Application Instructions

1. Eligibility and Requirements: Senior PhD students enrolled in a PhD program before December 2022 or researchers holding postdoctoral positions (including faculty positions) who obtained PhD degree after April 2023.
2. Applicants are required to submit the following materials:
   • CV (including a list of publications).
   • Research statement (up to 2 pages, single column, excluding references), including your research accomplishments and future research directions.
   • A 5-minute video recording for your research summary.
   • Two letters of recommendation uploaded to this form by the referees before July 31st, 2026 (AoE).
3. The awardee must attend the COLM AdvML-Frontiers × CoTMA Workshop and give a presentation in person.
4. Submit the required materials (a),(b),(c) to this form by July 24th, 2026 (AoE).

Important dates

Application materials	Jul 24, 2026
Recommendation letters	Jul 31, 2026

Submission Instructions

Submission Format

We invite paper submissions of up to 6 pages (excluding references and supplementary material).

Please ensure that all submissions conform to the COLM template and submit via OpenReview (submission link to be announced). Accepted papers are non-archival (which will not appear in formal conference proceedings). Concurrent submissions are allowed, but it is the responsibility of the authors to verify compliance with other venues' policies. Accepted papers will be allocated either a spotlight talk or a poster presentation.

Important Dates

Submission deadline	Jun 23, 2026 (AoE)
Notification to authors	Jul 24, 2026
Camera-ready deadline	TBD
Workshop date	Oct 9, 2026

Topics

The topics for AdvML-Frontiers × CoTMA include, but are not limited to:

• Adversarial machine learning for foundation models
• Multi-agent AI security and compositional threats
• Inter-agent attacks and trust exploitation
• Prompt injection and indirect prompt attacks
• Memory poisoning and RAG security
• Tool-use and API exploitation
• Model provenance, fingerprinting, and watermarking
• Model stealing and supply-chain attacks
• Backdoors and poisoning in frontier model pipelines
• Adversarial attacks on reasoning models and chain-of-thought
• Test-time adaptation and intervention security
• Robustness of agentic and embodied AI systems
• Security for multimodal models and VLAs
• Alignment and safety evaluation for frontier systems
• Theoretical foundations of robustness, controllability, and identifiability
• Governance, auditing, and accountability for AI agents
• Benchmarks and red-teaming frameworks for agentic systems

Contacts

For website-related questions, please contact the Workshop Publicity Student Chair. For paper submission and logistics questions, please contact the organizing committee at advml_frontiers24@googlegroups.com.