Be thorough. Be fair. Be courteous. Provide evidence and references for your statements. Above all, be open to new ideas and be constructive.
If the paper is a resubmission with reviewer continuity, please assess how well this revision takes into account previous reviews, and how it improved. The reviewing system gives you access to the former reviews and the authors were instructed to upload a cover letter outlining the changes to their work as supplementary material.
Your evaluation will be forwarded to the authors during the rebuttal period. Please be judicious in asking questions to be answered during rebuttal. Papers should be judged as submitted and thus authors should not be expected to produce new results or conduct additional experiments during the rebuttal period.
Review Ethics
SIGGRAPH Asia 2023 strongly discourages the use of automated text generation tools (for example Large Language Models, online chatbots, etc. such as ChatGPT) in any part of the reviewing process. Reviewers are fully responsible for the text of their reviews and decisions about papers need to be thoughtfully justified, fact-checked, and carefully proofread. Reviews must faithfully represent the reviewers' original opinions on the papers being reviewed and accurately state technical issues with complete, correct references. It is generally unethical to use LLM or other tools to automatically generate reviewing comments that do not originate from the reviewer’s own opinions. Furthermore, online chatbots (e.g. ChatGPT) collect conversation history to generate learned output and their use in any part of the reviewing process would violate the SIGGRAPH/SIGGRAPH Asia confidentiality policy.Please read the Reviewer Instructions and Ethics of Review page before reviewing.
Please be tolerant of papers that do not conform to the usual paper template.
Remember that the authors may not be able to share their code/data for various reasons. Submissions should not be penalized for the lack of code/data, even though such additional material often has a positive effect on the review process. Regardless of whether the code/data is provided, one should be able to recreate it from the paper and the supplementary material, and you may request additional information if needed.
For Conference Paper track submissions, please be open to riskier but potentially-significant work, even if not extensively compared against a variety of alternative options.Remember that different types of contributions require different types of validation and that not everything needs to be demonstrated on an end-to-end application or with a user study. Adjust the expectation to the originality of the contribution; while extensive testing is expected for a refinement of a well-established approach, a lighter-weight validation may be acceptable for a groundbreaking idea.If asking for additional experiments, be mindful that the required resources should not exceed those of a typical lab or those already demonstrated in the submission. When requesting comparisons to previous papers, provide links to the code/data of these papers (if no such code/data exists, consider limiting the request to a discussion). If needed, include suggestions to keep the paper length adequate.
