OpenJarvis

mirror of https://github.com/open-jarvis/OpenJarvis.git synced 2026-07-27 21:05:34 +00:00

Files

T

Andrew Park 08916ec468 feat(orchestrator): faithful ToolOrchestra unified-tool + rejection-sampling SFT pipeline

Replace the ADP-relabel cold-start with a ToolScale-backed, execution-grounded
rejection-sampling pipeline in the paper's real action space (arXiv:2511.21689).

- expert_registry: faithful unified tool calling — one named tool per model
  (gpt_5, qwen3_32b, ...) + basic tools, per-instance subset/price sampling (§3.1/3.3)
- toolorchestra/: split the 1663-line module into a package (prompts/experts/
  sandbox/clients/parsing/workers/agent), behavior preserved
- toolorchestra/rollout.py: faithful reason->act->observe loop (pure, injectable)
- toolorchestra/unified.py: real teacher caller + tool dispatch
- sft_data/toolscale.py: nvidia/ToolScale loader
- sft_data/unified_serialize.py: trajectory -> conversations JSONL (<tool_call> tags)
- sft_data/reject_sample.py: generate_sft_dataset (sample N -> verify -> keep)
- scripts/orchestrator/build_unified_sft.py: CLI
- 25 new offline tests; 127 orchestrator-learning tests green

Verifier is a gold-action-coverage proxy (full ToolScale DB checker + LLM judge
are follow-ups); deployed _run_rl still uses the 3-meta-tool eval port.

2026-06-17 11:35:55 -07:00

install

fix(install): use requires-python range; bootstrap shims are Python-free (#476 , #484 ) (#492 )

2026-06-03 16:59:29 -07:00

mining

feat(mining): validate Gemma E4B and Qwen 3.6 Pearl artifacts (#328 )

2026-05-14 20:11:24 -07:00

orchestrator

feat(orchestrator): faithful ToolOrchestra unified-tool + rejection-sampling SFT pipeline

2026-06-17 11:35:55 -07:00

pearl

feat(mining): validate Gemma E4B and Qwen 3.6 Pearl artifacts (#328 )