Skip to content

docs: refresh MT-MIA public asset state#296

Merged
DeliciousBuding merged 1 commit into
mainfrom
research/public-asset-refresh-20260525
May 24, 2026
Merged

docs: refresh MT-MIA public asset state#296
DeliciousBuding merged 1 commit into
mainfrom
research/public-asset-refresh-20260525

Conversation

@DeliciousBuding
Copy link
Copy Markdown
Owner

Summary

  • add a 2026-05-25 metadata refresh to the existing MT-MIA evidence note
  • record that the official repo still has no releases/tags and remains at commit d02aebb9241b383f08a4f89cc32054cf283c2ec6
  • confirm sampled JSONL packets still expose full.results and full.scores arrays without per-score row IDs

Decision

  • decision unchanged: Research-side cross-modal support only
  • no GPU release, dataset/model download, Platform/Runtime admission, or relational-tabular consumer lane

Checks

  • git diff --check
  • python -X utf8 scripts/check_public_surface.py
  • python -X utf8 scripts/check_markdown_links.py
  • python -X utf8 scripts/export_admitted_evidence_bundle.py --check
  • python -X utf8 scripts/run_pr_checks.py

Copilot AI review requested due to automatic review settings May 24, 2026 19:22
@DeliciousBuding DeliciousBuding merged commit e46eda1 into main May 24, 2026
2 of 3 checks passed
@DeliciousBuding DeliciousBuding deleted the research/public-asset-refresh-20260525 branch May 24, 2026 19:23
Copy link
Copy Markdown

@gemini-code-assist gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request adds a 'Public Asset Refresh' section to the MT-MIA documentation, detailing a metadata check performed on 2026-05-25. The update confirms that the repository state and the previous decision regarding its status remain unchanged. Feedback was provided regarding a potential inconsistency in the listed result paths, specifically the inclusion of 'rtf', which contradicts the claim that the asset surfaces remain the same as previous assessments.

The refreshed recursive tree still exposes the same public split, synthetic
table, and result-packet surfaces: `data/*/split/{mem,non_mem,ref}/`,
`synth_data/{clava_synth,reldiff_synth}/*/seed_{42,43,44}/`, and
`results/{clava_ddpm,reldiff,rtf}/*/mtmia_seed_*.jsonl`. A path-name scan for
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The path pattern now includes rtf in the results/ directory, which was not present in the initial assessment (see lines 15, 46, and 47). This addition contradicts the statement that the tree "still exposes the same" surfaces. If rtf is a newly discovered generator or result set, it should be documented in the 'Candidate' table and 'Parsed Metric Summary' sections to maintain consistency. If it is a typo, it should be removed.

Suggested change
`results/{clava_ddpm,reldiff,rtf}/*/mtmia_seed_*.jsonl`. A path-name scan for
`results/{clava_ddpm,reldiff}/*/mtmia_seed_*.jsonl`. A path-name scan for

@DeliciousBuding DeliciousBuding review requested due to automatic review settings May 24, 2026 19:45
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant