Launched in 2016, the Doctoral Dissertation Award is awarded annually to recognize a recent doctoral candidate who has successfully defended and completed his or her Ph.D. dissertation in computer graphics and interactive techniques. Recognizing young researchers who have already made a notable contribution very early during their doctoral study, the award is presented each year at the SIGGRAPH Conference and is accompanied by a plaque, complimentary full conference registration and travel to the award ceremony. Honorable Mentions may also be awarded.

Current Recipient

Guy Tevet’s doctoral dissertation, Toward Democratizing Human Motion Generation, represents a landmark contribution to computer graphics, animation, and generative modeling. Rather than presenting a collection of disconnected papers, the thesis advances a coherent research agenda: making high-quality human motion generation controllable, accessible, physics-based, and broadly useful across research and production settings. Through a sequence of foundational contributions, it has reshaped how the community approaches character motion synthesis and established many of the core methods now driving the field.

Among its influential achievements is the introduction of MotionCLIP, which first demonstrated that powerful vision-language representations could be aligned with human motion, enabling intuitive text-based control of animation. This was followed by MDM (Human Motion Diffusion Model), one of the earliest and most successful demonstrations that diffusion models are exceptionally well suited to modeling the richness, diversity, and compositionality of human movement. MDM rapidly became a standard benchmark and conceptual template for subsequent work across graphics, vision, and robotics. The dissertation then extended these ideas through PriorMDM, showing how diffusion models can function as reusable motion priors for editing, composition, and learning with limited data.

A defining strength of the thesis is its breadth and long-term vision. Beyond generation from text, it addresses classical animation challenges such as motion in-betweening, style transfer, refinement from sparse blocking poses, and transfer across diverse character topologies. In CLoSD, the dissertation closes the gap between generative AI and physics-based control by integrating diffusion models with simulation, enabling real-time, text-guided characters that interact plausibly with their environment.

The overall impact of this dissertation is exceptional. It establishes a unifying framework for motion generation, has inspired a large body of follow-up research, and connected recent advances in machine learning to the practical needs of animators and embodied AI systems. Guy Tevet’s thesis is more than the sum of its highly cited publications: it defines a research program that has already transformed the field and will continue to shape it for years to come.

Budmonde Duinkharjav  budmonde@gmail.com      
Ethan Tseng                        eftseng@princeton.edu

Nomination Procedure

All doctoral dissertations successfully defended (or thesis accepted) during the calendar year prior to the nomination deadline are eligible for consideration. There is no limit on the number of nominations that can be made from any single institution or advisor. The key criteria used to evaluate the nominations include technical depth, significance of the research contribution, potential impact on theory and practice, and quality of presentation.

The submitted dissertation should be a finalized version. Nominations are welcomed from any country, but only English language versions will be accepted. Nominations are evaluated by the Outstanding Doctoral Dissertation Award Committee. Nominations, including all supporting materials and endorsement letters, are due by January 31 of each year. Click the button below to submit a nomination.

Requirements