Terms of Use

The short version

SanskritKatha Feedback is a research platform. By using it, you agree to let us use your review data for research and publication. Your reviews may appear in an anonymized dataset. You will be acknowledged in publications if your reviews meet quality standards. The stories you review are research materials and cannot be copied or redistributed.

What this platform is

SanskritKatha Feedback is a human evaluation platform for AI-generated Sanskrit stories, operated by Mahesh Ramakrishnan as part of an independent research project. The platform collects structured reviews from Sanskrit-literate individuals to validate and complement automated quality scores.

By creating an account and submitting reviews, you agree to these terms.

Your review data

When you submit a review, you grant a perpetual, non-exclusive, royalty-free, worldwide license to use your review data (scores, comments, and associated metadata) for:

• Academic research and analysis
• Publication in research papers, conference proceedings, and journals
• Inclusion in anonymized open research datasets
• Derivative research that builds on the SanskritKatha project

In plain terms: once you submit a review, the scores and comments become part of the research record. You cannot retract them from a published dataset, though you can request deletion of your personal information (name, email) at any time. See our Privacy Policy for details.

Anonymized datasets

Your review data may be released as part of an open research dataset alongside our academic publications. In any such dataset, your personal information is removed. You would appear only as a pseudonymous reviewer ID with your proficiency level and review scores.

The purpose of releasing this data is to allow other researchers to reproduce and build on our work, which is standard practice in computational linguistics research.

Acknowledgment in publications

Reviewers whose trust score exceeds 0.5 will be acknowledged by name in research publications arising from this project. Trust scores are computed based on review consistency, calibration accuracy, and engagement quality (see Quality Control below).

If you prefer not to be named, let us know at [email protected] and we will list you as an anonymous contributor.

Quality control and trust scoring

To maintain the integrity of our research data, we use automated quality checks:

• Calibration stories: A small percentage of stories you review have known expected scores. Your accuracy on these affects your trust score.
• Consistency checks: Reviews that are unusually fast, suspiciously uniform, or consistently at odds with other reviewers may be flagged.
• Trust scores: Each reviewer has a computed trust score (0.0 to 1.0) based on these signals. Reviews from low-trust accounts may be down-weighted or excluded from analysis.

We reserve the right to exclude reviews that fail quality checks from our research analysis. This is not a judgment on you personally; it is a standard data quality practice in annotation research.

Account suspension and termination

We may suspend or terminate your account if:

• Your trust score drops below acceptable thresholds due to consistently low-quality reviews
• We have reason to believe you are gaming the system (e.g. submitting random scores, using multiple accounts)
• You violate these terms

If your account is suspended, we will notify you by email and explain the reason. You may respond to discuss the suspension.

The stories you review

The Sanskrit stories on this platform are AI-generated research materials created as part of the SanskritKatha project. They are the intellectual property of the project.

You may not:

• Copy, redistribute, or publish the stories outside this platform
• Use the stories for commercial purposes
• Scrape or bulk-download stories from the platform

You may discuss stories informally (e.g. mentioning a story in conversation), but please do not reproduce full story texts. The stories may eventually be released as part of the research dataset, at which point they will have their own terms.

No guarantees

Using this platform does not guarantee:

• That the research will be published
• That your reviews will be included in the final analysis
• That you will be acknowledged in publications (this depends on trust score)
• That the platform will remain available indefinitely

This is an independent research project, not a commercial service. We will do our best, but cannot make promises about outcomes.

Limitation of liability

This platform is provided "as is" without warranties of any kind, whether express or implied. To the fullest extent permitted by law, Mahesh Ramakrishnan and the SanskritKatha project shall not be liable for any indirect, incidental, or consequential damages arising from your use of the platform.

In practical terms: this is a volunteer research tool. If something breaks, we will fix it, but we are not liable for any inconvenience caused.

Governing law

These terms are governed by the laws of India. Any disputes arising from these terms or your use of the platform shall be subject to the exclusive jurisdiction of the courts in Bengaluru, Karnataka, India.

Changes to these terms

We may update these terms from time to time. Changes will be reflected on this page with an updated "Last updated" date. For material changes, we will notify active reviewers by email.

Continued use of the platform after changes constitutes acceptance of the updated terms.

Questions?

Email [email protected] with any questions about these terms.