Masked Diffusion for Travel Diaries Synthesis

Pascal Jutras-Dubé; Raziq Zaman; Fabian Bastin; Cinzia Cirillo; Ruqi Zhang

cs.LGOct 12, 2026

Masked Diffusion for Travel Diaries Synthesis

Authors: Pascal Jutras-Dubé, Raziq Zaman, Fabian Bastin, Cinzia Cirillo, Ruqi Zhang

Organizations: Department of Computer Science, Purdue University, West Lafayette, United States · Department of Civil and Environmental Engineering, University of Maryland, College Park, United States · Department of Computer Science and Operations Research, Université de Montréal, Montréal, Canada

Abstract

Because household travel surveys combine household characteristics, variable-length person rosters, vehicle rosters, and trip chains, existing synthesis pipelines split them across separate models or flatten them into cropped fixed-length vectors. Representation, rather than modeling, becomes the bottleneck: splitting severs cross-segment dependence, and cropping truncates the household-structure distribution. We instead tokenize the diary and keep the object whole: each household-day becomes one token sequence with count tokens for vehicles, persons, and trips, and distinct symbols for padded, hidden, unobserved, and inapplicable positions. We introduce DiaryBridge, a masked diffusion model that bridges the fully hidden sequence to the distribution of complete diaries by unmasking positions in any order. One trained model, conditioned on any observed subset, performs synthesis and partial-diary completion alike. Experiments on real data extracted from the 2017/2018 Regional Travel Survey of the Baltimore region show that DiaryBridge generates travel reflecting vehicle availability and employment status, recovers shared trips using other household members' observed diaries, and better reproduces aggregate demand across modes, purposes, and departure periods than the learned baselines.

Figures & tables

Figure 1: The top ribbon shows one encoded sample in flattened canonical order. The lower panels unfold the same sample into household, vehicle, person, and person-wise trip blocks. The example is illustrative: it uses reduced caps (Vmax = 3, Pmax = 4, Tmax = 4) and simplified attribute labels; the attributes used in the experiments are listed in Table 2 .

Figure 2: Simplified illustration of the two-phase count-first discrete diffusion sampler.

	Records	Mean	99th pct.	Max	After caps
Households	8,313	—	—	—	8,201
Persons per household	17,858	2.15	6	10	17,552
Vehicles per household	15,311	1.84	5	12	14,913
Trips per person	59,913	3.35	11	41	57,759

Table 1: Linked BMC extract of the 2017/2018 RTS. Records, means, and percentiles describe all 8,313 household-days; the last column gives the records retained after the capacity caps of Section 5.1.2 .

Block	Attributes (type)	Slots
Household	area type, home type, home ownership, household income (cat.)	$1\times 4$
Vehicle	body type, fuel type, model year (binned), toll transponder (cat.)	$V_{\max}\times 4$
Person	age (binned), gender, employment status, license, disability (cat.)	$P_{\max}\times 5$
Trip	origin activity, destination activity, mode (cat.), departure time (binned), co-travelers (bitmask)	$P_{\max}T_{\max}\times 5$
Counts	$c_{V}$ , $c_{P}$ , $c_{T,1},\dots,c_{T,P_{\max}}$	$2+P_{\max}$

Table 2: Feature selection.

Method	Household	Person	Household–	Member
	$\times$ person	$\times$ trip	person–trip	pairs
	$k=2$	$k=2$	$k=3$	$k=2$
Resample	$.138\pm.005$	$.359\pm.016$	$.744\pm.027$	$.165\pm.009$
IPF/IPU (train controls)	$.135\pm.008$	$.353\pm.008$	$.743\pm.026$	$.163\pm.009$
BN	$\mathbf{.229\pm.015}$	$\underline{.921\pm.009}$	$\underline{1.646\pm.026}$	$.583\pm.091$
VAE	$.481\pm.065$	$1.168\pm.095$	$2.154\pm.240$	$.883\pm.051$

Table 3: Cross-panel SRMSE at the indicated feature orders (lower is better; mean

\pm

std). Bold and underlined entries mark the best and second-best learned generators, respectively. The number of attributes is

k

	Conditional TVD		Empirical rate (%)
Method	Mode given	Purpose given	Zero-vehicle	Work trips by
	resources	employment	households with	nonworkers
			driver trips
Real	—	—	0.0	2.8
Diary donor	$.035\pm.009$	$.050\pm.006$	$14.1\pm 3.0$	$19.6\pm 1.1$
BN	$\underline{.095\pm.005}$	$\underline{.060\pm.006}$	$69.7\pm 2.2$	$9.8\pm 0.8$

Table 4: Travel conditional on resources and employment (mean

\pm

SD). Lower TVD and empirical rates closer to the real reference are better. Bold/underline: best/second learned generator.

Figure 3: Missing-member recovery with and without observed peer diaries. Recovery is the percentage of 892 real peer references with a compatible generated target trip. Bars show five-run means and error bars sample SD; attributes and trip counts are observed in both conditions.

Method	Trips/person	Demand error (%)
Real	3.419	—
Diary donor	$3.407\pm 0.061$	$16.55\pm 0.68$
BN	$\mathbf{3.278\pm 0.062}$	$39.95\pm 1.62$
AR	$1.700\pm 0.126$	$52.92\pm 3.87$
DiaryBridge count-first	$2.306\pm 0.340$	$\underline{37.38\pm 8.57}$
DiaryBridge joint	$\underline{2.752\pm 0.246}$	$\mathbf{28.67\pm 4.70}$

Table 5: Travel generated for a fixed population of 820 households (mean

\pm

SD). Trips/person is ranked by closeness to real data; lower demand error is better. Bold/underline: best/second learned generator.

Method	Trip mode	Vehicle count
Resample	.945	.999
IPF/IPU (train controls)	.976	1.001
BN	.613	$\underline{1.020}$
VAE	.737	.690
CTGAN	.522	.431
AR	$.833$	.971

Table 6: Train-on-synthetic, test-on-real macro-F1 relative to training on real data (higher is better). Bold/underline: best/second learned generator.

	Encoded errors		Chain error	Coverage
Method	Count–PAD	Invalid	Activity	Sampling
		co-traveler	chaining	zeros
Resample	0	0	.05	0
IPF/IPU (train controls)	0	0	.06	0
BN	$\mathbf{0}$	$\mathbf{0}$	$\mathbf{0}$	$38.4\pm 6.0$
VAE	$96.5\pm 0.7$	$4.5\pm 0.7$	$64.7\pm 6.4$	$\mathbf{57.6\pm 4.9}$

Table 7: Raw encoded errors, activity-chaining errors, and sampling-zero coverage (%, mean

\pm

SD). Lower errors and higher coverage are better. Bold/underline: best/second learned generator.

Conditional Generative Modeling

Conditional generative modeling learns to produce data distributions steered by labels, inputs, text, or physical and structural constraints.

See all 393 papers

Generating Individual Travel Diaries Using Large Language Models Informed by Census and Land-Use Data
Sep 7, 2025Sepehr Golrokh Amin, Devin Rhoads, Fatemeh Fakhrmoosavi +2Synthetic Data Generation Synthetic Data Evaluation
School of Civil and Environmental Engineering, University of Connecticut, United States
AugMask: Training Diffusion Models on Incomplete Tabular Data via Stochastic Augmentation and Masking
Jun 2, 2026Jungkyu Kim, Taeyoung Park, Kibok LeeLearning with Missing Data Tabular Diffusion Models
Department of Statistics and Data Science, Yonsei University, South Korea.
Feasible and Novel Synthetic Population Generation with Tabular and Sequential Travel Attributes
Aug 16, 2026Farbod Abbasi, Zachary Patterson, Bilal FarooqSynthetic Population Generation Synthetic Data Generation
Concordia University, Montreal, Quebec, Canada · Toronto Metropolitan University, Toronto, Ontario, Canada

Sep 7, 2025J/K move · Enter open · S save

Generating Individual Travel Diaries Using Large Language Models Informed by Census and Land-Use Data

Sepehr Golrokh Amin, Devin Rhoads, Fatemeh Fakhrmoosavi, Nicholas E. Lownes+1

School of Civil and Environmental Engineering, University of Connecticut, United States

Synthetic Data Generation Synthetic Data Evaluation Text Generation

Open paper AskPDF arXiv

This study introduces a Large Language Model (LLM) scheme for generating key attributes of travel diaries in agent-based transportation models, including purpose, mode and distance, to assess the underlying viability of LLMs for activity generation tasks. While traditional approaches rely on large quantities of proprietary household travel surveys, our method generates personas stochastically from open-source American Community Survey (ACS) and Smart Location Database (SLD) data, then synthesizes diaries through direct prompting. Our study features a novel one-to-cohort realism score: a composite of four metrics (Trip Count Score, Interval Score, Purpose Score, and Mode Score) validated against the Connecticut Statewide Transportation Study (CSTS) diaries, matched across demographic variables. Our validation utilizes Jensen-Shannon Divergence to measure distributional similarities between generated and real diaries. When compared to diaries generated with classical methods (Negative Binomial for trip generation; Multinomial Logit for mode/purpose) calibrated on the validation set, LLM generated diaries achieve comparable overall realism (LLM mean: 0.692 vs. 0.628). The LLM excels in determining trip purpose, and its trip mode predictions demonstrate greater consistency (a narrower Realism Score distribution). Meanwhile, classical models lead to better numerical estimates of trip count and activity duration. Aggregate validation confirms the LLM's statistical representativeness (LLM mean: 0.779 vs. 0.706), demonstrating LLM's zero-shot viability and establishing a quantifiable metric of diary realism for future synthetic diary evaluation systems.

Abstract

Conditional Generative Modeling

Explore similar work

Generating Individual Travel Diaries Using Large Language Models Informed by Census and Land-Use Data

AugMask: Training Diffusion Models on Incomplete Tabular Data via Stochastic Augmentation and Masking

Feasible and Novel Synthetic Population Generation with Tabular and Sequential Travel Attributes