Skip to content

Multi-objectives: Implement helper to add token usage tracking and validation in objectives/guide #78

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
doxav wants to merge 2 commits into
base: experimental
Choose a base branch
from
Open

Multi-objectives: Implement helper to add token usage tracking and validation in objectives/guide #78

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d4247e5
Implement helper to add token usage tracking and validation in object…
Jun 8, 2026
b098b95
addressed ChingAn review
Jun 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -253,7 +253,7 @@ Key features:
- **Vector scores** — `Guide.get_score_dict()` returns `Dict[str, float]` with named metrics
- **Weighted scalarization** and **Pareto dominance** ranking via `ObjectiveConfig`
- Supported in `BasicSearchAlgorithm`, `BeamsearchAlgorithm`, and `BeamsearchHistoryAlgorithm`
- Token-minimization pattern using `UsageTrackingLLM` + `TokenUsageAugmentingGuide`
- **Token-aware objectives** — wrap an agent LLM with `UsageTrackingLLM`, wrap the task guide with `TokenUsageAugmentingGuide`, then set `ObjectiveConfig.required_metrics` so runs fail early if `error`, `tokens_in`, or `tokens_out` are missing. This supports goals such as "solve GSM8K correctly while minimizing prompt and completion tokens."

Canonical notebooks:

Expand All @@ -262,6 +262,7 @@ Canonical notebooks:
| [multiobjective_quickstart](examples/notebooks/multiobjective_quickstart.ipynb) | Core vector-score infrastructure and BasicSearch integration |
| [multiobjective_trainers](examples/notebooks/multiobjective_trainers.ipynb) | Beamsearch and PrioritySearch multi-objective support |
| [multiobjective_bbeh_langgraph](examples/notebooks/multiobjective_bbeh_langgraph.ipynb) | Real LLM task: BBEH boolean expressions with accuracy + execution time |
| [multiobjective_token_usage_gsm8k_demo](examples/notebooks/multiobjective_token_usage_gsm8k_demo.ipynb) | GSM8K-style objective that minimizes answer error, prompt tokens, and completion tokens |

Trace-Bench multi-objective benchmarks (in [AgentOpt/Trace-Bench](https://github.com/AgentOpt/Trace-Bench)):

Expand Down
Loading
Loading