses and concise itineraries suitable for real-time use, while GPT-5 models produce more detailed and consistent plans but with higher latency. The openweight gpt-oss- 120 b achieves stability comparable to GPT5 but with significantly reduced response time, offering a balanced compromise between efficiency and reliability. A composite score integrating response time, itinerary steps, and semantic similarity confirms these findings. Overall, GPT-4 achieves the most balanced performance, GPT-5 favors stability and detail, and gpt-oss-120b provides a strong middle ground. Personalevel analysis shows that accessibility needs and multimodal preferences significantly influence itinerary design, while triplength analysis reveals that journey distance affects itinerary detail and stability. These findings underscore the potential of agentic AI to deliver personalized trip planning experiences.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".