feat(hints, summary, docs): Refactor O3 parameters to generic names and restructure GAIA validation docs#58
Conversation
There was a problem hiding this comment.
Pull Request Overview
This pull request refactors O3-specific parameter names to generic model-agnostic names and restructures GAIA validation documentation. The changes make the codebase more flexible and maintainable by removing model-specific naming conventions.
Key changes include:
- Renamed parameters:
o3_hint→hint_generation,o3_final_answer→final_answer_extraction - Updated function names and references throughout the codebase
- Extracted common GAIA validation content into separate documentation files
- Added support for configurable base URLs for hint generation and final answer extraction
Reviewed Changes
Copilot reviewed 35 out of 36 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| utils/util_llm_simple_voting.py | Updated string replacement to use generic "LLM" instead of "O3" |
| utils/util_llm_parallel_thinking.py | Updated string replacement and improved function parameter formatting |
| src/utils/summary_utils.py | Renamed functions to generic names and added base_url parameters |
| src/core/orchestrator.py | Updated function calls and variable names to use generic terminology |
| config/*.yaml | Updated all configuration files to use new parameter names |
| docs/mkdocs/docs/*.md | Restructured GAIA validation docs and updated references |
| scripts/*.sh | Updated script to use new config file names |
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
|
|
||
| yield task | ||
| return |
There was a problem hiding this comment.
Remove the unnecessary return statement at the end of the generator function. Generator functions don't need explicit empty return statements.
| return |
src/utils/summary_utils.py
Outdated
| reasoning_effort="high", | ||
| ) | ||
|
|
||
| # response = await client.chat.completions.create(messages = [{"role": "user", "content": content}], model="dummy") |
There was a problem hiding this comment.
Remove the commented-out debug code. This appears to be leftover debugging code that should be cleaned up.
| # response = await client.chat.completions.create(messages = [{"role": "user", "content": content}], model="dummy") |
src/utils/summary_utils.py
Outdated
| model="o3", | ||
| messages=[{"role": "user", "content": f"[{message_id}] {full_prompt}"}], | ||
| reasoning_effort="medium", | ||
| # reasoning_effort="medium", |
There was a problem hiding this comment.
Remove the commented-out parameter. If this parameter is no longer needed, the commented line should be removed to keep the code clean.
| # reasoning_effort="medium", |
There was a problem hiding this comment.
Pull Request Overview
Copilot reviewed 35 out of 36 changed files in this pull request and generated 3 comments.
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
src/utils/summary_utils.py
Outdated
| reasoning_effort="high", | ||
| ) | ||
|
|
||
| # response = await client.chat.completions.create(messages = [{"role": "user", "content": content}], model="dummy") |
There was a problem hiding this comment.
This commented-out line appears to be debug/test code that should be removed from production code.
| # response = await client.chat.completions.create(messages = [{"role": "user", "content": content}], model="dummy") |
src/utils/summary_utils.py
Outdated
| async def get_gaia_answer_type( | ||
| task_description: str, api_key: str, base_url: str = "https://api.openai.com/v1" | ||
| ) -> str: | ||
| # client = AsyncOpenAI(api_key=api_key, timeout=600) |
There was a problem hiding this comment.
This commented-out line should be removed as it's redundant with the active line below it.
| # client = AsyncOpenAI(api_key=api_key, timeout=600) |
| message_id = _generate_message_id() | ||
| response = await client.chat.completions.create( | ||
| model="o3", | ||
| messages=[{"role": "user", "content": f"[{message_id}] {full_prompt}"}], |
There was a problem hiding this comment.
This commented-out parameter should either be removed or properly documented if it's intentionally disabled for generic LLM compatibility.
| messages=[{"role": "user", "content": f"[{message_id}] {full_prompt}"}], | |
| messages=[{"role": "user", "content": f"[{message_id}] {full_prompt}"}], | |
| # The 'reasoning_effort' parameter is intentionally disabled for compatibility with generic LLM APIs. | |
| # Uncomment and set as needed if using an LLM that supports this parameter, e.g., OpenAI's proprietary models. |
…nd restructure GAIA validation docs (MiroMindAI#58) * update o3 hints and summary to be more general and update docs for gaia validation * to pass lint * improve retry * to incorporate copilot
Summary
gaia_validation_prerequisites.mdagent_gaia-validation_mirothinker.yaml,agent_gaia-validation_claude37sonnet.yamlChanges
o3_hint→hint_generation,o3_final_answer→final_answer_extraction, etc.orchestrator.pyandsummary_utils.py