-
Notifications
You must be signed in to change notification settings - Fork 18
Pull requests: James-CUDA/Gittensor-TinyRouter
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(novelty): keep n_agree consistent with agreement_rate when no king
#394
opened Jul 21, 2026 by
marktech0813
Contributor
Loading…
1 task done
Fix #390: [bug] novelty_report n_agree=0 disagrees with agreement_rate=0.5 when no king...
#393
opened Jul 21, 2026 by
marktech0813
Contributor
Loading…
3 tasks
fix(novelty): align n_agree with agreement_rate when no king exists
#392
opened Jul 20, 2026 by
RealDiligent
Contributor
Loading…
1 task done
fix(fugu): resolve livecodebench_v6 in format_hint
#391
opened Jul 20, 2026 by
RealDiligent
Contributor
Loading…
1 task done
feat(analysis): zero-shot transfer report (SPEC §6.1 in-dist vs §6.2 held-out)
#388
opened Jul 20, 2026 by
minion1227
Contributor
Loading…
fix(standings): rank from competition.history, not the always-empty benchmarks subtree
#387
opened Jul 20, 2026 by
galuis116
Contributor
Loading…
feat(analysis): verify SPEC R3 (TRINITY > best multi-agent baseline) offline
#386
opened Jul 20, 2026 by
joaovictor91123
Contributor
Loading…
feat(submission): verify the competition king record's integrity (anti-cheat on best_composite_score)
#384
opened Jul 20, 2026 by
galuis116
Contributor
Loading…
feat(adapters): add the AIME benchmark adapter (harder math, more routing headroom)
#383
opened Jul 20, 2026 by
minion1227
Contributor
Loading…
feat(analysis): verify SPEC R8 (sep-CMA-ES > SFT > RS > REINFORCE) offline
#382
opened Jul 20, 2026 by
joaovictor91123
Contributor
Loading…
feat(analysis): cross-benchmark union view of the sampling & selective diagnostics
#380
opened Jul 20, 2026 by
minion1227
Contributor
Loading…
fix(pr_eval): restore the submitter's theta after novelty — 2 of 3 benchmarks were scored as the king
#379
opened Jul 20, 2026 by
minion1227
Contributor
Loading…
feat(results_table): oracle / headroom / gap-closed columns (ORACLE §7 item 3)
#378
opened Jul 20, 2026 by
minion1227
Contributor
Loading…
feat(analysis): SPEC definition-of-done roll-up — is the replication actually done?
#377
opened Jul 20, 2026 by
minion1227
Contributor
Loading…
feat(analysis): verify SPEC R9 (removing SVF/Thinker/tri-role/penultimate-token all hurt) offline
#376
opened Jul 20, 2026 by
philluiz2323
Contributor
Loading…
6 tasks done
feat(analysis): verify SPEC R11 (trained coordinator > LLM-as-coordinator) offline
#373
opened Jul 20, 2026 by
philluiz2323
Contributor
Loading…
6 tasks done
ProTip!
What’s not been updated in a month: updated:<2026-06-21.