All pull requests

Pull requests

• Add BCal authority contract scenarios

  #1011·Esben Nyhuus Kristoffersen (esbenk) opened Nov 3, 2026Collaborator·

  ·

• Add Sales & Operations Control Tower BCal challenge

  #1004·Sun Haoran (haoranpb) opened Nov 2, 2026Collaborator·2/2·

  ·

• Use BCal compilation evidence in NL2AL scoring

  #1003·Sun Haoran (haoranpb) opened Nov 2, 2026Collaborator·

  ·

• Add repeatable BCal evaluation cohorts

  #1002·Sun Haoran (haoranpb) opened Nov 2, 2026Collaborator·1/2·

  ·

• Add separate BCal completion review model

  #1001·Sun Haoran (haoranpb) opened Nov 2, 2026Collaborator·3/3·

  ·
  1

• Use BCal scenario evidence in evaluations

  #1000·Sun Haoran (haoranpb) opened Nov 2, 2026Collaborator·2/3·

  ·
  1

• Handle reasoning models with BCal tools

  #999·Sun Haoran (haoranpb) opened Nov 2, 2026Collaborator·

  ·
  1

• Refresh BCAL gold dataset criteria and remove unused manifest

  #996·Sun Haoran (haoranpb) opened Nov 2, 2026Collaborator·1/3·

  ·
  3

• [BC-Bench] Add Sales & Operations Control Tower BCal challenge

  #992·Esben Nyhuus Kristoffersen (esbenk) opened Nov 2, 2026Collaborator·

  ·

• Experiment: evaluate seven fixture repairs on the frozen precision baseline

  #989·Wenjie Fan (gggdttt) opened Oct 28, 2026Collaborator·1 task done·

  ·
  5

• [BC-Bench] Use BCal compilation evidence in NL2AL scoring

  #985·Esben Nyhuus Kristoffersen (esbenk) opened Oct 22, 2026Collaborator·

  ·
  4

• [BC-Bench] Use BCal scenario evidence in evaluations

  #984·Esben Nyhuus Kristoffersen (esbenk) opened Oct 22, 2026Collaborator·

  ·
  5

• Experiment: align review gold and compare knowledge with any-match scoring

  #983·Wenjie Fan (gggdttt) opened Oct 22, 2026Collaborator·

  ·
  1

• Experiment: frozen reviewer findings vs single and majority verification

  #982·Wenjie Fan (gggdttt) opened Oct 21, 2026Collaborator·

  ·
  2

• dataset(code-review): add reaction gold answers from BCApps PR 12152-neg-pr12152-20261020

  #981·Wenjie Fan (gggdttt) opened Oct 20, 2026Collaborator·

  ·
  1

• dataset(code-review): add reaction gold answers from BCApps PR 12261-neg-pr12261-20261020

  #980·Wenjie Fan (gggdttt) opened Oct 20, 2026Collaborator·

  ·
  1

• dataset(code-review): add reaction gold answers from BCApps PR 12254-neg-pr12254-20261020

  #979·Wenjie Fan (gggdttt) opened Oct 20, 2026Collaborator·

  ·
  1

• dataset(code-review): add reaction gold answers from BCApps PR 11904-neg-pr11904-20261020

  #978·Wenjie Fan (gggdttt) opened Oct 20, 2026Collaborator·

  ·
  1

• dataset(code-review): add reaction gold answers from BCApps PR 12207-neg-pr12207-20261020

  #977·Wenjie Fan (gggdttt) opened Oct 20, 2026Collaborator·

  ·
  1

• dataset(code-review): add reaction gold answers from BCApps PR 12418-neg-pr12418-20261020

  #976·Wenjie Fan (gggdttt) opened Oct 20, 2026Collaborator·

  ·
  1

• dataset(code-review): add reaction gold answers from BCApps PR 11555-neg-pr11555-20261020

  #975·Wenjie Fan (gggdttt) opened Oct 20, 2026Collaborator·

  ·
  1

• Enable native BCAL multi-turn evaluation

  #968·martinsrui-msft opened Oct 12, 2026Collaborator·

  ·
  1

• [Experimental] Correct HTTP clean fixture and prepare v4 smoke head - AB#652984

  #941·Wenjie Fan (gggdttt) opened Oct 6, 2026Collaborator·4/4·

  ·
  1

• [Experimental] Second semantic-normalization smoke pin - AB#652973 (no product merge/release)

  #940·Wenjie Fan (gggdttt) opened Oct 6, 2026Collaborator·3/4·

  ·

• [Experimental support] Accept strict normalization manifests

  #938·Wenjie Fan (gggdttt) opened Oct 6, 2026Collaborator·2/4·

  ·