Skip to content

[GDPVal] Add three reference anchors to the multistage overlay - #3319

Open
agronskiy wants to merge 1 commit into
gdpval-develfrom
agronskiy/feat/gdpval-refs-promote-three
Open

[GDPVal] Add three reference anchors to the multistage overlay#3319
agronskiy wants to merge 1 commit into
gdpval-develfrom
agronskiy/feat/gdpval-refs-promote-three

Conversation

@agronskiy

Copy link
Copy Markdown
Contributor

The reference ladder stops at DeepSeek V4 Pro / 1223, and the AA GDPval-AA v2 board now runs to 1764. A candidate anywhere near the frontier therefore has no anchor above it, and since the anchors are held fixed by the fit, its elo falls out of a saturated win rate rather than out of the comparison — a model that beats every reference on nearly every task has no evidence separating 1300 from 1600.

The September Super-3.5 campaign shows it happening: Inkling-Small fitted at 1296–1338 against this ladder, where its own AA row reads 1191.

Three anchors added, elos from the AA board read 2026-09-11:

key elo why
deepseek_v4_flash_0731 1468 new ceiling
inkling_small 1191
qwen35_122b 925 closes the 905–992 gap

deepseek_v4_flash_0731 is AA's DeepSeek V4 Flash 0731 row, not the April DeepSeek V4 Flash (1116) — the deliverables are from the 0731 release. The original nine are untouched and keep their 2026-09-09 reading.

Paired with the eval-factory MR that adds the same three to configs/benchmarks/gym/gdpval/refs.yaml and the corresponding mounts. Deliverables have to be staged under the refs root before either lands in a run.

🤖 Generated with Claude Code

Mirrors the eval-factory refs.yaml change. The nine-ref ladder stopped at
1223 while the AA GDPval-AA v2 board runs to 1764, so a frontier candidate
had no anchor above it and its elo came out of a saturated win rate rather
than out of the fit.

  deepseek_v4_flash_0731  1468  new ceiling
  inkling_small           1191
  qwen35_122b              925  closes the 905-992 gap

Elos are the AA rows read 2026-09-11. Each resolves under GDPVAL_REFS_ROOT
like the existing nine, so a local run picks them up by pointing that env
var at a directory holding the same per-reference subdirectory names.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Signed-off-by: Alex Gronskiy <agronskiy@nvidia.com>
@yaoyu-33 yaoyu-33 added area:environment Individual environments, benchmarks, verifiers, and environment-specific resources servers feature New capabilities, enhancements, or enablement work needs-review PR is ready for code review and waiting on a reviewer labels Sep 11, 2026
@github-actions github-actions Bot added the sla:triage-overdue Review assignment is over the one-business-day SLA label Sep 14, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

area:environment Individual environments, benchmarks, verifiers, and environment-specific resources servers feature New capabilities, enhancements, or enablement work needs-review PR is ready for code review and waiting on a reviewer sla:triage-overdue Review assignment is over the one-business-day SLA

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants