SoCS Doctoral Consortium
The Doctoral Consortium (DC) aims to provide students with the opportunity to engage closely with established researchers, receive constructive feedback on their research, gain advice on career paths, foster professional networking, and present their work at the conference. While the DC primarily targets Ph.D. students, submissions are open to all students, including master’s students and undergraduates working on combinatorial search.
As part of the Doctoral Consortium, SoCS offers accepted students in-depth guidance from senior members of the community on both research and career development. Each student accepted to the DC (see below for the acceptance procedure) will be paired with an established researcher in the field, who will provide personalized advice on research direction and career planning.
We encourage all eligible students to apply.
Student Support
SoCS will aim at providing student support, partly covering travel and/or registration costs through a separate scholarship application process. Participation in the DC, however, is a requirement for students receiving travel assistance.
More information will be announced soon.
Timeline (all times are 23:59 Anywhere On Earth, UTC-12)
May 18, 2026, submission deadline.
May 25, 2026, notification of accept/reject.
June 8 June 13, 2026, camera-ready deadline for extended abstracts.
Application Procedure for the Doctoral Consortium
To participate in the Doctoral Consortium, an application is required and needs to be submitted at https://easychair.org/conferences/?conf=socs2026 (Doctoral Consortium track)
Applicants for the Doctoral Consortium should submit the following:
Either an extended abstract or a long/short paper submitted to SoCS or another conference/journal, as detailed below. The research topic of the abstract or paper must fall into the topics of interest of SoCS-26.
Extended abstract: An abstract outlining the student's research. The abstract should not exceed 2 pages, inclusive of references, and must adhere to the AAAI two-column, camera-ready style. It is essential to designate "(Student Abstract)" at the end of the title. Papers that surpass the specified length and formatting guidelines may face rejection without review. Applicants have the option to indicate whether they wish for their abstract to be included in the proceedings.
Long/short paper: A paper, with the student as the first author, submitted to either SoCS-26 or another conference or journal within the past year (not including SoCS-25). The venue of submission and its current status (accepted, rejected, or under review) needs to be indicated in the submission page. Note that papers submitted under this option cannot be included in the proceedings. Consequently, there are no restrictions on the format or length of the paper.
Proof of being a student. Applicants must provide documentation confirming their current student status. Acceptable documents include a student ID (with graduation date), an enrollment letter, or a letter from the candidate's advisor, program coordinator, or school administration verifying the applicant's enrollment.
Accepted Papers
The first author of each paper is a student participating in the DC.
Extended Abstracts:
Scalable Robust Multi-Agent Path Finding (Student Abstract)
Amit Bouzaglo
Verification and Automated Construction of Social Laws for MAPF (Student Abstract)
Jakub Mestek
Papers Accepted at SoCS 2026:
Bidirectional Search for Longest Paths: Case for Front-to-Front Heuristics
Tzur Shubi; Ariel Felner; Solomon Eyal Shimony; Shahaf Shperberg
Learning and Completeness in Real-Time Heuristic Search
Devin Wild Thomas; Wheeler Ruml
Make Use of Your Search Effort: Data Augmentation for Learning Effective Planning Heuristics from Optimal Plans
Shanhe Zhong; Eldan Cohen
Maximizing Throughput in Lifelong Multi-Agent Path Finding with Unassigned Agents
Omer Onn; Ariel Felner; Roni Stern
Tree-MAPF: On the Complexity of Optimizing Multi Agent Path Finding on Tree Graphs
Daniel Koyfman; Dor Atzmon; Shahaf Shperberg; Ariel Felner
Bucket Aggregation in Bucket-Based Priority Queues for Heuristic Search
Ryan Goodwin; Eric Hansen; Garrett Fereday
Petri Net Induced Heuristic Search for Resource Constrained Scheduling
Ido Lublin; Dor Atzmon; Izack Cohen
Evaluation of Baseline Methods for SSD-based External Memory Search
Yuki Suzuki; Alex Fukunaga
Previously Published Papers:
Graph-Based Attention for Differentiable MaxSAT Solving
Sota Moriyama; Katsumi Inoue
Advances in Neural Information Processing Systems 38, 137740-137768, 2025.
Other Papers:
Towards Resilient Multi-Agent Pathfinding via Mitigating Cascading Delays with Diverse Plans
Mrinalini Subramanian
Revisiting Landmarks: How To Learn from Previous Plans to Generalize over Problem Instances
Issa Hanou; Sebastijan Dumančić; Mathijs de Weerdt