Search Results

Documents authored by Huang, Haosheng

Document
Short Paper
DOI: 10.4230/LIPIcs.COSIT.2026.16
From Trajectories to Relations: Interpreting Overtake Manoeuvres with Large Language Models and Qualitative Spatial Representations (Short Paper)

Authors: Jana Verdoodt, Changbo Zhang, Lars De Sloover, Haosheng Huang, and Nico Van de Weghe

Published in: LIPIcs, Volume 393, 17th International Conference on Spatial Information Theory (COSIT 2026)

Abstract
This paper explores whether qualitative spatial encodings make overtake manoeuvres more interpretable to a large language model (LLM). Using simulated traffic trajectories from a microscopic road-network model, we encode the same overtake event at three levels of complexity through Point-Descriptor-Precedence (PDP) inequality matrices. An LLM is then used in a staged, zero-shot API workflow in which longitudinal and lateral descriptor encodings derived from these matrices are interpreted separately and then merged into a final manoeuvre interpretation. Outputs are assessed through expert-based qualitative evaluation focusing on relational correctness, spatial precision, completeness, and interpretive adequacy. For the selected case, the results suggest that the model can produce meaningful natural-language interpretations at all three levels, but that these change as complexity increases. The 2-point representation yields the clearest focal manoeuvre, the 5-point representation offers the best balance between contextual richness and interpretability, and the 10-point representation produces the most nuanced structural reading, although in a less concise form. The paper should be read as an exploratory study of representation effects rather than as a benchmark of LLM performance.
Cite as

Jana Verdoodt, Changbo Zhang, Lars De Sloover, Haosheng Huang, and Nico Van de Weghe. From Trajectories to Relations: Interpreting Overtake Manoeuvres with Large Language Models and Qualitative Spatial Representations (Short Paper). In 17th International Conference on Spatial Information Theory (COSIT 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 393, pp. 16:1-16:8, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{verdoodt_et_al:LIPIcs.COSIT.2026.16,
  author =	{Verdoodt, Jana and Zhang, Changbo and De Sloover, Lars and Huang, Haosheng and Van de Weghe, Nico},
  title =	{{From Trajectories to Relations: Interpreting Overtake Manoeuvres with Large Language Models and Qualitative Spatial Representations}},
  booktitle =	{17th International Conference on Spatial Information Theory (COSIT 2026)},
  pages =	{16:1--16:8},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-438-3},
  ISSN =	{1868-8969},
  year =	{2026},
  volume =	{393},
  editor =	{Timpf, Sabine and Filomena, Gabriele and Kapaj, Armand and Zhu, Rui and Giudice, Nicholas A. and Manley, Ed},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.COSIT.2026.16},
  URN =		{urn:nbn:de:0030-drops-275607},
  doi =		{10.4230/LIPIcs.COSIT.2026.16},
  annote =	{Keywords: micro-scale traffic analysis, overtaking behaviour, symbolic movement encoding, spatial reasoning, expert evaluation}
}
Document
Short Paper
DOI: 10.4230/LIPIcs.COSIT.2026.32
The Same Street, Different Safety Judgments: Evaluative Framing Effects in Daytime and Nighttime Street View Imagery (Short Paper)

Authors: Tong Qin, Nico Van de Weghe, Wim Fias, and Haosheng Huang

Published in: LIPIcs, Volume 393, 17th International Conference on Spatial Information Theory (COSIT 2026)

Abstract
Street View Imagery (SVI) is widely used to study perceived street safety, yet most studies treat safety as a general impression of urban scenes and rely mainly on daytime imagery. It remains unclear whether safety judgments differ across evaluative framings, defined here as the safety question asked of participants, and whether this difference varies between daytime and nighttime scenes. We analyzed 7,320 safety ratings from 61 participants for 120 SVIs, including 60 daytime and 60 nighttime images. Participants judged scenes under two evaluative framings: crime-related safety and traffic-related safety. Using cumulative link mixed models with random intercepts for participants and stimuli, we found that the stimulus set received higher safety ratings in the crime-related framing than in the traffic-related framing. This evaluative-framing effect was stronger for daytime than for nighttime scenes. These findings indicate that perceived street safety in SVI is not a fixed property of image content, but also depends on the safety question asked during judgment.
Cite as

Tong Qin, Nico Van de Weghe, Wim Fias, and Haosheng Huang. The Same Street, Different Safety Judgments: Evaluative Framing Effects in Daytime and Nighttime Street View Imagery (Short Paper). In 17th International Conference on Spatial Information Theory (COSIT 2026). Leibniz International Proceedings in Informatics (LIPIcs), Volume 393, pp. 32:1-32:7, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2026)

Copy BibTex To Clipboard

@InProceedings{qin_et_al:LIPIcs.COSIT.2026.32,
  author =	{Qin, Tong and Van de Weghe, Nico and Fias, Wim and Huang, Haosheng},
  title =	{{The Same Street, Different Safety Judgments: Evaluative Framing Effects in Daytime and Nighttime Street View Imagery}},
  booktitle =	{17th International Conference on Spatial Information Theory (COSIT 2026)},
  pages =	{32:1--32:7},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-438-3},
  ISSN =	{1868-8969},
  year =	{2026},
  volume =	{393},
  editor =	{Timpf, Sabine and Filomena, Gabriele and Kapaj, Armand and Zhu, Rui and Giudice, Nicholas A. and Manley, Ed},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.COSIT.2026.32},
  URN =		{urn:nbn:de:0030-drops-275763},
  doi =		{10.4230/LIPIcs.COSIT.2026.32},
  annote =	{Keywords: street view imagery, spatial cognition, safety perception, evaluative framing, geographic information science}
}
Document
Vision Paper
DOI: 10.4230/LIPIcs.COSIT.2019.19
The Future of Geographic Information Displays from GIScience, Cartographic, and Cognitive Science Perspectives (Vision Paper)

Authors: Tyler Thrash, Sara Lanini-Maggi, Sara I. Fabrikant, Sven Bertel, Annina Brügger, Sascha Credé, Cao Tri Do, Georg Gartner, Haosheng Huang, Stefan Münzer, and Kai-Florian Richter

Published in: LIPIcs, Volume 142, 14th International Conference on Spatial Information Theory (COSIT 2019)

Abstract
With the development of modern geovisual analytics tools, several researchers have emphasized the importance of understanding users' cognitive, perceptual, and affective tendencies for supporting spatial decisions with geographic information displays (GIDs). However, most recent technological developments have focused on support for navigation in terms of efficiency and effectiveness while neglecting the importance of spatial learning. In the present paper, we will envision the future of GIDs that also support spatial learning in the context of large-scale navigation. Specifically, we will illustrate the manner in which GIDs have been (in the past) and might be (in the future) designed to be context-responsive, personalized, and supportive for active spatial learning from three different perspectives (i.e., GIScience, cartography, and cognitive science). We will also explain why this approach is essential for preventing the technological infantilizing of society (i.e., the reduction of our capacity to make decisions without technological assistance). Although these issues are common to nearly all emerging digital technologies, we argue that these issues become especially relevant in consideration of a person’s current and future locations.
Cite as

Tyler Thrash, Sara Lanini-Maggi, Sara I. Fabrikant, Sven Bertel, Annina Brügger, Sascha Credé, Cao Tri Do, Georg Gartner, Haosheng Huang, Stefan Münzer, and Kai-Florian Richter. The Future of Geographic Information Displays from GIScience, Cartographic, and Cognitive Science Perspectives (Vision Paper). In 14th International Conference on Spatial Information Theory (COSIT 2019). Leibniz International Proceedings in Informatics (LIPIcs), Volume 142, pp. 19:1-19:11, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2019)

Copy BibTex To Clipboard

@InProceedings{thrash_et_al:LIPIcs.COSIT.2019.19,
  author =	{Thrash, Tyler and Lanini-Maggi, Sara and Fabrikant, Sara I. and Bertel, Sven and Br\"{u}gger, Annina and Cred\'{e}, Sascha and Do, Cao Tri and Gartner, Georg and Huang, Haosheng and M\"{u}nzer, Stefan and Richter, Kai-Florian},
  title =	{{The Future of Geographic Information Displays from GIScience, Cartographic, and Cognitive Science Perspectives}},
  booktitle =	{14th International Conference on Spatial Information Theory (COSIT 2019)},
  pages =	{19:1--19:11},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-115-3},
  ISSN =	{1868-8969},
  year =	{2019},
  volume =	{142},
  editor =	{Timpf, Sabine and Schlieder, Christoph and Kattenbeck, Markus and Ludwig, Bernd and Stewart, Kathleen},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.COSIT.2019.19},
  URN =		{urn:nbn:de:0030-drops-111113},
  doi =		{10.4230/LIPIcs.COSIT.2019.19},
  annote =	{Keywords: visual displays, geographic information, cartography, cognitive science}
}

Any Issues?
X

Feedback on the Current Page

CAPTCHA

Thanks for your feedback!

Feedback submitted to Dagstuhl Publishing

Could not send message

Please try again later or send an E-mail
© 2023-2026 Schloss Dagstuhl – LZI GmbH Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact