FazBrowse GitHub Viewer | Trending |
URL:
| Home
Tools: [Download Repo ZIP]   [Original HTTPS Page]

Recover translation-bench checkpoints after interrupted writes by datduyng · Pull Request #2769 · microsoft/TypeAgent · GitHub

Recover translation-bench checkpoints after interrupted writes - #2769

Merged
Dominic Nguyen (datduyng) merged 12 commits into
mainfrom
domnguyen/list-determiner-listname
Aug 21, 2026
Merged

Recover translation-bench checkpoints after interrupted writes#2769
Dominic Nguyen (datduyng) merged 12 commits into
mainfrom
domnguyen/list-determiner-listname

Conversation

Dominic Nguyen (datduyng) commented Jul 31, 2026
edited
Loading

Copy link
Copy Markdown
Contributor

Makes single-writer translation-bench checkpoints recover from interrupted JSONL writes while preserving validated resume and merge behavior.

Before: header\n{"caseId":"1"}\n{"caseId":
Append: {"caseId":"2"}
After:  header\n{"caseId":"1"}\n{"caseId":"2"}
Test:   PASS translationBench.checkpointPrimitives.spec.js (1 passed, 0 failed)

Treat determiner-only listName captures (the/my/list) as invalid via
checked_wildcard validation so "add ham to the list" falls through to
the LLM instead of creating a list named "the".
- Normalize: strip true determiners + trailing list(s) only (keep
  whole foods / active tasks / most wanted intact)
- Reject bare closed-class, of-led junk, closed-class "… of them",
  lots/rest partial anaphora; accept open-class "photos of them"
- Execute guards on all listName actions; removeItems throws if missing
- Salvage key __recovered__; checked_wildcard on createList
- Expand unit coverage for validator, coalesce, and edge names
Dominic Nguyen (datduyng) changed the title Reject placeholder list names from list grammar matches Recover translation-bench checkpoints after interrupted writes Aug 21, 2026
Dominic Nguyen (datduyng) marked this pull request as ready for review August 21, 2026 18:41
Dominic Nguyen (datduyng) added this pull request to the merge queue Aug 21, 2026
Merged via the queue into main with commit ae9c10c Aug 21, 2026
26 checks passed
euntaek hong (euntaek-hong) pushed a commit to wrongbutworks/TypeAgent that referenced this pull request Aug 28, 2026
Builds on microsoft#2769 by merging a complete set of checkpoint shards in stable
order and exposing the operation through `pnpm merge-checkpoints`.

```console
$ prettier --check packages/benchmarks/package.json packages/benchmarks/src/translationBench/synthesizer/generationSupport.ts packages/benchmarks/src/translationBench/scripts/mergeCheckpoints.ts
All matched files use Prettier code style!
Result: 3 files checked, 0 formatting errors
```

---------

Co-authored-by: typeagent-bot <typeagent-bot[bot]@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
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.

2 participants


Back | FazBrowse Home | New Git URL